Editing color images

Brief description: This lesson is devoted to methods of enhancing color and greyscale raster images - correcting image brightness, contrast, hue, and saturation; correction by histogram, and filtration. You will also learn about converting images to RGB, 8-bit indexed and greyscale; layering color and greyscale images to monochrome layers.

Correcting image brightness, contrast, hue, and saturation

You can tune brightness, contrast, hue and saturation of color and greyscale images.

Choose Brightness/Contrast from the Image menu.

Using Brightness/Contrast dialog, you can tune:

• Brightness of color and greyscale image
• Contrast of color and greyscale image
• Hue of color image
• Saturation of color image       

You can tune the parameter value using the sliders or entering the values in the corresponding fields.

Correction by histogram

Using correction by histogram you can tune the image brightness, hue and contrast. You can apply automatic correction by histogram or tune the parameters manually.

Choose Equalize from the Image menu.

To correct an image by histogram automatically, press the Auto button of the Equalize dialog.
To correct an image manually you need to specify three values: lightest and darkest levels, and image gamma - for each channel or for Master channel.

Darkest level (the left slider position) is set using the dark eyedropper , using which you can specify the colors to turn black All pixels included in the histogram area from 0 to the left slider, will turn black. Lightest level is set using the light eyedropper , using which you can specify the colors to turn white. All pixels included in the histogram area from the right slider to 256, will turn white.

Color Filters

Blur

This filter produces a blur effect on an image. For example, you can apply the Blur filter to an image which has texture fills.

Choose Blur from the Filters menu.

Set Radius for the filter. Increasing the radius value also increases image blurring. To modify the Radius value use the slider or enter the value in the field. When you achieve satisfactory results in the Preview window, press the Apply button.

Unsharp Mask

This filter is designed to increase the sharpness of color boundaries, producing an overall increase of image sharpness. For example, this filter can be used to correct images that became blurred after interpolation - e.g., after such operations as rotate, scale, alignment, calibration or four-point calibration, and also resize and resample.

Choose Unsharp Mask from the Filters menu.

Set Radius - "level" of the filter effect. Increasing the radius value increases the number of pixels surrounding the color boundary that are to be processed.

• Set Sensitivity for the filter using the slider or enter the value in the corresponding field. Increasing this value provides a stronger effect.
  Applying small Radius values results in an increase of sharpness only on color boundaries.
• When you achieve satisfactory results in the Preview window, press the Apply button.       

Median

This filter reduces image noise, analyzing all pixels within the specified radius and assigning the central pixel the averaged value of reviewed pixels. As a result, pixel color and brightness are evened. For example, Median filter can be used to eliminate image "grain".

Choose Median from the Filters menu.

Set Radius for the filter - area within which color values are to be to analyzed. When you achieve satisfactory results in the Preview window, press the Apply button.

Convert to RGB, greyscale, 8-bit indexed

Convert to RGB (greyscale)        

• If your document contains several raster images, select the one(s) to convert to RGB (greyscale).
• From the Image menu, choose Convert to, and then Convert to RGB (Convert to greyscale).

Convert to 8-bit indexed

You can convert full-color images (24 bit per pixel) and monochrome images to 8-bit-indexed (8 bits per pixel).
This operation can be used to reduce the size of a full-color file, and therefore speed up working with it. Note that in some cases the converted image might have lower quality due to some information loss.

• If your document contains several raster images, select the one(s) to convert to 8-bit indexed.
  See section Raster Image Selection for detailed information.
  
• From the Image menu, choose Convert to, and then Convert to 8-bit indexed.      

	Execution order
Calculating the optimal palette for the selected image(s)	Press Set auto palette .
Recalculating of palette for specified color number	Specify the color number in the Colors  field. Press the Reset Palette button.
Selecting colors from the palette	Press Shift, select colors from the palette by clicking on them (if only selecting one color there is no need to press Shift). NOTE: Select the Light Paint checkbox button to view selected colors in the Preview window.
Merging selected colors	Press Merge Colors  button.
Deleting selected colors	Press Delete Colors  button.
Changing selected color(s)	To change selected color(s): To the specified color- Press Get color to set and choose a desired color from the Color box. To one color of the image - press the button with eyedropper , and then specify a desired color on the screen.
Adding color(s) to the palette	Enter desired number of colors in the Colors field. Press Set auto palette - or - Click on a blank palette cell. Click on the added field, and assign the color as described above in section Changing selected color(s).

Separating color and greyscale images to monochrome layers

This section describes the operation of separating color or greyscale raster images to monochrome layers.
Binarization allows you to extract pixels of specified colors (or close to them) to a monochrome layer. For example, when you binarize a map, first you get a monochrome layer that contains isolines; then a layer that contains roads, etc.
Color Separation allows you to separate a color or greyscale image to several monochrome layers with a single operation; each image pixel is transferred to one of the resulting monochrome layers.

Binarization

Using binarization you need to specify the color range(s) for a created layer, e.g., for hydrography layer - from light-blue to dark-blue.

• Choose Binarization from the Image menu.
• Choose binarization method - Range or Threshold. Set the parameters for the chosen method (Color model (RGB, HSV, Grey), channel (grey, red, green, blue, hue)).
• Specify a layer and its color to place the new monochrome image to.
• Using one of these eyedroppers specify the colors on the screen; they are added to the list of the Colors list.
• The estimated binarization results are highlighted in the preview window.
  You can choose any incorrectly selected color from the colors list, and then remove it using the Delete button.
• Tune the results using the sliders located in Ranges (Threshold).
• You can delete small speckles that appear using the slider Noise Reduction.
• When you achieve satisfactory results in the Preview window, press Apply.       

Color separation

To execute Color separation set color categories (thematic layers), by which you want to separate the image, and then set a list of colors for each category.

• Choose Color separation from the Image menu.
• Specify the first category color on the image using the eyedropper Add category Choose a standard color; specify a name for the category in the displayed Category name and color dialog.
• Using the eyedropper Add color, specify all colors of the image, which should fall within this category.
• Repeat the previous two steps for all categories, which you want to set on a new layer.
  Incorrectly defined categories and colors can be selected and deleted from the list using the Delete button.
• Using the Noise Reduction slider you can adjust the sensitivity of Color separation - smooth contours.
• Deleting small-sized speckles is executed when the Despeckling checkbox is set to on.
• When you achieve satisfactory results in the Preview window, press Apply