Working with raster images

nanoCAD contains tools for creating and editing of raster images.

According to used tool you can edit monochrome, color and/or grayscale images saved in internal or external format (TIFF, BMP, JPG, JPEG, PNG, PCX, GIF).

Most of the functionality of professional raster image processing (elimination of geometric distortions, alignment, calibration, automatic and semi-automatic vectorization, text recognition, color correction and color raster filters) is available in the Raster module. Description of the module commands is marked with the phrase:

This functional is available only in the Raster module.

The Raster module provides an extended set of raster image editing functions:

• Calibration of raster images to eliminate complex distortions in geometry;
• An extended set of raster selection methods with support for intelligent geometry recognition;
• Changing the size and resolution of an image;
• Filtering monochrome and color images using various algorithms to eliminate noise and improve the raster quality;
• Binarization and adaptive binarization for layer-by-layer separation depending on the established criteria;
• Automatic search and selection of text areas and linear objects on a raster.

It also contains a powerful functionality for converting a raster to a vector representation and vice versa:

• Automatic vectorization with text recognition;
• Semi-automatic vectorization (tracing) of polylines, contours, hatches and basic geometric entities;
• Various tracing modes, including intelligent deletion from the raster;
• Rasterization of vector objects.

Insert Raster Image

Command line: IAT, IMAGEATTACH, ROPEN, INSERTRASTER

You can insert images into the drawing. The list of supported raster formats is displayed in the Raster file formats section of the Options dialog box (the Tools menu – Options):

Raster images can be referenced and placed in drawing files but, like external references, they are not actually part of the drawing file. The image is linked to the drawing file through a path name. Linked image paths can be changed or removed at any time.

Once you’ve attached an image, you can reattach it multiple times, treating it as if it were a block. Each insertion has its own clip boundary and its own settings for brightness, contrast, fade, and transparency.

You can insert in the drawing a number of raster image files with the same name but different content. In such a case, a sequence number will automatically be added to the names of such images via an underscore sign ”_” starting with 1.

To insert a raster image, specify the necessary options in the opened Insert Raster Image dialog box.

Parameters:

Name: Includes the list of names of the inserted images.

Opens the Insert Raster Image dialog box.

Path type: Sets the way to describe the location of the source file for the raster image insertion. Dropdown list contains several variants:

• Full path (absolute);
• Relative path;
• No path.

For more information on specifying a path, see the Insert External Reference section.

Insert Point

Specify On-

screen

Selects the box to set the X,Y,Z coordinates values in the command line or specify the position on the screen. The «X» «Y» «Z» fields of this section are inaccessible.

X: Y: Z: Enters the X, Y, Z coordinate values for the raster image insertion in the corresponding fields.

Scale

Specify

Onscreen Selects the box to set the scale value of the raster image in the command line or specify the position on the screen. The «X» «Y» «Z» fields of this section are

W: Sets the scale factor width.

inaccessible.

H: Sets the scale factor height.

Uniform Scale

Specifies the scale factor for the Width or Height values. A value specified for Width is also reflected in the Height value.

Rotation

Specify On- Specifies the rotation angle for the inserted image, using the pointing device.

screen

Angle: Sets the rotation angle for the inserted image.

Insert image relative to

UCS Sets the insert image mode relative to the User Coordinate System (UCS).

View Sets the insert image mode relative to the World Coordinate System (WCS)

Auto Switches on/off the full screen mode of the inserted reference.

zoom

To use raster image georeferencing information stored in a World or TAF file, set the Use World or TAF file option in the Georeferencing section of the OPTIONS dialog:

When inserting raster images with geodata, the coordinates of the insertion point, scale and rotation angle are automatically substituted in the Insert Ratser Image dialog box.

Moving, rotating and scaling a raster image is conveniently done using grips: the upper right grip is responsible for scaling, the lower right grip is responsible for rotation (relative to the lower left), and the left grips (lower and upper) are responsible for movement.

Create New Image and Rasterize Objects

New Image from Selection

Ribbon: Raster – File > New Image from selection

Menu: Raster – New Image from Selection

Toolbar: Raster –

Command line: NEWRASTERFROMSELECTION

Selection for new image can contains:

• raster images;
• vector objects.

When there is no selection, the operation will apply to all data in the current workspace.

When you select vector data, they are rasterized. Color of rasterized objects depends on color of the resulting raster image:

• Monochrome images rasterized objects will have color of monochrome image.
• Color Image rasterized objects will have original color.

How to create new image from selection

• 1. Select objects and run New Image from Selection command;
• 2. Set options in New Image from Selection dialog.

3. Click OK.

New Image from Selected dialog settings:

Image Parameters

If your selection contains raster images it’s better to set Resolution and Color Space not lower than best raster image to avoid quality reduction. For example, if you have raster grayscale raster image with 300 dpi, you should set Grayscale or TrueColor, and Resolution not lower than 300 dpi.

Source

Select objects/area for rasterization:

Position, Margins and Scale

The image scale is set by selecting the desired position:

Using Image Scale, you can create more than one image with specified scale.

In this case files will be created with order number. In this case, the results will be saved to separate files with adding page numbers to the set name:

Set margins and image position. Positions the image relative to the sides of the fixed aspect. Enter indentation values in the section fields corresponding to the image sides. If the image fits into one page of the current format, you can set alignment to one of the sides by left-click on the arrow with the desired alignment type

Embed or save as

Settings for saving the resulting image.

Finish your work

Buttons.

Save Objects to a File

Menu: Raster > Save Objects to a File

Toolbar: Raster –

Command line: RASTEROUT

The command exports part of drawing to graphic format.

After start you will see prompt with a list of selectable options in the command line.

When the command runs in Model space:

Specify area to rasterize or [Screen/Extents/Window]:

When the command runs in Layout:

Specify area to rasterize or [Screen/Layout/Window]:

Options:

Screen Exports the current view of the screen into graphic format.

Window Selects export area by frame.

Specify first window corner: - Specify the first corner of rectangle

frame.

Specify second window corner: - Specify opposite corner of rectangle

frame.

Extents Exports of all objects. Option is displayed when the command runs in

Model space.

Layout Exports of all layout contents. Option is displayed when the command

runs in Layout.

Further command prompts:

Input file folder <C:\ProgramData\Nanosoft AS\nanoCAD Int 25.0\Samples> or [Browse]:

Press ENTER to save the file in the previous path, or browse the path in dialog by selecting option Browse.

Input file name <Untitled0_Layout1>: Press ENTER to save the file

under the previous name or enter new name.

Choose file type or [EMF/WMF/TIF/TIFF/BMP/JPG/JPEG/PNG/PCX]:

Press ENTER to save the file in the previous format or enter other format.

Specify DPI <300> or [75/150/200/300/400/600/1000/1200]:

Press ENTER to save the file in the previous format or enter other format.

Save Screenshot to a File

Ribbon: Raster – File > Save screenshot as to file

Menu: Raster – Save Screenshot to a File

Toolbar: Raster –

Command line: RASTERPRINTSCREEN

Save the current view of workspace to EMF-file.

• 1. Run Save Screenshot to a File command.
• 2. Specify the path for file.
• 3. Click OK.

Screenshot saves to specified folder with the name of drawing file at the first time and nameN, N – is the number of saved file:

Untitled0.EMF

Untitled0_1.EMF

Untitled0_2.EMF etc.

Create New Raster Image

Ribbon: Raster – File > New Image

Menu: Raster – New Image

Toolbars: Raster –

Command line: NEWRASTER

Create new (empty) raster image and insert it to current document.

• 1. Run New Image command.
• 2. In New Image dialog set desired options.
• 3. Click OK.
• 4. Specify insertion point.

New Image dialog options

See the New Image from Selection command for options details.

Separate Raster Image

Command line: SEPARATERASTER

You can separate raster images from original reference and assign new name and path, if raster image was inserted to drawing more than once or its copies were created in a session.

• 1. Run command Separate raster.
• 2. Select raster images to create separate references to files.
• 3. Press ENTER.

Separated raster images marked in External references dialog by adding _clone_N to the name, N – ordinal number of copy.

You can give raster images a new name and storage path using the External References toolbar:

If separate raster images are not saved, then program prompts you to save them on exit.

Embed Raster Image

In nanoCAD there is a mechanism for embedding of raster images. Inserted images saved in external raster files can be embedded and saved in drawing file.

To embed the raster:

• 1. Open dialog External References (menu Insert External References).
• 2. Select reference to a raster image.
• 3. Click the Embed image button.

Embedded image has no saved path, and type of reference becomes Embedded.

Attention

Embedded raster images increase the file size greatly.

Attention

In programs that do not support the mechanism for embedding bitmap images, there may be problems with their display.

Using Save as context menu command the embedded image can be made inserted into the drawing again as a reference by assigning a storage path in a separate raster file.

Image settings

nanoCAD allows you to convert the inserted images into other supported raster file formats. The list of supported raster file formats is shown in the Options dialog box (the Tools menu – the Options command).

For the TIFF format, you can also change settings such as color mode, compression, organization, byte order; for JPG and JPEG formats – Image Quality.

To save an image to another format or to change the format parameters:

• 1. In the External References dialog box select raster references and click Save.
• 2. In the opened Save Image File dialog box, type the new file name, select the file type from the dropdown list and click the Options button.
• 3. Depending on the selected file type, the TIFF Options dialog box or the JPEG Options dialog box will be open.

Displaying quality of raster images

The commands in this section adjust the display of raster images on the screen without modifying their raster data. Changes made by these commands will not in any way affect the display of the same images in another document or another program.

Image Adjust

Ribbon: Raster – Settings > Image Adjust

Menu: Raster – Image Adjust

Toolbar: Modify Object –

Command line: IMAGEADJUST, IAD

You can adjust the brightness, contrast and fade for the display of the raster image without affecting the original raster image file. The image adjustment is intended for improvement of the display of raster images (adjust contrast to make poor-quality images easier to read) or special effects.

Note

To change the intrinsic brightness, contrast, hue and saturation of raster images, use the Brightness/Contrast (LEVELS) command. This command, unlike the Adjust Raster command, modifies the raster image data.

Bitonal images cannot be adjusted for brightness, contrast or fade.

Raster image is selected by its contour (frame or clip border).

To adjust raster image options, use the Image Adjust dialog box.

Parameters:

Raster Image: List of the raster images inserted in the drawing.

Adjustment

Brightness: Controls the brightness of the raster image display.

Contrast: Controls the contrast of the raster image display.

Fade: Controls the fading effect of the raster image display.

These parameters are also available in the Image Adjust section of the Properties functional bar:

Visibility

Show Image Controls the display of the image content on the screen. If this checkbox is not selected, then only the contour of the raster image is shown.

Show Clipped If this checkbox is selected, only the clipped area of the raster image is displayed. Otherwise, the raster image is displayed completely, even if a clip has been set for it.

To set a clip for the raster image use the Image Clip command (ribbon > Insert tab > Reference > Hatch > Image Clip).

Transparency Used only for bitonal raster images and images with transparent pixels. Makes the background color of the image transparent.

When enabling the transparency mode, it becomes possible to view through transparent pixels of the raster image of objects that are in the graphics area behind the raster.

The transparency property is supported for those raster file formats in which transparent pixels exist, for example, in monochrome images (* .BMP) background pixels are transparent.

These parameters are also available in the Misc section of the Properties functional bar:

Resets the values for brightness, contrast and fade to the default settings.

You can set transparency level for the raster image as for drawing object.

To change the transparency of the raster image as drawing object:

• 1. Select the raster image.
• 2. Select a transparency value in the Transparency drop-down box in the General section of the Properties panel.

The IMAGEFRAME system variable allows you to manage the visibility of the clipping contour and raster contour. If the system variable is set to value 1 (set by default), the contour is displayed on the screen and you can select it and print it. If system variable is set to value 0, the contour visibility is turned off and you cannot select it and print it. If system variable is set to value 2, the contour is displayed on the screen, but you cannot print it.

There are commands in the Raster menu – Object > Image > to make work with IMAGEFRAME system variable easier:

Frame On - Sets IMAGEFRAME = 1 Frame Off - Sets IMAGEFRAME = 0 Print Off - Sets IMAGEFRAME = 2

Quality of Raster Images

This function changes quality of raster images from high (by default) to draft, accelerates image downloading and improves system performance when processing large images.

Ribbon: Raster – Settings > Image Quality Menu: Raster – Image Quality Command line: IMAGEQUALITY

To change the quality of raster image:

1. Run command Image Quality.

2. Select the required option in the command line: IMAGEQUALITY : [High/Draft].

Transparency of Raster Images Background

Changing the transparency of raster images background on the screen.

The command allows you to switch the transparency of the background pixels of bitonal raster images (for example, TIF format), as well as the alpha channel pixels of color images (for example, PNG format).

Note

The command switches the transparency of raster images background only on the screen. Raster images are not modified.

To change the transparency of raster image background:

• 1. Run the command.
• 2. Select the required option: Enter transparency mode [ON/OFF] :

The background transparency of selected raster images can also be changed in the Properties bar.

Correcting Raster Image Geometry

This section provides procedures for correcting geometric distortions in monochrome, color, and grayscale raster images. Such operations are applied before using more complex procedures such as interleaving or vectorization.

Resizing a Raster Image

Menu: Raster – Change Size

Command line: IMAGESIZE

This functionality is available only in the Raster module.

It is used to bring the image size in accordance with predetermined values.

It is advisable to carry out after performing such operations as deskewing and cropping the image, as well as after scanning, which results in an image of a non-standard size. If the new image is smaller than the original one, the image is cropped. If the new image is larger than the original one, margins are added to the image. Changes are tracked in the preview window.

The operation can be applied to only one image.

Select and image and run the command.

In the dialog box that appears, set the following parameters:

• 1. In the Units field, select measurement units from the list;
• 2. Determine the required image size by one of the following ways:
  • Automatically determine the closest format click the Find closest button in the Standard Paper Size section;
  • In the Standard Paper Size list, select one of standard paper formats and the required orientation (Landscape or Portrait). You can customize the Standard Paper Size list in the Papers section of the Options dialog;
  • Enter values in the Width and Height fields of the Image Size section;
  • If you work with a color or grayscale image, the color of added image edges is specified by the Layout Background Color parameter in the Color settings section of the Options dialog box;
• 3. Align the image using one of arrows of the Alignment/Margins field. For example, you can align the image to the top left by clicking the top left corner arrow. To center the image, select the central button;

• 4. Specify sizes of the image margins by one of the following ways:
  • Resize the image by increasing or decreasing the margins. To do this, clear the Fixed size checkbox and, using the appropriate fields in the Alignment/Margins section, increase or decrease the size of margins in required directions by entering positive or negative values in the appropriate fields;
  • Align the fixed size of the image by changing the margin values. Check the Fixed size box. Then for the required image margins, enter the changed values in the appropriate fields;
• 5. Press OK.

Changing Image Resolution

Ribbon: Raster – Modification > Resample

Menu: Raster – Resample

Command line: RESAMPLE

This functionality is available only in the Raster module.

The operation allows you to proportionally change geometric dimensions (scale) of an image and its resolution (DPI).

When scaling, the image dimensions in pixels can be changed or left unchanged. In the latter case, the image is resized in relative units (millimeters or inches) by changing the resolution value.

You can change the image in pixels by:

• Changing the actual image size while maintaining the resolution value;
• Changing the resolution;
• Changing the actual image size and its resolution.

Attention

When changing the image resolution without a corresponding change in its size, the image file size will be proportional to the square of its resolution. For example, a 200 ppd image file is four times the size of an image file of an image file of the same size and 100 ppi resolution.

The operation can be applied to only one image.

• 1. Select an image and run the command.
• 2. Specify the parameters in the Resample dialog box that appears.

• 3. If you want to save the original image size in pixels, check the Fix size in pixels box;
• 4. To increase or decrease the image size by changing the value of image pixels, clear the Fix size in pixels box.
• 5. Select the measurement units from the list;
• 6. Specify the image size in the required way:
  • Enter new values for the image width or height in the Width or Height fields (with selected pixels as a measurement unit and the checked Fix size in pixels box these values are blocked). When you enter any value in one field, the value in another one changes automatically to maintain the image proportions;
  • Enter the scale factor as a percentage of the current image size in the Scale field. If the Fix size in pixels box is checked, then setting a new actual size affects the resolution value. The image size in pixels remains unchanged. If the Fix size in pixels box is cleared, then setting a new actual size affects the size in pixels. The image resolution remains unchanged;
• 7. To change the image resolution, enter the required values in dots per inch (dpi) in the Resolution field. If the Fix size in pixels box is checked, then the program remains the image size in pixels unchanged, changing only the image resolution;
• 8. Click OK.

Cropping of Raster Images

The size of raster image can be reduced to determined area by using Crop operations. This area can be determined by set clip of image or rectangle. Also, raster can be cropped automatically using procedure that determines empty image fields and crops them.

These operations apply to multiple images at once. For example you can crop pack of images by placing them on top of each other.

Automatic crop

Automatic crop trims the “empty” image field to the smallest possible rectangle that includes all raster data.

Ribbon: Raster – Crop – Crop > Auto

Menu: Raster – Crop – Auto

Toolbar: Raster –

Command line: CROPAUTO

• 1. Select raster to be processed on the screen. If there was no selected images, then the command processes all visible images placed in non-locked layers.
• 2. Run the command.

Original raster image

The result of the Automatic crop command

Auto crop by frame

If the raster has a frame, then the command automatically changes the size of raster by the frame.

862

Ribbon: Raster – Crop – Crop > By Frame

Menu: Raster – Crop – Auto by frame

Toolbar: Raster –

Command line: CROPBYFRAME

If the image has deformations, use the Deskew command for correct CROPBYFRAME behavior.

• 1. Select raster to be processed on the screen. If there was no selected images, then the command processes all visible images placed in non-locked layers.
• 2. Run the command.

If the raster has no frame, then the result of Auto crop by frame is similar to the Crop → Auto.

Crop by Rectangle

Cropping can reduce an image size to a specified rectangular area size.

Ribbon: Raster – Crop > By Rectangle

Menu: Raster – Crop by Rectangle

Toolbar: Raster –

Command line: CROPBYRECT

• 1. Start the command.
• 2. Specify two opposite corners for the rectangular area in the raster.

Parts of an image outside the specified area are cropped.

Crop by clip

Ribbon: Raster – Crop > By Clip

Menu: Raster – Crop – By Clip

Toolbar: Raster –

Command line: CROPBYCLIP

To crop the raster by clip, the clip of the raster image should be predefined (Modify → Clip → New… command).

• 1. Select raster to be processed (visible and placed in non-locked layers). If there was no selected images, then the command processes all visible images placed in non-locked layers that have the clip.
• 2. Run the command.

The raster image will be cropped by the clip of the raster image.

Mirroring

The operation allows mirroring an image around either vertical or horizontal axis crossing the image center.

Ribbon: Raster – Align > Mirror By X axis

Ribbon: Raster – Align > Mirror By Y axis

Menu: Raster – Mirror > By X Axis

Menu: Raster – Mirror > By Y Axis

Command line: MIRRORX

Command line: MIRRORY

Toolbar: Raster –

1. Select images to mirror. If no image is selected, then this command processes all visible images located on unlocked layers.

2. Start the command Raster Modify – Mirror > select condition:

• By X Axis, to flip image vertically;

• By Y Axis, to flip image horizontally.

Rotation

The operation rotates image around a central point, using three fixed rotation angles (90°, 180° or 270°) or any angle. When an image is rotated at any angle the image size increases to inscribe rotated initial image.

To rotate at 90°, 180°, 270° angles:

Ribbon: Raster – Align > Rotate 90 ccw

Ribbon: Raster – Align > Rotate 180

Ribbon: Raster – Align > Rotate 270

Ribbon: Raster – Align > Rotate 90 cw

Menu: Raster – Align – Rotate > 90 ccw

Menu: Raster – Align – Rotate > 180

Menu: Raster – Align – Rotate > 90 cw

Command line: ROTATEAT90

Command line: ROTATEAT180

Command line: ROTATEAT270

Toolbar: Raster – 90 ccw Toolbar: Raster – 180

Toolbar: Raster – 90 cw

Select the images to rotate. If no image is selected, then this command processes all visible images located on unlocked layers.

Start the command Raster Modify – Rotate > select rotation condition.

To rotate at custom angle:

Ribbon: Raster – Align > Rotate Custom Angle Menu: Raster – Rotate > Custom Angle… Toolbar: Raster –

• Command line: ROTATEATANGLE
• 1. Select the images to rotate. If no image is selected, then this command processes all visible images located on unlocked layers.
• 2. Start the command.
• 3. Enter rotation angle to Angle field.

Or press Measure and specify line by 2 points. Angle will be measured between this line and X-axis.

4. Select OK.

Deskewing

This operation enables an image skew resulting from scanning to be corrected. The whole image is rotated about its central point in order to eliminate either horizontal or vertical skew. When an image is rotated at any angle the image size increases to inscribe rotated initial image.

Menu: Растр – Deskew > Manual

Command line: DESCEWAUTO

Command line: ROTATEATANGLE

Toolbar: Raster – Auto

Toolbar: Raster – Manual

To deskew an image automatically:

• 1. Select the images to deskew. If no image is selected, then this command processes all visible images located on unlocked layers.
• 2. Start the command Deskew > Auto.

If the program is able to estimate the rotation angle, then raster images will be deskewed. Otherwise, use the manual deskewing procedure.

To deskew an image manually:

• 1. Select the images to deskew. If no image is selected, then this command processes all visible images located on unlocked layers.
• 2. Start the command Deskew > Manual.
• 3. Enter angle value to the Angle field.

Or press Measure and specify line for deskew by start and end points.

4. Press OK.

Correction by Four Points

Menu: Raster – 4 point correction

Toolbar: Raster –

Command line: FRAMING

The command is used for correction of scanned images with frame. It is suggested that the image frame and its content are distorted equally.

To perform this procedure, specify the desired frame size – its height and width, and the appropriate current position of the frame corner dots on the image. After correction the image is transformed so

that the frame corners are moved to the rectangular frame corners of the specified size, whose sides are parallel to image sides.

If no image is selected, then this command processes all visible images located on unlocked layers.

To perform four point correction:

• 1. Select the images to correct by four-point correction. If no image is selected, then this command processes all visible images located on unlocked layers.
• 2. Start the command. The 4-point Correction dialog will be displayed:

• 3. Click the Find frame button. If the program is able to find the drawing frame, then you will see a blue polygon over the image close to the raster lines. If a frame is not found specify it manually.
• 4. To specify the frame corners manually, press the button and click the frame corners on the image. These points can be specified in an arbitrary order because the program always sorts them so they form the frame without intersections. Watch the red rubberline to control, press BACKSPACE to go to previous frame corner if necessary.
• 5. In the Frame size field specify Width and Height.
• 6. Select orientation Landscape or Portrait.
• 7. Press OK.

Find closest format button starts searching closest standard paper format.

Or you can select format manually in Frame Size combo-box.

For standard paper formats you can use the Use internal frame option to correct frame size depending on paper format settings. Internal frame size can be defined in Papers section of the Options dialog ( Tools - Options).

Set the internal page frame size:

• 1. Select paper format in Paper formats section.
• 2. Press Modify button.
• 3. In the Modify paper format dialog set Internal frame and specify margins in Frame margins fields.

For paper formats without internal frame specified, the Use Internal Page Frame option in 4-point Correction dialog will be disabled.

Calibration

Menu: Raster – Calibration

Command line: CALIBRATION

This functionality is available only in the Raster module.

The calibration operation affects the entire image. When multiple images are selected, the command is applied to visible images located on unlocked layers.

The calibration transformation is determined by the transformation model and a set of calibration pairs.

When preparing calibration, it is necessary to specify the vectors of raster points displacement. To do this, specify a set of calibration pairs. Each of these pairs determines two coordinates - the current position of a point on the image (measured point) and its required theoretical position (real point).

A transformation model is a type of parametric transformation used in calibration. Each model determines a family of transformations of the same type.

When using some sets of calibration pairs and individual methods, the program is not able to perform the transformation of a given type in such a way that all measured points move to the corresponding real points. This leads to deviation of the points obtained as a result of transformation from the corresponding real points. The criterion for choosing the conversion parameters is to minimize the rootmean-square error at all calibrated points.

Each of the calibration pairs is one of the following types:

• Grid if a pair is part of the calibration grid; used in calculating the calibration parameters and assessing the calibration accuracy;
• Check if a pair is used when calculating the calibration parameters and assessing the calibration accuracy;
• Control if a pair is used only to assess the calibration accuracy and does not affect the calibration parameters;
• Unused if a pair is not used when calculating the calibration parameters and assessing the calibration accuracy.

Basic Calibration steps

• 1. run the calibration command;
• 2. create a set of calibration pairs;
• 3. specify the position of measured points;
• 4. choose a suitable calibration method;
• 5. calibrate.

Before performing the calibration, it is recommended to set the coordinate system - origin and scale.

Calibration command

The command opens the calibration dialog box.

Calibration dialog buttons

Specifying a set of calibration pairs

When creating calibration pairs, their definitions are added to the list in the Calibration dialog box:

• 1. Specify the known theoretical coordinates of points (real points) in one of two ways: by specifying a calibration grid or adding points one by one, or both at the same time. When created, each calibration pair has the same coordinates of measured and real points;
• 2. Set the corresponding measured points for all real points by selecting them in the image or by entering their coordinates from the keyboard.

Setting the calibration grid

When specifying a calibration grid, it creates a set of calibration pairs, the points of which are located at the nodes of a rectangular grid. Such calibration pairs relate to the Grid type.

The position of the calibration pair points specified when creating a grid is determined by the origin of the grid, the cell size, and the number of cells in the horizontal and vertical directions.

There can be only one grid in a calibration object. Re-setting the calibration grid will delete all calibration pairs belonging to the existing grid.

Click the Define grid button

The Calibration grid dialog box appears:

1. Specify the calibration grid origin;

• 2. Enter coordinates in the Origin field or click the Choose origin button and specify with the mouse the location of the grid origin in the image. The bottom left corner of the grid is taken as the grid origin, and the grid is generated in the positive direction of X and Y axes;
• 3. Set cell dimensions along X and Y axes;
• 4. If necessary, you can also add columns in the negative direction of X or Y axes by specifying a negative value for the X or Y dimensions;
• 5. Specify the number of cells along X and Y axes using the X cells and Y cells respectively;
• 6. To avoid errors, click the Preview button and view the specified grid. If necessary, correct errors;
• 7. Select OK to create a calibration grid and return to the Calibration dialog box.

You can create a rectangular grid rotated by a specified angle. Otherwise, the grid rows and columns will be orthogonal to the X and Y axes.

Setting the calibration grid through the loaded file

To add an arbitrary calibration grid, it is possible to create a text file with the RPT extension. After specifying the coordinates and parameters of all real calibration points in it, and starting the calibration dialog before starting to change the measured calibration points, you should load this file through the Import grid button.

RPT file format:

first line:

• Unsigned Int Calibration method second line:
• Unsigned Int Number of points next lines (separated by a space):
• Unsigned Int Sequential number of the calibration pair
• Double Real point x coordinate
• Double Real point y coordinate
• Double Measured point x coordinate
• Double Measured point y coordinate
• Double Calculated point x coordinate (identical to x-measured point before calculation)
• Double Calculated point y coordinate (identical to y-measured before calculaiton)
• Bool Check point
• Bool Control point?
• Bool Used point?
• Unsigned Int Sequential number of point in the grid along x axis (starting from 0)
• Unsigned Int Sequential number of point in the grid along y axis (starting from 0)

String – Point label (name)

Example of RPT file

10 16 0 414.250000 -312.500000 415.789786 -311.284805 414.250000 -312.500000 1 1 1 0 0 Grid(01,01) 1 436.250000 -312.500000 437.410740 -310.659621 436.250000 -312.500000 1 1 1 1 0 Grid(01,02) 2 458.250000 -312.500000 460.386260 -308.679871 458.250000 -312.500000 1 1 1 2 0 Grid(01,03) 3 480.250000 -312.500000 481.694622 -309.044562 480.250000 -312.500000 1 1 1 3 0 Grid(01,04) 4 414.250000 -288.500000 414.799911 -285.027068 414.250000 -288.500000 1 1 1 0 1 Grid(02,01) 5 436.250000 -288.500000 431.221632 -307.526974 436.250000 -288.500000 1 1 1 1 1 Grid(02,02) 6 458.250000 -288.500000 464.815333 -257.650194 458.250000 -288.500000 1 1 1 2 1 Grid(02,03) 7 480.250000 -288.500000 419.868235 -297.301982 480.250000 -288.500000 1 1 1 3 1 Grid(02,04) 8 414.250000 -264.500000 414.250000 -264.500000 414.250000 -264.500000 1 1 1 0 2 Grid(03,01) 9 436.250000 -264.500000 436.250000 -264.500000 436.250000 -264.500000 1 1 1 1 2 Grid(03,02) 10 458.250000 -264.500000 458.250000 -264.500000 458.250000 -264.500000 1 1 1 2 2 Grid(03,03) 11 480.250000 -264.500000 480.250000 -264.500000 480.250000 -264.500000 1 1 1 3 2 Grid(03,04) 12 414.250000 -240.500000 414.250000 -240.500000 414.250000 -240.500000 1 1 1 0 3 Grid(04,01) 13 436.250000 -240.500000 436.250000 -240.500000 436.250000 -240.500000 1 1 1 1 3 Grid(04,02) 14 458.250000 -240.500000 458.250000 -240.500000 458.250000 -240.500000 1 1 1 2 3 Grid(04,03) 15 480.250000 -240.500000 480.250000 -240.500000 480.250000 -240.500000 1 1 1 3 3 Grid(04,04)

Adding calibration pairs one at a time

When using this method, pairs are added one by one. The created pairs can be of one of the following types: Check, Control or Unused. This procedure is designed so that you can create pairs by specifying only the real point coordinates. The measured points can be specified later.

Adding points:

1. Select Add in the context menu or click the Add point button located in the Calibration dialog box. The dialog box for adding a point will appear;

• 2. Enter the real point coordinates in the Real field;
• 3. Enter the pair name in the Label field, otherwise this pair will be named PointNN by default;
• 4. Enter the measured points coordinates in the Measured field or specify them on the screen using button. If this is not done, their coordinates will coincide with the real ones and can be changed later;
• 5. If necessary, in the Use as field, change the type of pair from Control to Check or, by clearing the checkbox, define the pair type as Unused;

6. Press ENTER or select Add to create the pair and continue the operation.

Specifying measured points on the screen

The measured points can be specified on the screen using the mouse:

• 1. Select the calibration pair to be changed from the list of the Calibration dialog box or on the screen. The program will highlight the selected point with help of “grips”;
• 2. Specify the selected pair using button or by the Zoom to command of the context menu. The program pans the image so that to show the measured point of the selected pair in the center of the screen;
• 3. Change the location of the measured point. Place the cursor over the “grips” and left-click. Move the cursor and click again to determine a new location of the measured point;
• 4. Move to the next or previous calibration pair. Select Next or Previous in the context menu or click or button on the toolbar of the Calibration dialog box. The program will pan the image so that to show the next (previous) measured point of the selected pair in the center of the screen.

Choosing a calibration method

A calibration method is chosen taking into account the nature of image distortion, as well as the number and location of calibration pairs.

If there is no information regarding image distortion, you can use the Choose automatically method. In this case the program itself will choose the calibration method that is optimal for a given set of calibration pairs.

The table lists possible distortions and calibration methods used to correct them.

Calibration methods to correct distortions

Each calibration method assumes the minimum number of calibration pairs the model can use. If the number of calibration pairs exceeds a certain value, all models except Surface Splines will produce nonzero distortion.

The following table lists the limits on the number of calibration pairs for each calibration method:

Limitations on the number of pairs for each distortion correction method

Calibration accuracy estimation

Calibration transforms the entire raster image through a calculated transformation. Usually not only the points specified in the calibration pairs are moved, but all the image points. Accuracy estimation allows you to determine the offset of each raster point for the selected calibration method prior to performing the calibration procedure.

During the estimation, another point is created for each calibration pair, called the calculated one. It shows the position of the measured point after performing the selected calibration method. Labels of such points are highlighted in yellow (labels of real points – blue, labels of measured points - red; you can change these default colors in the Options dialog box). Then the program calculates and displays the distances between each calculated point and the corresponding real point. These distances determine the calibration deviations for each pair.

The program calculates the parameters of the selected method, therefore, after transformation, each measured point is placed as close to the corresponding real point as possible. For these calculations the program uses only calibration pairs of the control and grid types.

To estimate the displacement of any point in the image after calibration, it is necessary to create a calibration pair with measured and real points that have the coordinates of the required image point and assign this pair a check type. This pair will not be considered when determining the transformation parameters, but the program will find the calculated point for it and calculate the deviation from the actual point location.

To estimate the calibration accuracy

• 1. Create calibration pairs required to correct image distortions.
• 2. Specify the location of the measured points.
• 3. If necessary, create test pairs to determine the direction of movement of arbitrary pairs.
• 4. Select the desired method from the list of calibration methods. Only models applicable to a given set of calibration pairs can be used. If these requirements are not met, the estimation cannot be completed.
• 5. Click the button located in the Calibration dialog box.

The Mean error parameter shows the mean error of the chosen method. Note also the Xerr and Yerr values, which are used to estimate the error of each point when using the chosen method.

Color Correction

This functionality is available only in the Raster module.

A real map or color scheme, as a rule, is performed using a small number of colors, however, when scanning a paper original, a color raster image is obtained, the points of which have several tens or even hundreds of colors. Color filtering and color correction tools are used to prepare images for subsequent complex operations such as binarization, slicing, raster editing, and vectorization. Color filters are also used to improve the quality of images after applying operations that move objects in the image or change the resolution (scaling, alignment, rotation, calibration, or four-point correction).

Conversion to True Color, Grayscale and 256 Colors

This functionality is available only in the Raster module.

A raster image can be converted to color modes: TrueColor (full color), Grayscale (half-toned), and to indexed colors.

Conversion operations can be applied to several images of the same type at the same time.

Converting monochrome images to grayscale, TrueColor or indexed color models makes it possible to apply color filters to them (Blur, Adaptive blur, Contour Sharpness, Average).

Conversion to True Color (Full Color)

Ribbon: Raster – Processing – Convert to 24 bits > Conversion to True Color

Menu: Raster – Processing the raster – Convert to – True Color.

• 1. Select images.
• 2. Perform the command.

Conversion to Grayscale

Ribbon: Raster – Processing – Convert to 24 bits > Convert to Grayscale

Menu: Raster – Processing the raster – Convert to – Grayscale

Command line: CONVERTTOGRAYSCALE

• 1. Select images.
• 2. Perform the command.

Conversion to the Indexed Colors

The command allows you to reduce the image file size by changing the color depth. The command also provides the ability to fine-tune the colors. As a result of its application to an image represented by the True color model, the number of colors used is reduced to 256 (or less). You can further reduce the number of colors by deleting selected colors or merging several colors into one. You can replace selected colors and add them to the palette.

• 1. Select the images to be converted.
• 2. Run the command.

The Convert to indexed colors dialog has a preview window, a table of current colors and editing tools.

Convert automatically

To obtain an image with the maximum possible number of colors, enter the value 256 (suggested by default) in the color field and click Reset palette button.

To automatically reduce the number of colors in the selected image, click Set auto palette button. If the results are satisfactory, click OK in the preview window

Setting the number of colors in the palette

Set the desired value in the color counter , click Reset palette button. The palette will be automatically rearranged in order to display the image in the best possible way in the colors, the number of which is set in the counter.

Removing a color from the palette

• 1. Select the color(s) to be removed from the table. Multiple colors are selected while holding down the SHIFT key.
• 2. Click Delete selected colors button. The image will be redrawn in a new palette.

Adding a color to the palette or replacing one color with another

• 1. Select a color in the table to be replaced, or point to an empty square if you want to add a color.
• 2. Click Get color to set button and specify the required color to be added or replaced in the dialog box.

To get color from an image: click Choose color to change palette button, specify the point on the raster image.

Combining multiple colors into one

• 1. Select multiple colors in the table while holding down the SHIFT key.
• 2. Click Merge colors button. The selected colors will be reduced to one of the colors existing in the palette, which is as close as possible to the average value of all selected ones.

Changes are controlled in the preview window.

Click OK to apply changes to the image.

Changing Brightness, Contrast, Hue and Saturation

Ribbon: Raster – Processing > Brightness/Contrast

Menu: Raster – Processing the raster > Brightness/Contrast…

Command line: LEVELS

This functionality is available only in the Raster module.

Using the Brightness/Contrast command, you can adjust brightness, contrast, hue and saturation of pixels of one or several color or grayscale images.

Brightness characterizes the relative lightness or darkness of a color and is measured as a percentage: from 0 (black) to 100 (white).

Changing the contrast allows you to increase or decrease the brightness differences in the image pixels. An image with the same pixel brightness has zero contrast. Increasing the contrast increases the

differences in brightness by darkening dark colors and brightening bright colors. The contrast of an image changes as a percentage: from 0 (solid gray) to 100. The hue is usually understood as the color, and the saturation is the color purity. The command allows you to completely change the color content of an image.

Hue is the wavelength of light reflected or transmitted through an object. Typically, a color name (red, orange, green, etc.) is used to describe a color tone. Each hue occupies a specific position on the standard color wheel and is characterized by an angle ranging from -180° to + 180°.

Saturation is the degree of a color purity. It defines the ratio of gray to a given hue and is expressed as a percentage: from 0 (gray) to 100 (fully saturated).

• 1. Select color or grayscale raster images to be processed.
• 2. Run the Brightness/Contrast… command.
• 3. In the dialog box, use the corresponding input fields or sliders for changing values of Brightness, Contrast, Hue and Saturation parameters.

In the preview window, you can observe the change in parameters.

Click Apply.

The Reset button sets 0 value for all parameters.

Correction by Histogram

Ribbon: Raster – Processing > Equalize

Menu: Raster – Processing the raster > Equalize…

Command line: EQUALIZER

This functionality is available only in the Raster module.

The operation is used to adjust the brightness, color hue and contrast of an image. To do this, a histogram correction algorithm is used with the setting of two threshold brightness levels - the darkest and brightest pixels, as well as the image gamma, which determines the position of the average brightness value relative to the current threshold values.

Gamma specifies the ratio of length of the brightness range between the average and the brightest values to the length of the brightness range between the dark threshold and the average value. As a result of using the command:

• pixels with a brightness value below the dark threshold receive a zero brightness value;
• pixels with brightness values higher than the brightest maximum brightness value (255);
• the brightness values of pixels lying between the darkest and average values and between the average and the brightest levels are redistributed evenly in accordance with the lengths of the intervals given to them, which are determined by the image gamma.

Increasing the gamma value decreases the interval provided for brightness in the dark range and therefore increases the contrast in it, while simultaneously decreasing the contrast in the light range, and vice versa.

The command allows you to redistribute both the average brightness of the image pixels and the brightness by individual color components of pixels (Red, Green and Blue). This allows you to correct the color of pixels in the image - for example, make a pink background pure white.

• 1. Select raster images to be processed.
• 2. Run the command.

The Equalize dialog box presents the image histogram that shows the averaged number of pixels corresponding to each brightness value:

• low brightness values correspond to the left side of the histogram;
• high (lightest tones) the right side of the histogram.

The slides at the bottom of the histogram show the threshold values: black on the left is the darkest, gray in the middle is the average, and white on the right is the brightest pixel.

It is possible to choose one of four histograms:

• Master shows the summarized distribution of pixel brightness;
• Red, Green and Blue show the brightness distribution of the corresponding color components of pixels (red, green, and blue).

Select the histogram corresponding to the color component to be corrected.

Set the brightness values of the darkest, lightest pixel and gamma of the image using the Levels fields or the eyedropper. For fine tuning, use the histogram sliders.

The Levels fields contain a numerical expression of the current thresholds. Using the sliders of the Master histogram, you can proportionally change the values of the thresholds of all components at the

same time. The Red, Green, and Blue histogram sliders change the brightness thresholds separately for their respective color components. The eyedroppers are used to select thresholds and gamma in an image. If a color sample is taken by an eyedropper Pick low level (or Pick high level), then the value of the darkest (brightest) threshold component is set. If a color sample is taken by an eyedropper Pick middle level, the middle tone position is defined and thereby – the image gamma. The brightness values of all other pixels will be proportionally redefined within the boundaries of the new tone range. Moving the middle slider changes the gamma value and redistributes the contrast between

The Auto button automatically sets the values of the light and dark thresholds so as to cut off the brightness values for each color component that are not found in the image.

Click Apply to perform the operation.

the light and dark parts of the image.

This correction can be performed repeatedly by sequentially changing the brightness distribution of the image pixels.

Gamma Correction

Menu: Raster – Processing the raster > Gamma Correction …

Command line: GAMMAEDIT

This functionality is available only in the Raster module.

This operation is used to enhance the overall quality of an entire image by changing its so-called “color profile” through an algorithm that changes the distribution of brightness of image points.

Points with intermediate values of color intensity are usually unevenly distributed and form a curve (gamma) of an arbitrary shape. The gamma value determines the slope of the curve exactly halfway between the white and black points. Changing the gamma increases or decreases the brightness of points that fall within a certain brightness range, as well as changes the brightness of red, blue and green colors.

• 1. Select raster images to be processed.
• 2. Run the command.

In the Gamma Correction dialog box that opens:

• 3. Set white and black points by selecting them with the eyedroppers on the image.
• 4. Select a color channel.
• 5. Create a gamma curve automatically or adjust it manually.
• 6. Click Apply.

Setting black and white points

To select the values of white and black points directly on the image, use the eyedroppers Black point and White point. Select the appropriate eyedropper and outline an area on the screen.

You can fine tune the color values in the Color point section of the dialog box.

If color values other than (0,0,0) for black and/or (255,255,255) for white points are specified in this section, then points outside the specified range will not be gamma corrected.

The Auto set button calculates the white and black point values for each color channel. The histogram in the middle of the dialog box displays the distribution of brightness of the colors. The curve can be changed for each channel separately.

Press to return the gamma curve to its original view, click .

Changing the gamma curve

There are three ways to change the shape of gamma curve of the Master channel and curves of individual channel curves. These ways are inteconnected.

• 1. Check or clear the S-Curve box. S-curve is a form of gamma curve that is always symmetrical about the center point of the distribution range. This means that by adding brightness to dark areas, we automatically darken bright areas to the same level and as a result the image looks better balanced in brightness. If the S-Curve checkbox is cleared, you can create a gamma curve by moving the overall brightness balance in the resulting image.
• 2. Entering a value in the Gamma field changes the slope of the curve.
• 3. Move the curve of the Master channel or curves for individual channels in the histogram window.

If the Master is selected in the Channel list, all curves are shown.

When you move the mouse cursor over the curve, a marker appears in the form of a point, which can be moved by changing the slope of the curve. If you hold the mouse cursor on the curve, you can see the real distribution of brightness of the image colors.

To change the shape of curves for red, blue and green channels, select the corresponding channel from the Channel list.

In some images you can also “split” the Master curve into components (red, blue and green) by clicking the Auto set button.

Editing Color Palette (Color Classifier)

Ribbon: Raster – Processing > Color classifier

Menu: Raster – Processing the raster > Color Classifier…

This functionality is available only in the Raster module.

With the help of edit operations, the color palette of the image can be calculated automatically or changed based on the colors specified by the user.

Colors in the image are managed in the Color Classifier dialog box. The dialog tools allow you to:

• reduce the palette by deleting selected colors or combining several colors into one;
• replace the selected colors;
• add new colors in the palette;
• save the customized palette for use when processing the same type of color images.

Set auto palette

Click the button. The program will automatically determine the set and number of colors that represent the image the most accurately.

Create a new palette

Set the number of colors in the Color Counter field or edit the palette in the Color samples table.

Click button. It is used to recalculate the palette based on the number of colors specified in the Color counter. The changes are controlled in the preview window. When the required result is achieved, click OK.

Color selection

Methods:

• specify the color in the Color samples table by the mouse; several colors are selected while holding down the SHIFT key;

• click button, specify the point in the image; the selected color will be highlighted in the table;
• Click button, select the area in the image with a polygon; colors contained within the specified polygon will be highlighted in the table (to close the polygon, press the right mouse button or ENTER).

You can control the selected colors in the image using the View selection tool.

Delete color from the palette

• 1. Select color(s) to be deleted.
• 2. Click Delete colors button.

You can reduce the number of colors by setting their number in the counter window and clicking button.

Add color to the palette or replace existing one

• 1. Select the colors to be replaced or specify an empty square if you need to add a color.
• 2. Click button and specify in the dialog that opens the color to be added or replaced.

Merging several colors into one

• 1. Select several colors in the table while holding down the SHIFT key or specify them in the image using tool.
• 2. Click button.

The Apply button starts application of the settings made.

Filtering Color Raster Images

Blur

Command line: GAUSSBLUR

This functionality is available only in the Raster module.

The filter has the effect of blurring the image, creating the impression that the image is slightly out of focus. Blur filtering reduces the image sharpness, but makes the image areas containing texture fills more uniform, which makes it easier to binarize or separate the image by color.

When calculating a pixel color, the program replaces its color value with the one averaged over the neighborhood.

Radius is the only filter parameter. The higher its value, the more blurry the image becomes.

1. Select images on the screen to be processed and start filtering in one of the following ways.

• 2. Run the command.
• 3. In the Radius field enter the value from 0.1 to 10.0 to set the degree of the image blurring. The higher the value, the stronger the blur effect.
• 4. Select Apply to start filtering.

Adaptive Blur

Adaptive blur allows, while maintaining sufficiently clear boundaries between contrasting colors, to smooth out areas of similar colors. This makes it possible to eliminate grain and remove “printing raster” in color and grayscale images.

The Radius parameter determines the number of pixels at the border of the transition of colors analyzed during the operation.

Run the command. In the Adaptive Blur dialog box:

• 1. In the Radius field, enter the value from 0.1 to 10.0 or regulate the parameter with the slider. The changes are controlled in the preview window;
• 2. When you get a desired result, click Apply to start the operation.

Contour Sharpness

Ribbon: Raster – Filters > UnSharp Mask

Ribbon: Raster – Filters > Unsharp mask…

Command line: UNSHARPMASK

This functionality is available only in the Raster module.

The filter finds the boundaries of color transitions in the image and sharpens them.

The filter changes the contrast of pixels at the boundaries of color transitions, producing an overall sharpening of the image. It can be used to correct images that are blurry as a result of interpolation (for example, after scaling, resizing, or calibrating).

• 1. Select the images on the screen to be processed.
• 2. Run the command.
• 3. Enter in the Sensitivity field or specify using the corresponding slider the value in percent (from 0 to 100). The higher the value, the stronger the filter effect.
• 4. Enter the Radius value. The radius determines the “depth” of the filter’s effect. The higher the value, the more pixels surrounding the transition border will be processed. Low values provide sharpening only at the boundaries themselves. To find the optimal value for the parameters of the entire image, first set them for a small typical area of the image in the preview window.
• 5. Select Apply to start filtering.

Median

Command line: MEDIANER

This functionality is available only in the Raster module.

Averaging filtering suppresses noise in the image by analyzing all pixels within a given radius and assigning the median value of the analyzed pixels characteristics to the central pixel. As a result, the color and brightness of the pixels are averaged.

• 1. Select the images on the screen to be processed.
• 2. Run the command.
• 3. Specify the Radius value in the range from 1 to 5. This value determines the radius of the neighborhood (in pixels) within which the filter will analyze color values.
• 4. To start filtering, select Apply.

Color Reducing

This functionality is available only in the Raster module.

This section describes the procedures for producing monochrome raster images (raster layers) from color and grayscale images.

The original image is a raster file obtained as a result of a color or grayscale scan. The binarization operation allows you to create monochrome raster images containing a black-and-white representation of objects in a color image. For example, objects of different colors (level lines, roads, rivers, etc.) can be sequentially extracted from one color image of a scanned map and placed on separate monochrome layers. This separation method allows you to place objects on one monochrome layer that correspond to one or more different colors in the original image.

Another way to get monochrome images from color ones is color layering. The program can convert a color image to a set of monochrome raster layers. At the same time, it is guaranteed that the black and white representation of all pixels of the original image will fall on this or that layer.

Binarization

Ribbon: Raster – Processing > Binarization

Menu: Raster – Processing the raster > Binarization…

Command line: BINARIZATION

This functionality is available only in the Raster module.

When performing binarization, a new monochrome image of the specified color is created and placed on the specified layer. Using a certain criterion, the program determines the pixels of the original image (color or grayscale), which should become black (image pixels) and white (background), and then generates a monochrome image and places it on a new raster layer. The criterion for dividing pixels into two sets is determined by the selected binarization method and its parameters (threshold values or a set of color range). Pixels are placed on the image, they are selected in accordance with the setting specified in the tabs of the Binarization dialog box. The new monochrome image is named Original Image Name_N, where N is an integer.

This operation can be applied to multiple images at the same time. If no image is selected, the command will be applied to all visible images located on unlocked layers.

Binarization works on images that have a display border. Using this property, you can limit the binarization area on any image by setting the display border for it.

Binarization methods

To convert color and grayscale images to monochrome, various conversion algorithms are used, called binarization methods. It is recommended to select the conversion method appropriate for the type of image.

Brightness threshold

The Brightness threshold method converts color pixels with brightness values above the specified level to background points, and below that level to image points.

This method can be used to convert both color and grayscale images. When converting a grayscale image, the program uses the grayscale of this image. When converting a color image, the grayscale is determined by the brightness value of colored points.

Threshold by RGB

When using the Threshold in RGB method, specify three separate thresholds for the Red, Green and Blue components. The program converts colored points with Red, Green and Blue values below the corresponding thresholds to black points (image points) of a monochrome image.

Brightness ranges

The Brightness ranges method converts colored pixels with any brightness value to image points. This method selects a number of base gray levels. These levels are used as the midpoints of the ranges. For each selected level, you can define the half-lengths of the ranges. The half-length of the range is the number of gray levels below and above the selected level.

Brightness ranges method converts pixels whose gray level is within the specified ranges to image points. The rest of the pixels are converted to background points.

This method can also be used to convert color and grayscale images. The gray level calculation for colored points is described in the Brightness Threshold section.

Ranges by RGB

Using this method, you can convert color pixels belonging to specified RGB ranges to image points.

To set the RGB range, first select the center color of the range. Red, Green and Blue components of this color determine the position of the center point of the RGB range. For each color component (R, G and B), specify the corresponding half-lengths of the ranges. The red, green, or blue range half-length is the

number of R, G, or B levels below and above the selected R, G, or B level. For example, if the R level for the selected color is 50 and the range half length is 10, then the RGB range contains colors with an R component from 40 to 60.

Ranges by HSV

The Range by HSV method simplifies the conversion of RGB color images. It converts pixels of similar colors to image points. Colors that are similar from the point of view of human perception are called analogous (red - orange, dark green - light green, etc.).

To convert an image using the Range by HSV method, specify one or more HSV ranges. The HSV range is determined by the selected color and the half-lengths of the H, S, V ranges. The HSV range is similar in structure to the RGB range. Hue is expressed as an angle from 0° to 360°, and Saturation and Value - as a percentage from 0 to 100.

When converting grayscale images, the results of applying this method are of poor quality.

Configuring binarization

To fine-tune the binarization procedure, you need to choose the appropriate method. The type of the selected method determines the tuning method. For each of the two threshold methods, you should assign one or three threshold values in the histogram. For any range method, you should specify a set of ranges of appropriate types that contain the colors to be extracted.

In addition, for any method, it is necessary to assign a layer on which the image obtained as a result of binarization will be placed.

The Binarization dialog box.

Until options are specified, the preview shows the original view of the image. But as soon as the selection of parameters begins, the results of binarization settings will be dynamically displayed in this window.

Setting range binarization

In the Binarization dialog box, select the Range tab.

In the By list, select the desired type of the range method.

The adjustable parameters depend on the selected method.

Specify a set of ranges that capture the colors of those color image data (objects) that should be transferred to a separate monochrome layer.

Adding a range

Click or button and specify in the image a pixel of the color of the data to be binarized . Or click button and specify the area. An elements corresponding to the created range will appear in the Colors list. By successively clicking the buttons and indicating similar colors, achieve an acceptable result in the preview window.

When configured, the preview window dynamically displays the results of adding each range. The most reliable results are obtained by viewing the image at a scale of 1:1.

Changing parameters of the created range

If adding a range causes unwanted capture of image pixels, you can change its parameters using the channel histogram sliders or sliders and input fields in the Ranges section

.

Deleting a range

If you cannot achieve acceptable results by adjusting the range or it is selected by mistake, select it in the Colors list and click Remove color.

Use the Noise reduction slider to improve the binarization quality.

To binarize large “filled” areas, increase the value of this parameter (closer to Low mark) to reduce the amount of raster “garbage” and unfilled holes in the objects obtained during binarization.

To obtain a monochrome image of small or thin objects (texts, symbols, level lines or grids), decrease the value of this parameter to prevent distortion of the shape of small objects.

Set the color and name of the layer on which the binarization result will be placed:

• enter the name in the Layer field;
• specify the layer’s color by clicking on its swatch. Select Color in the windows that appears. Click OK.

Configuring threshold binarization

In the Binarization dialog box, select the Threshold tab.

In the By list, select the required method

Find threshold values for the selected method.

With the selected Threshold by brightness method, select Gray in the Channels list to see the gray level histogram. Specify the threshold value using the slider or G slider.

With the selected Threshold by RGB method, configure the threshold methods R, G and B. For this you can also use sliders on histograms of Red, Green and Blue channels.

Use the Noise reduction slider to improve the binarization quality.

Specify the color and name of the layer on which the binarization result will be placed.

Enter the name in the Layer field. To set the color, click on its swatch and make a selection in the window that appears. Click OK.

The binarization procedure is started by the Apply button.

The binarization result is a monochrome image of data having black and close to it pixel colors on the original color raster.

Adaptive Binarization

Ribbon: Raster – Processing > Adaptive Binarization

Menu: Raster – Processing the raster > Adaptive Binarization…

Command line: FONESMOOTHER

This functionality is available only in the Raster module.

It is recommended to binarize scanned images made in grayscale mode from originals with a nonuniform background (blue, sepia) using the Adaptive Binarization tool. This operation combines binarization and image quality enhancement procedures. The program analyzes the boundaries of color transitions in an image and distributes pixels to the background and information.

The command can be applied to color images as well.

All parameters of the Adaptive Binarization dialog box are set using sliders or by entering values into the corresponding fields.

Radius is a “depth” of the filter effect. The higher the value of the radius, the more pixels surrounding the border of the color transition will be taken into account in the processing.

Smooth – smooths the background and makes the image objects more distinct, and removes “garbage” when performing binarization.

Binarize – creates a monochrome image of selected information.

When satisfactory results are achieved in the preview window, click the Apply button.

Color Separation

This functionality is available only in the Raster module.

The operation allows you to separate points of a color image into disjoint sets – categories.

Each category is defined by a set of base colors. The purpose of such procedures is to highlight the colors by which the original image was made. Usually, objects of the same kind are indicated by one color, therefore, by dividing image points by color, you get the opportunity to select the necessary image objects into separate monochrome raster layers (images).

The example shows the creation of monochrome background layers, rivers and level lines using the procedure of separation by color.

• 1. Run the command**.**
• 2. Create a set of categories using the Add category button:

Use the eyedropper to indicate an object on the image, the points of which should fall into the category being created.

On the screen the Category Name and Color window will appear, in which you can assign a symbolic color and a category name.

• 3. In the same way, use the eyedropper to indicate the level lines. Specify the layer name Level lines and select the color – dark brown.
• 4. Add basis colors to the Level lines category. Click the eyedropper button in the Colors field and specify point on the image. The color of the specified point will be added to the list of the Colors field.

In case of an error when choosing a basic color, delete it. The error in the color selection is determined by the separation results, which are dynamically displayed in the preview window.

5. In the same way, create a category for rivers and set the basic colors.

• 6. To process a larger number of images of the same type, it is convenient to save the color separation parameters in a special file. Select the Template button and save the setting in the template. The saved setting can be later loaded to work with another image.
• 7. Having achieved an acceptable separation quality, click the Apply button.

Color Reduction

Menu: Raster – Processing the raster > Color Reduction…

This functionality is available only in the Raster module.

The procedure for reducing the number of colors is the same as for separating by color described in the previous section. The only difference is that you do not need to define the category names, since the procedure for reducing the number of colors does not create additional layers.

The procedure is set up in the same way as in the separation procedure. The purpose and functions of elements of this dialog box are identical to the purpose and functions of the elements in the Color Separation dialog box.

Monochrome Filtering

This functionality is available only in the Raster module.

Monochrome filters are used to process monochrome (black and white) images. Monochrome images are also called bitonal because they only use one bit of information to describe the color of each pixel (black or white).

Filtering operations can be applied to images located on visible and unlocked layers. One or several raster images can be filtered.

Removing “Speckle”

Ribbon: Raster – Filters > Speckle Remover

Menu: Raster – Filters > Speckle Remover…

Command line: SPECKLEREMOVER

This functionality is available only in the Raster module.

The filter removes raster objects (isolated groups of pixels), the size of which is less than the specified value, and automatically estimates the size of spots in the image. This filter can be used after binarization or separation procedures to remove small raster objects from the resulting monochrome layer images.

Fragment of a monochrome raster obtained as a result of binarization

After “speckle” removal

8. Select images to be processed and run the Speckle Remover command.

• 9. Enter the maximum size of raster objects to be deleted in the current units (left input field) or in points (right input field).
• 10. Or check the Auto estimating box to let the filter estimate the size of the raster “speckle” before filtering.

To measure the size of a raster object on the screen, click button and specify a point inside the object, or click button and specify two points on the screen. The program will set the “speckle” size equal to the distance between the specified points.

11. Click Apply.

Filling “Holes”

This functionality is available only in the Raster module.

The filter fills small holes in raster objects. In this case, only those “holes” are filled, the size of which is less than the specified value. This filter can automatically estimate the size of holes in image objects.

The filter can be used after binarization or separation procedures to fill in unwanted holes in raster objects.

Original image

After removing “holes”

1. Select images to be processed and run the Hole Remover command.

• 2. Enter the maximum size of the raster “holes” to be removed in the current units (left input field) or in points (right input field).
• 3. Or check the Auto estimating box to let the filter automatically estimate the size of raster “holes” before filtering.

To measure the size of the raster “hole” on the screen, click button and specify a point inside the “hole”, or click button and specify two points on the screen. The program will set the “hole” size equal to the distance between the specified points.

4. Click Apply.

Smoothing

This functionality is available only in the Raster module.

The filter smooths outlines of raster objects, fills in edge and internal background speckles, and also partially removes raster “garbage”.

Fragment of the original image

After smoothing

Work of the Smoothing filter consists of two stages. At the first stage it works like the Median filter on a grayscale image, analyzes the neighborhood of a given radius of each pixel and replaces the brightness of the central pixel with the average brightness of the neighborhood. As a result, the outlines of the raster object are blurred. Increasing the Medianning value makes gray strips wider.

At the second stage, the filter converts to black those pixels whose brightness value is less than the value of the Threshold parameter. Wide black lines show the binarization boundaries. Pixels inside boundaries turn black, outer pixels turn white. Increasing the Threshold value makes the objects thicker, while decreasing makes them thinner.

First stage – blurring edges

Second stage - binarization

1. Select images to be processed and run the Smoothing command.

• 2. Set the Medianning the degree of medianning in percent. This value determines the degree of blurring the outlines of raster objects.
• 3. Set the Threshold gray level from 0 to 255. This value determines the degree of cropping of blurred edges. Higher values increase the amount of smoothing, but the raster lines get thicker.

Use the preview window to select the optimal values of the parameters.

4. Click OK.

Thinning

Ribbon: Raster – Filters > Thinning

Menu: Raster – Filters > Thinning…

Command line: THINNING

This functionality is available only in the Raster module.

The filter in one pass thins raster fragments by one point (pixel) simultaneously in all directions. This filter has an additional parameter that allows you to thin raster objects to the skeleton (only the pixels located in the middle remain).

After thinning to skeleton

Select images to be processed and run the command.

Specify the number of filter passes, or select Thin image up to skeleton to turn all objects into one-pixel lines.

Click Apply.

Thickening

This functionality is available only in the Raster module.

This filter thickens raster objects by a specified number of pixels in a specified direction - horizontal, vertical, or simultaneously in all directions.

The pictures below show how thickening works when all directions are selected. On the left there is the original raster image, and on the right there is the result of applying the filter after three passes.

Thickening filtration in all directions

The figure below illustrates how the Thickening filter works when only the vertical direction of the thickening is selected. On the left there is the original raster image, on the right there is the result of applying the Thickening filter after three passes.

Vertical thickening filtration

Select the images to be processed and run the command.

Select the required direction of thickening – Horizontal, Vertical or All (horizontal, vertical or diagonal). Click Apply.

Contour

Contour filters turn filled raster areas into single-pixel contours. In a four-connected contour, adjacent pixels can only be joined by the sides, in an eight-connected contour, contiguity is also allowed at the corners, so such contours look thinner.

The images below show the results of applying contour filters.

Parts of contours obtained with four-connected and eight-connected filtering

Select the images to be processed and run the command.

Inversion

This functionality is available only in the Raster module.

Inversion changes the color value of each point of a monochrome image to the opposite (creates a “negative image”). The background points become the image points, and the points transmitting the image of the objects - the background ones. The figure below shows the result of applying the filter.

Result of applying the Inversion filter

Select the images to be processed and run the command.

Selection of Data in Raster Images

This functionality is available only in the Raster module.

To edit the content of monochrome raster images, apply the raster selection. Using various types of selection, you can select on raster images:

• raster objects raster lines, arcs and circles;

• areal fragments of a raster image;

• raster lines segments fragments of raster lines of any shape limited by points of intersection with other raster lines or endpoints;

• isolated raster fragments multiple interconnected raster points.

The Order of Raster Selection

• set the selection mode: add data to selection, remove from selection, single selection;
• run one of the selection methods: specifying, within the rectangle, clipping polygon, etc. The selection method is assigned based on the type of raster data being selected;
• carry out the procedure for pointing on the image in accordance with the running method.

Quantitative Raster Selection Modes

Ribbon: Raster – Raster Selection – Basic methods > …

Menu: Raster – Raster Select > …

Basic Selection Modes

Ribbon: Raster – Raster Selection > …

Menu: Raster – Raster Selection > …

906

Object Raster Selection Methods

The trace selection of raster data is based on the principle of recognition of individual raster objects: raster lines, arcs and circles in the selected area or on its border.

Trace Selection Methods.

Trace selection of raster data is based on the principle of recognizing raster line segments in a given area or before crossing with other segments.

Fill Selection Methods

The fill selection of raster data is based on the principle of recognition in a given area of contacting (merging) raster points.

Setting Selection Options

For correct recognition of raster objects when using the object selection methods, it is necessary to make settings in the Options tab of the Conversion options dialog box:

Selection options for recognizing raster objects

The values of Max. thickness and Max. break parameters can be measured in a raster image using buttons next to the fields.

Separation by Type and Size

Ribbon: Raster – Processing > Separation by size

Ribbon: Raster – Processing > Separation Linear Objects

Ribbon: Raster – Processing > Separate Text Areas

Menu: Raster – Processing the raster > Separation by size

Menu: Raster – Processing the raster > Separation Linear Objects

Menu: Raster – Processing the raster > Separate Text Areas

Command line: SEPARATIONBYSIZE

Command line: SEPARATELINEAROBJECTS

Command line: SEPARATETEXTAREAS

This functionality is available only in the Raster module.

Selection operations allow you to transfer specific monochrome raster objects to new raster images placed on specified layers. From the original image, you can select hatches, text, linear objects and objects by size (isolated groups of adjacent points).

When performing an operation, the program finds objects of the specified type with the specified parameters in the image and transfers them to a new raster image. The new raster image created as a result of the operation has the same parameters (size, insertion point, resolution, scale) as the original one, but is placed on the specified layer. In this case, the loss of objects removed from the original image does not occur - they are simply moved to a separate raster layer.

The separation procedures can be used:

• instead of speckle remover, when it is necessary to save small-sized image objects, which the program can attribute to raster “garbage”;
• when it is necessary to use operations only to objects of a certain type, for example, editing texts or hatches.

After making selection, the displaced objects can be saved as a separate raster image, selected and returned to the original image, or the entire resulting raster layer can be deleted.

The order of performing object separation operations

• select the images to be processed. If no selection is made, the operation will be applied to all visible images located on unlocked layers;
• Run the appropriate command;
• In the dialog box that opens, configure the parameters by which the selection of objects will be carried out. The type and settings of the dialog box depend on the type of the selected object. The buttons in the dialog boxes located next to the fields are designed to measure the set parameters in the image. To measure the object parameter in the image, click the button and indicate the required on the screen. The measured value appears in the corresponding field. The principle of filling in the Output layer field, in which it is necessary:
  • enter the name of the layer on which the created image with the selected objects will be placed;
  • set the color of the created layer by left-clicking on the color swatch located next to the field and choosing a color in the dialog box that appears.

• The settings results are controlled in the preview window of the dialog box. When satisfactory results are achieved, click Apply.

Setting parameters in the dialogs of object separation commands

Separation by size

• set the minimum and maximum size of objects in the corresponding fields;
• select linear objects;
• in the Max.width field, set the value of thickness of the raster line of objects to be selected;
• in the Max. break field, specify the ignored break size in lines.

Separate text areas

• in the Max.width field, set the maximum line width of the raster text;
• in the Text Height field, specify the maximum size of an upper case letter in the text;
• in the Text Orientation field, select from the drop-down list the orientation of texts Horizontal, Horizontal and Vertical or Arbitrary Oriented;
• if necessary, check the boxes Overlapped texts and Single Letters.

Saving and deleting the created raster image

The created raster image with the selected objects automatically receives the status of embedded in the document and the name of the layer specified in the Output layer field.

Rasterization

The following commands rasterize the selected vector and raster data onto underlying raster images. Selection can consist of:

• raster images;
• raster selection;
• vector data.

When these rasterization commands are executed, only the part of the selection that is located within the bounds of the underlying image is rasterized. The part of the selection outside the destination image bounds is ignored.

The commands will rasterize the selection to all underlying visible images located on unlocked layers.

Merge

Add selected data to raster image.

• Command line: MERGE
• 1. Select data for rasterization.
• 2. Place it over the destination image.
• 3. Run the command Merge.

Source data will be deleted.

Merge a Copy (Rasterize)

Add the copy of data to raster image and save original.

• 1. Select data for rasterization.
• 2. Place it over the destination image.
• 3. Run the command Merge a Copy (Rasterize).

Copy of selected data will be rasterized. Original data will be kept.

Raster Drawing

With pixel-based painting tools, you can draw and erase raster lines on monochrome, grayscale, and color images. When working on a monochrome image, these tools allow you to paint images with the

913

color or erase – that is, draw with the background color. When working on a color or grayscale image, you can choose the color (grayscale) that will be used for drawing.

The fill tools can work on both monochrome and color images and let you fill raster areas within closed contours with the color of raster objects and erase isolated raster objects by filling them with the color of the raster background.

Pixel drawing

Tools for pixel drawing used for drawing and erasing rater data of an image.

To draw on raster image:

Options:

Color Setting the color of the pencil. The option opens the Select Color dialog box.

The option is available only if the drawing contains color raster images.

On monochrome images, the raster line will be displayed in the main color. On grayscale images - by one of the shades of gray that most closely matches the selected color in terms of brightness.

On color images - by the selected color.

Command prompts:

Current color <R: 255 G: 255 B: 255> or [Color]: Shows current color and an option to change it. Draw or input pencil width <1.00000>: Shows current width of a line. Can be changed with user input. Draw curve on raster or “Enter” to finish: Draw a line.

To erase on raster image:

Options:

Color Setting the color of the eraser. The option opens the Select Color dialog box.

The option is available only if the drawing contains color raster images.

On monochrome images, the raster line will be displayed in the main color. On grayscale images - by one of the shades of gray that most closely matches the selected color in terms of brightness.

On color images - by the selected color.

Command prompts:

Current color <R: 255 G: 255 B: 255> or [Color]: Shows current color and an option to change it. Draw or input eraser width <1.00000>: Shows current width of a line. Can be changed with user input. Draw curve on raster or “Enter” to finish: Draw a line.

Floodfilling

The floodfilling tools allows the filling of closed raster outlines with the raster object color (foreground) and also to erase isolated raster objects by filling them with the raster background color.

The commands only work on a monochrome raster image.

To fill rater contour with flood:

Specify a point within a closed outline on a monochrome raster image.

To erase with flood:

Specify any point in the isolated raster area to be erased.

Editing Raster Text

Ribbon: Raster – Rasterize > Editing Raster Text

Menu: Raster – Rasterization > Editing Raster Text

Command line: RTEXTEDITCMD

This functionality is available only in the Raster module

1. Run the command

In the image set a rectangular area above the text being edited. In the process of specifying the area, you can specify its parameters Base point, Area angle (coordinates of the second point) in the corresponding fields of Options window and the text rotation angle in the Angle field.

• 2. Specify text base point (bottom left corner of the text).
• 3. Specify text base line.
• 4. Draw a rectangle around the raster text, the content of which will be removed from the raster.

• 5. Enter a text line in the command line or edit the recognized one, if text recognition is enabled.
• 6. Select the text insertion or rasterization mode.

Digitizing Raster Data (Vectorization)

This functionality is available only in the Raster module

• Trace (semi-automatic vectorization);
• Automatic vectorization.

Trace (Semi-Automatic Vectorization)

This functionality is available only in the Raster module

Tracing is an interactive procedure that allows you to vectorize a raster. Tracing is based on the technology of local recognition of raster geometric entities. Using this technology, the program

identifies raster lines as a line, arc or circle and generates the corresponding vector objects. You specify raster on the image, and the program creates vector objects that approximate the selected raster images. As a result of tracing, a vector copy of the raster object is created.

When tracing, you have the ability to transform and place only selected objects on different layers, as well as obtain vector objects with lineweights that depend on the width of the raster line.

Tracing is performed only on monochrome raster images.

The trace order:

• configure tracing parameters;
• select the tracing mode;
• select the method corresponding to the object being vectorized;
• specify object in the image.

Configuring Trace

During tracing, local recognition of vector objects on a raster image occurs. For correct identification of objects, the program always uses two numerical parameters that determine the maximum width and value of the ignored raster line break, as well as the parameter that sets the degree of acceptable deviations of the shape of raster objects from vector prototypes.

Trace is configured in the Conversion Options dialog box, in the Options and Trace tabs.

Best parameters when configuring trace

Set the values of parameters by entering them in the appropriate fields or measure them directly in the image by pressing the ruler button.

Click OK.

Options tab

Trace configuration options

Trace Modes

While trace methods determine how to recognize objects in the raster and which objects to recognize, the trace mode determines the final result of tracing – what should be the output.

The trace result can be:

• Creating vector objects ( Make Vector and Keep Raster),
• Deleting raster objects when creating vector ones ( Make Vector and Erase Raster),
• Deleting raster objects ( Erase Raster),
• Smoothing raster objects ( Smooth Raster).

The trace mode remains active till you choose another mode.

Make Vector

This mode creates vector objects while saving the original raster – the raster image is vectorized. The Make Vector and Keep Raster trace mode is used by default.

Make vector and erase raster

Menu: Raster – Trace > Vector (erase raster)

Command line: MAKEVEKTORANDERASERASTER

In this mode trace creates approximating vector objects and deletes the traced parts of raster lines. The mode converts raster objects into vector ones.

Erase raster

Ribbon: Raster – Trace – Mode > Erase Raster

Manu: Raster – Trace > Erase Raster

Command line: ERASERASTER

In this mode trace removes the traced parts of the raster line and does not create vectors. Intersections or parts of other raster objects under the traced data remain unchanged.

Smooth raster

Command line: SMOOTHRASTER

In this mode trace smooths raster lines.

Trace Methods

The trace method is determined by the type of vector object used to approximate the raster line. You can trace objects using automatic detection or indication of the object type.

Trace with Automatic Detection of Object Type

Ribbon: Raster – Trace – Trace > Trace Auto

Menu: Raster – Trace > Trace Auto

Command line: TRACEAUTO

This functionality is available only in the Raster module

This method is used for tracing raster entities: line segments, arcs and circles.

Raster entities are raster objects that coincide in shape with basic vector objects – vector image entities (lines, arcs, circles). We will use the terms of raster circle, raster arc and raster line to mean circular, arc and line-shaped raster objects. Raster entities can also be defined as objects obtained by rasterizing the corresponding vector analogs, which happens, for example, when a vector drawing is output to a printer. It should be borne in mind that real raster entities may have defects that make it difficult to identify them by the program, for example, breaks, elliptical circles and arcs, uneven width, etc.

When tracing with automatic object detection, after specifying a point on the raster line, the program selects the type of vector object most suitable for approximating the specified raster line (line, arc or circle) and traces the specified raster object. If this object cannot be approximated by any of the above objects and its width exceeds the value set in the Max. thickness field of the Options tab, the program creates its vector contour, consisting of closed polylines.

When tracing a raster line by indication, the program determines the type of raster entity (line, arc or circle) and tries to trace the object as large as possible. In this case, the intersections of the entity with other objects are ignored and the set of points of the object is extended as long as the shape of the raster object corresponds to the recognized entity. Since real raster objects may differ from ideal raster entities, the recognition algorithm uses the Accuracy parameter, which sets the acceptable degree of mismatch between the shape of real raster objects and ideal ones.

Trace with automatic detection of the object type

• Run the command;
• Click on the raster object to be traced.

The point should be selected on the least distorted and largest segment of the raster object, outside of its intersections with other objects.

If the program recognizes a raster entity, then it will be traced with the most suitable vector object.

If the specified object is not a raster line, arc, or circle, a vector contour approximating it will be created. The graphics editor distinguishes raster lines.

Forced Trace

This functionality is available only in the Raster module

Forced tracing allows you to replace raster objects with vector ones of a specified type. The user selects one of the three object types (line, arc or cirle) to be used as reference object for tracing (line, arc or circle), and then specifies points on a raster line, as if drawing a vector entity above the raster entity. The program tries to trace the specified object with a vector of the same shape. If possible, the raster line is traced. Unlike the TraceAuto command, these methods allow you to trace a part of a raster entity, as well as entities with significant shape distortions.

If the object width exceeds the value of Max. thickness parameter, or its shape cannot be approximated by the object of the specified type, then it is not traced.

When tracing by Line and Arc method, the ends (both or one of them) of reference lines and arcs can be specified outside the traced raster objects – on their imaginary continuation. In this case, the raster object is traced to its endpoints. If the ends of the reference objects lie on a raster object, then the part of the raster object located under the reference object is traced. If you want to recognize the line up to its finite limits, Check the Auto-extend vectors box in the Trace tab of the R2V Conversion Options dialog box.

Trace Line

Command line: TRACELINE

• Run the command;

• specify the beginning and end of the line lying on the raster object to be traced.

If the program cannot recognize the object under the vector line, a beep sounds.

Trace Arc

Command line: TRACEARC

• Run the command;

• Specify the first, middle and end points of the arc, lying on the raster arc to be traced.

To trace the next raster arc, repeat steps 1 and 2.

Trace Circle

Command line: TRACECIRCLE

• Run the command;

• Specify two points – the ends of diameter of the raster circle to be traced.

To trace the next circle, repeat steps 1 and 2.

Trace Outline

Command line: TRACEOUTLINE

The Trace outline method is intended for tracing raster objects with closed boundaries. As a result of this operation, a closed polyline is created, it reproduces the object outline and the hatched area inside the outline.

To trace the outline:

• In the Trace tab of the R2V Conversion Options dialog box, set parameters for tracing outline with the selected method;
• Run the command;
• Specify a point inside the outline.

The outline trace results depend on the parameters set in the Trace tab of the R2V Conversion Options dialog box.

Trace options

Trace Polyline

Ribbon: Raster – Trace – Trace > Line Following

Menu: Raster – Trace > LineFollowing

Command line: LFCMD

This functionality is available only in the Raster module

The method belongs to the forced trace commands.

The method allows you to trace arbitrary raster lines, approximating them with a vector polyline. When tracing, specify a point on the raster line, and the command automatically follows this line (the tracing direction must be determined) to the nearest node point of the line or to the intersection point and creates an approximating vector polyline - a polyline consisting of linear segments. The node point of a raster line is either its endpoint or the point of intersection with another raster object. Thus, in one step,

a part of an arbitrary raster line bounded by two node points is traced – a raster polyline segment. One trace step can create an arbitrary number of segments, the number of which depends on the line complexity and the trace settings.

After tracing each polyline segment, select the further direction of tracing or complete the procedure. Direction is selected by specifying a point on the next raster curve segment adjacent to the last traced segment. The steps and polyline segments can be undone.

The command provides for the mode in which the direction of continuing trace is determined automatically. In this mode, the program offers one of possible directions by showing a special marker on the segment selected as a continuation. Within a specified period of time (by default, the waiting time is 0), you can choose another direction or agree with the proposed one. If the direction is not selected, then after the waiting time has elapsed, the program will automatically continue tracing in the selected direction.

Setting auto detect direction

The tracing of raster polylines is influenced by the parameters set in the R2V Conversion Options dialog box on the Options tab: Max. thickness, Max. Break and Accuracy, which specify the maximum thickness of the raster polyline, the size of the ignored break, and the accuracy of the raster polyline approximation.

The color and width of the polyline created by tracing depend on the checked Use table box and specified values of parameters in the Separation tab in the R2V Conversion Options dialog.

Setting parameters of vector polyline being created

Context menu commands when tracing polyline

The order of tracing a polyline

• 1. Set the command parameters in the R2V Conversion Options dialog box;
• 2. Specify the trace mode (Make vector, Erase Raster etc.);
• 3. Run the polyline trace comamnd;
• 4. Pick points on the raster curve. When the Auto Detect Direction box is checked, a “rubber” line is drawn from the mouse cursor showing the current trace direction;
• 5. Manage the tracing process from the context menu;
• 6. To complete the polyline creation, select Enter or press ENTER;
• 7. To exit the command, press ESC.

Tracing polyline with orthogonal segments

When tracing raster polylines, you can use the orthogonalization mode, which allows you to align the generated polyline segments at a predefined base angle:

• 1. In the R2V Conversion Options dialog box, when tracing, check the Line Follow Orthogonalization box on the Trace tab;
• 2. Check the Auto Estimate Base Angle box or set the first segment direction in the Base angle field (or using button measure it on the screen);
• 3. Decrease the Accuracy value in the Options tab to avoid creating superfluous segments;
• 4. Select the required trace mode and run the polyline trace command;
• 5. Specify the point on the screen to start the trace.

Results of trace with orthogonalization

Setting orthogonalization when tracing

Automatic Vectorization (Raster to Vectors)

As a result of automatic vectorization, vector objects are generated that approximate the raster image.

Vectorization can work on a raster selection set or on a group of images. If you have created a raster selection set and started vectorization, it will start to run on this set.

If there is no raster selection, and several images are inserted into the current drawing and they are available (visible and located on unlocked layers), then when performing vectorization, you need to select a group of raster images.

Vectorization also works on images that have a display border. Using this property, you can limit the vectorization area on any image by setting a cropping border for it.

Vector object thicknesses can be rounded to predefined values. Vector objects that correspond to raster lines of different thickness can be placed on different layers or assigned different colors.

The order of vectorization:

• 1. configure vectorization parameters (R2V Conversion Options);
• 2. select images to be converted;
• 3. start the automatic vectorization.

Vector objects appear above the source raster.

Conversion Options

Ribbon: Raster – Conversion > R2V Conversion Options

Menu: Raster – Convert > Conversion options

Command line: R2VSETUP

This functionality is available only in the Raster module

Raster selection and vectorization are configured using the R2V Conversion Options dialog box. The main options are set in the Recognition and Options tabs. If text search and recognition algorithms used, then it is also necessary to configure options specified in the Texts tab described below in this chapter.

The thicknesses of the objects obtained during vectorization can be rounded to predefined values, as well as placed on different layers and assigned different colors. These operations are performed using the Separate tab.

Recognition tab

In this tab you can select a set of algorithms that will be used when vectorizing a raster image.

Recognition tab

Different types of raster images should be vectorized using different sets of entities. For example, to vectorize images of maps or sketches, you should use polylines that approximate raster curves of arbitrary shape, and when vectorizing engineering drawings - algorithms that create segments, circles, arcs, then the vector drawing will most accurately reproduce the original drawing.

To ensure the possibility of optimal recognition of images of different structure, the Graphics Editor uses several vectorization algorithms that recognize raster analogs of vector entities and generate approximating vector objects of the corresponding types. You can use one or more recognition algorithms in one vectorization operation.

The tree of algorithms is located in the right part of the Recognition tab. At the first level, there are recognition algorithms. To enable the desired algorithm, check the box next to its name. The second level of the tree contains additional functions and parameters of the corresponding algorithms.

The vectorization process is also directly influenced by the geometric recognition parameters located in the Options tab of this dialog box.

Points

This algorithm is used to recognize geodetic marks and similar labels on specific images, for example, on geographic maps.

The graphics editor recognizes as a point an object with dimensions of at least 2x2 pixels (smaller objects are considered “speckle” and are deleted) and no more than the Max. thickness vertically and horizontally.

The Point type object can be recognized only in an automatic mode.

It is not recommended to use this algorithm when recognizing images of low quality or containing many small spots and “speckle”.

Lines

Includes line segment recognition algorithm. As a result of the algorithm wor, straight lines are created.

The figure illustrates the results of vectorization using a single enabled Lines algorithm. The original raster fragment is shown on the left; the right drawing shows the result of vectorization with disabled display of object thicknesses.

Lines vectorization

It has the following additional options:

• Arrows when the option is enabled, it recognizes dash and dash-dot straight lines, creating vector lines with the corresponding line type;
• Line Types when the option is enabled, it recognizes raster analogs of dimension lines (lines with one or two arrows at the end points) and, if the arrows are found, saves them as end markers of the line.

This algorithm is influenced by the following parameters of the Options tab.

The algorithm recognizes raster objects, the length of which is greater than the value of the Min. length option, and thickness is less than the value of the Max. thickness option.

The Max. break option sets the maximum length of an ignored raster line break.

The Approximation Accuracy option sets the accuracy of approximation of raster lines by vector objects. If the raster quality is poor, the option value should be decreased so that the algorithm recognizes raster objects with significant shape distortions.

When the Orthogonalization parameter is activated, the recognized line segments are aligned parallel and perpendicular to the direction specified in the Base angle field. Sections with small angles of deviation from the corresponding directions are aligned. The acceptable deviation for orthogonalization is determined by the Accuracy parameter. The closer this parameter is to the Low value, the greater the deviation can be.

Arcs and circles

Enables the algorithm for recognizing raster circles and arcs.

The figure illustrates the results of vectorization using the Lines and Arcs and Circles algorithms. The original raster fragment is shown on the left; the right drawing shows the result of vectorization with disabled display of object thicknesses.

Circle vectorization

It has the following additional options:

• Arrows – when the option is enabled, the program searches for raster analogs of dimensional arcs (arcs with one or two arrows at the end points) and, if arrows are found, creates the corresponding dimensional objects;

• Line Types - when the option is enabled, the program recognizes dash and dash-dot arcs and circles, creating vector objects with the appropriate line type.

This algorithm is influenced by the following parameters of the Options tab.

The algorithm recognizes raster circles and arcs that are larger than the Min. length, and thickness is less than the value of Max. thickness.

The Max. break option sets the maximum length of an ignored raster arc and circle break.

The Approximation Accuracy option sets the accuracy of approximation of raster arcs and circles by vector objects. If the raster quality is poor, the option value should be decreased so that the algorithm recognizes raster objects with significant shape distortions.

Polylines

This algorithm approximates the center lines of raster objects by polylines. The algorithm creates polylines consisting of straight segments only. You can use this algorithm alone or in conjunction with the Outlines algorithm when vectorizing maps and other images consisting of arbitrary lines (i.e., handdrawn lines).

It has the following additional options:

• Line Types when this option is enabled, the program recognizes dash and dash-dot arcs and circles, creating vector objects with the appropriate line type;
• Create Vertex on Intersection when this option is enabled, the algorithm creates vertices at the intersections of polylines.

This algorithm is influenced by the following parameters of the Options tab.

Max. length determines the maximum length of the recognizable segment added to the polyline. Longer segments are not added to the polyline. This allows, for example, to trace curved contours on maps that intersect with coordinate lines; the algorithm automatically stops at the intersection of the traced contour with a long straight coordinate grid line.

Approximates raster lines that are thinner than the Max. thickness.

The Max. break option sets the maximum length of an ignored raster line break.

The Approximation Accuracy option sets the accuracy of the raster curve approximation by the vector polyline.

Outlines

This algorithm is designed to approximate the outlines of filled areas by polylines by creating closed polylines that approximate the boundaries of raster objects. Contour polylines consist of straight segments only.

The figure shows the results of vectorization using the only enabled Outlines algorithm. The original raster fragment is shown on the left; on the right there is the result of vectorization.

Outlines vectorization

This algorithm is influenced by the following parameters of the Options tab.

The algorithm approximates raster lines that are thicker than the Max. thickness. To get the outlines of all raster objects, enable only this algorithm and set the Max. thickness equal to zero.

The Max. Break option sets the maximum length of an ignored raster line break.

The Approximation Accuracy option sets the accuracy of approximating the raster object boundaries by vector polylines.

Text areas

Enables the text recognition algorithm. The program finds image fragments containing raster texts and applies an operation specified as an additional parameter of the algorithm to the found texts.

The settings of search for raster texts and the OCR module are configured in the Texts tab of the same dialog box. The setting is described below in this chapter.

Options of text areas vectorization

• None does not vectorize raster texts. Areas containing found raster texts are not vectorized.
• Text Areas creates vector rectangles that limit raster texts. Areas containing found raster texts are not vectorized. Vector texts can be entered manually using the procedure for viewing and correcting recognized texts described later in this chapter.
• Polylines approximates the center lines of raster texts with polylines.
• Outlines approximates the boundaries of raster texts with outline polylines.
• OCR recognizes raster texts and creates text objects.
• Raster recognizes raster texts without creating vector text objects.

The algorithm uses the values in the Text Height list of the Options tab as the maximum height of upper case raster text characters.

Options tab

To optimally configure the raster entities recognition algorithms on a certain raster image, you can use information on measurement values of raster objects in the Options tab.

Options tab

Options values can be entered from the keyboard or measured on the screen.

Measuring option value on the screen

Click the button located next to the corresponding dialog field ( or ).

Specify two points in the image.

In the process of measurement, the program draws a “rubber” line connecting the specified points, therefore, in the text below the process of measurement in the image will be defined by the draw a line term. At the end of the measurement, the value is entered into the corresponding field.

Raster vectorization options

Separate Tab

Using the Separation tab of the R2V Conversion Options dialog box, you can manipulate the properties of the vector objects being created: calibrate the thicknesses of the resulting vector objects, distribute vector objects corresponding to raster lines from the specified ranges of thicknesses to different layers and / or assign different colors to such objects.

Separate tab

Separation options

Thicknesses Table

The criterion for dividing vector objects by layers and / or by colors is the thickness of the original raster lines.

The table specifies the intervals of widths of the image raster lines, which, during vectorization, will be converted into vector objects with a specified line width and a specific color, and then placed on the specified layers.

The thickness table can contain an arbitrary number of elements – thickness intervals. Each interval is determined by two values - the lower and upper boundaries of the thicknesses of vector objects that fall into this interval. Each interval is assigned properties (thickness, color and layer), which are assigned to objects with thicknesses that fall within the interval.

All intervals are created within the thickness range from zero to the Max. thickness specified in the Options tab of the same dialog box. The initial contents of the table are as follows:

Original content of the thicknesses table

Creating a new interval in the thicknesses table

Enter the value of the upper limit of the thickness interval in the New Interval field, click New Interval button.

You can create an interval by left-clicking on the desired place in the thickness table ruler.

Creating a new interval in the thicknesses table

The existing interval, which contains the specified thickness value, will be split into two intervals. The properties of the new interval are inherited from the existing one. An arrow appears on the ruler to show the position of the upper boundary of the created interval.

Specifying the upper limit of the specified interval

Changing the interval boundary

You can change the boundaries of all intervals, except for the upper boundary of the last and the lower first, which always have the Max. thickness and 0 respectively. Changing the upper (lower) boundary of the interval entails a change in the lower (upper) boundary of the adjacent interval. If you set the value of the upper (lower) boundary greater (less) than the value of the upper (lower) boundary of the adjacent interval, then this interval will be deleted.

Left-click on the row of the thickness table corresponding to the required interval, and then on the field in the Start or End columns in the selected row and change the value of the upper or lower boundary of the interval, or move the corresponding arrow on the ruler of the table of thicknesses.

Deleting interval in the table of thicknesses

In addition to the initial one, you can delete any interval.

Left-click on the row of the thickness table corresponding to the interval to be deleted, click Delete Interval button or move one boundary arrow of the interval in the table ruler until it connects with another boundary arrow of the same interval. The interval will be deleted and one of the adjacent intervals will be given a new width value. If the movement occurs from left to right, then the width of the left adjacent interval increases; the width of the right adjacent interval increases as you move the cursor from right to left.

Changing interval properties

Each interval has the following properties: measured thickness (Start and End columns), specified Width, Color and Layer.

During the vectorization process, these properties will be assigned to objects that have a thickness within the given interval.

Interval properties

When the Use table box is checked, the value specified in the Width field is assigned to all vector objects whose width is within the specified interval.

If you do not need to round off the line thickness within the interval to the values preset in the Width field, clear the checkbox in the corresponding line of the interval properties.

Specify the interval width

Select with the mouse the row of the thickness table corresponding to the desired interval, and then click in the Width column.

Enter the required value.

Assign layer to interval

You can assign name of any layer to the interval or select Current. All vector objects with thicknesses within the specified range will be created on the specified layer. If there is no such layer yet, it will be created automatically during vectorization. When you select the Current layer, objects will be created on it.

Select with the mouse the row of the thickness table corresponding to the desired interval, and then click in the Layer column.

Select the existing layer in the list or enter the name of a new layer that will be created during vectorization.

Assign color to interval

You can assign any color to the interval, including Current, By Layer, and By Block. All vector objects that have thicknesses within the specified interval inherit the specified color. If Current is selected, objects of the specified interval will be created with the current color.

Select with the mouse the row of the thickness table corresponding to the desired interval, and then click in the Color column.

Select the color from the list or specify it in the dialog box that opens when you click the Other option.

Saving settings of the table of thicknesses

Settings of the table of thicknesses can be saved in the settings file.

Click the Template button, select Save.

In the Save Template File dialog box:

• 1. set the file name with .TPL extension;
• 2. click OK.

By default, the file will be saved in the Recognition options folder located in the root folder.

Loading setting of the table of thicknesses

Click the Template button, select Load.

In the Open template File dialog box:

• 1. specify the file name;
• 2. click Open.

Trace Tab

To trace outlines, it is necessary to configure the parameters that affect the conversion procedure by this method.

In the R2V Conversion Options dialog box, open the Trace tab.

Trace tab

Trace tab options

Texts Tab

The order for automatic vectorization of texts:

• 1. configure vectorization parameters. In the R2V Conversion Options dialog box, in the Recognition tab, the Text area item should be checked, the required settings are made in the Texts tab;
• 2. select images to be converted;
• 3. run the automatic vectorization command Raster to vectors. Vector objects will appear above the source raster;
• 4. to interrupt vectorization process, press ESC.

The text search and recognition module parameters are configured in the Texts tab of the R2V Conversion Options dialog box.

Texts tab

Processing of raster texts is divided into two stages. First, the program searches for raster fragments containing raster texts. These fragments are called text areas.

Then the program applies to the found raster texts the operation specified as an additional parameter of the Text area algorithm in the Recognition tab.

One of such operations is raster text recognition using the built-in text recognition module (OCR).

The OCR module recognizes raster texts and creates text objects. It calculates the height and rotation angle of the created texts.

The program contains a file of OCR letter patterns (DEFAULT.OCR and CYRILLIC.OCR), with which the module recognizes English characters, digits, punctuation marks and special characters (the first half of the ASCII table). In addition, you can train the OCR module to recognize any other text characters.

If OCR cannot recognize a character, then that character is replaced in the text line with a ”~” (tilde) character. If all the characters in the word are not recognized, OCR does not generate a corresponding text object.

Text Recognition Options

Orientation

Specifies the acceptable orientation of raster texts:

• Horizontal searches for horizontal text lines. Text areas will be only horizontal;
• Horizontal and vertical searches for horizontal and vertical text lines. Text areas will only be horizontal and vertical;
• Arbitrary oriented searches for all text lines. Selecting this option can slow down the search for text areas.

Overlapped by graphics

When this option is enabled, the program searches for raster texts related to other raster objects. Selecting this option can slow down the search for text areas.

Standalone letters

Searches for single text characters. If this option is disabled, the program will not find single text characters, but it will also not recognize graphic objects, markers, dashes, etc. as text.

Patterns

To customize OCR, define a set of word patterns. A word pattern is a rule that defines the allowed sequence of characters within one recognized word. The OCR module will only recognize words that match one of the specified patterns (patterns are listed in the Texts tab of the R2V Conversion Options dialog box).

The Add Pattern and Delete Pattern buttons control the composition of the word pattern list. Below there is the formal description of the word pattern definition:

[% [length] character type] || [letter]]

Parameters for defining a word pattern

Character type encoding

For example:

• The Rz%D pattern corresponds to words that start with “Rz” followed by any sequence of digits, for example, “Rz40”, “Rz2.5”, “Rz5000”.
• The %1N%n pattern corresponds to words of the national alphabet with an uppercase first letter, for example “Hannover”, “Oslo”, “Moscow”.
• The %D%% pattern corresponds to words of the following type: “20%”, “1100%”, “12.50%”.
• The %DV pattern corresponds to words of the following type: “5V”, “220V”, “13.8V”.

Height Table

In this field you can set the possible text heights. If you check the box, when generating recognized texts, the OCR module will create text objects with heights from this list, rounding the recognized height to the nearest one specified in the list.

Template file

Specifies a sample letter library file used for recognition. Samples of letters are topological models of text symbols (letters, special characters, etc.), which are used to recognize raster text symbols.

It is possible to train the OCR module to recognize other text characters as well. During the learning process, OCR creates sample letters and writes them to the library. You can save sample letters both in an existing and in a newly created library file.

Note

If a custom library file with letter patterns is used, the OCR recognizes characters described only in that file.

Place to layer

In this list, you can enter the name of the layer on which the texts obtained as a result of OCR work will be located.

Train OCR

Ribbon: Raster – Conversion > Train OCR

Menu: Raster – Convert > Train OCR…

Command line: OCRTRAIN

This functionality is available only in the Raster module

The standard capacities of OCR module allow to recognize English letters, digits, punctuation marks and special characters (the first half of the ASCII table).

The OCR module can be trained to recognize any character. To do this, either replenish one of the existing libraries with new letter patterns, or create your own library.

The training process creates the examples (topological model) of text characters and place them in an open letter patterns library. Several patterns can correspond to one text character. Character patterns are stored in special letter pattern file libraries. These files are located in the OCR folder by default. You can replace any patterns in the existing library with your own.

Letter pattern library

Run the Train OCR command.

In the Train OCR dialog box:

To create a new pattern library, click New button or to add new patterns in the existing library, click Open button, select the required library file in the dialog box and click OK.

Create new, replace or delete patterns of existing symbols.

Save the library in its own or a new file using the Save button.

Click OK.

Train OCR dialog box

Creation, replacement and deletion of patterns in the library is performed using the buttons of the dialog box toolbar and the Character input field.

• tools to pick a raster character.

• Add and Delete buttons.

Creating a new letter pattern

Enter the required letter in the Character field.

Using one of the selection tools, pick the raster character corresponding to the specified letter:

• Select character by floodfill. To select, point with the cursor to an isolated raster character;
• Select character by window. To select, set the opposite corners of the window that bounds the desired raster character;
• Select character by polygon. To select, specify vertices of the polygon bounding the desired raster character; to complete the selection, press ENTER.

In case of mistake, repeat the selection procedure.

Click the Add character to library button.

The program will create a topological pattern of the letter and add it to the appropriate section of the patterns library.

The left part of the dialog box display the sections of the current pattern library: digits, English capital, English small, special, national capital, national small. Closed and non-empty sections are marked with ”+” sign.

Sections of the current pattern letter library

The box to the left of the section name, when checked, connects all patterns in this section for recognition. Clearing the checkbox removes the connection of all patterns in this section. Any section can be opened by clicking on the field with the ”+” sign, and enable or disable the necessary patterns located in this subsection. In order to connect/disconnect the patetrn, check/clear the checkbox to the left of the pattern name.

Multiple patterns can be specified for each letter. For example, as shown in the following illustration for the letter “a”. The pattern set corresponding to this letter can be expanded by clicking ”+”. The pattern can be selected by clicking on it with the mouse. When you select a pattern, its shape is displayed in the field on the right side of the window.

Creating a letter pattern

Patterns can be both disabled (in this case they will not be used for recognition), and removed.

Removing a letter pattern

• 1. select the required pattern.
• 2. click Remove character button.