Color Correct Node

The Color Correct node is a comprehensive tool for performing a wide variety of adjustments. It provides seven common color manipulation tools in one layer-based interface for fast, simple color correction.

The Color Correct node enables you to do any or all of the following:

  • adjust brightness, contrast, and gamma
  • adjust shadows, midtones, and highlights
  • perform HSL-based color adjustments
  • operate on individual channels
  • swap channels
  • invert colors
  • desaturate colors
  • simulate printer lights and stops

In the Color Correct node, adjustments are made directly in the work monitor, using the tools provided. For each type of adjustment you wish to make, you create a separate layer. Each layer features its own set of parameters, which you adjust to achieve the desired effect. You can add as many layers as you need, and any layer can be deleted, reordered, or temporarily disabled.

Speed and Accuracy

Using the Color Correct node to perform multiple color correction operations on an image is both simpler and more accurate than using a separate node for each adjustment.

The result is faster and more accurate because each successive operation is cumulative. Regardless of the bit depth of the input image, the node calculates each operation in floating point. All of the data in the result of the first operation is preserved and used in the next operation--even those values that go out of bounds and would have to be clamped on final output.

Note:
Three operations, by their nature, are exceptions to this rule: HSL, Channel Swap, and Monochrome. Whenever you create a layer of one of these three types you break the chain and potentially lose some color data.

Color Correct Parameter Tab

This tab gives you access to the Color Correct work monitor. The parameters available for each type of adjustment, which are accessed from within the work monitor, are described in the following section, "Using the Color Correct Work Monitor ."

Work Monitor / At Resolution

First, select a display resolution for the work monitor from the At Resolution popup menu. Then click the plus icon to display the work monitor for this node.

Channel Mask

The Channel Mask parameter enables you to inhibit the modification of any channel by deselecting the Red, Green, Blue, Alpha, or Other channel buttons.

Control Image Parameters

The Control Action and Control Channel menus enable you to specify how an optional control image input will govern the node operation, as explained in "Using Control Images with Filter Nodes ."

Using the Color Correct Work Monitor

The Color Correct work monitor provides a unique panel where you create a layer with specific parameters for each type of operation you wish to perform.

Color Correct Monitor Panel

Working with Layers

Click any layer button to create a correction panel for that operation. A corresponding entry will be created in the layer list.

Each entry in the layer list consists of four icons:

  • The Move Layer icon (up arrow) is used to move a layer up one position in the list. Each layer is calculated in order from the top of the list to the bottom. (Refer to "Speed and Accuracy " in the introduction of this node description for more information about why the order of layers is important.)
  • The Select Layer icon is used to select a layer and display its associated parameters in the layer panel.
  • The Delete Layer icon (red "X") is used to delete a layer.
  • The Disable Layer icon (checkbox) is used to temporarily disable a layer. When the box is checked, the layer is enabled and will contribute to the result; when the box is not checked, the layer is disabled and will remain so until you check the box again.

Selecting a Display Depth

The D and F buttons at the lower right corner of the layer panel enable you to control how values are displayed in the Color Picker and in the graph displays available in some layers: select "D" to use the Default bit depth in the graph displays; select "F" to use Floating Point (0-1) in the graph displays. This option enables you to display input values in whatever unit is most appropriate to the task.

For example, if you are working in 8-bit, choosing Floating Point display of input values would enable you to reduce the brightness of the image to 80 percent (0.8 on the graph) without calculating which value between 0 and 255 is equivalent. On the other hand, if you are working in Cineon linear and you know that your white point is 4095, you would probably prefer to use Default input display (0-65535).

Using the Color Picker

The Color Picker can display the RGBA value of any pixel, or pixel area, in an image. It is especially useful for comparing color values in two different images using A/B Wipe.

To use the Color Picker, simply click the mouse in one of the two Color Meter boxes. The cursor will change into an eye dropper--this is the Color Picker tool.

Then click on any pixel in the image that you want to meter. The color values of that pixel will appear in the Color Meter box and the cursor will revert to normal shape. (If you are sampling an area of pixels, their average value will be displayed.)

To sample an area of pixels, click the Sample Size Selector to select the appropriate size pixel grid from the popup menu: 1x1, 3x3, 5x5, 7x7, 9x9, or 11x11.

BCG ( Brightness, Contrast, Gamma) Layer

The BCG panel enables you to adjust Brightness, Contrast, and Gamma values using tuners. The BCG panel features five tabs: Master, Shadows, Midtones, Highlights, and Range.

BCG Master, Shadows, Midtones, and Highlights Tabs

The first four tabs provide the same set of parameters, which enable you to specify Brightness, Brightness Offset, Contrast, Contrast Offset, and Gamma for each channel (RGBAO) and for Luminance.

You can adjust the entire image in the Master tab, and/or adjust the Highlights, Midtones, and Shadows by selecting the appropriate tab, all of which feature the same set of parameters.

The default value for all of the Brightness, Contrast, and Gamma parameters is 1, which represents no change from the input image. Offsets for Brightness and Contrast can be reset for each channel:

  • Click the "Brightness" button to toggle to "Brightness Offset" (the default brightness offset is 0).
  • Click the "Contrast" button to toggle to "Contrast Offset" (the default contrast offset is 0.5).

BCG Range Tab

The fifth BCG tab, Range, enables you to control the effect of any adjustments you make in the Shadows, Midtones, and Highlights tabs. You can specify the range of values that will be affected, as well as the magnitude of the effect.

The following illustration demonstrates how the Range tab can be used. If you made this adjustment to the Midtones curve in the Range tab, it would affect any changes in parameter values that you made in the Midtones tab. A smaller range of colors would be affected, and the effect would be diminished by half.

You can adjust the graph by manipulating the graph display, or by keying specific values into the Input and Output tuners:

  • The Input value represents a point on the colorspace.
  • The Output value represents the level of effect at that point (in a range of 0-1).

You can turn the display of shadows, midtones, or highlights on and off in the graph by clicking the S, M, and H buttons on the left side of the Range panel. To bring up a Keyframe Editor for each, click the associated button.

The BCG Range tab interface works exactly like the interface provided in the Curve Editor layer (see the description of the Curve Editor Layer for more information). You can add and delete points and apply different functions to the distribution curves.

The only difference is that the BCG Range panel displays a histogram of the current image in the graph. You can adjust the display of the histogram, if necessary, to make it easier to see, by dragging the vertical sliders on the right side of the graph.

HSL ( Hue, Saturation, Lightness) Layer

The HSL layer enables you to perform hue-based color adjustments using the HSL color model.

The Hue Shift parameter enables you to rotate the color distribution vectors of the image around the axis of the colorspace. You can perform the hue shift visually by dragging the color wheel or you can use the tuner to enter a value in the range of -180 to 180 degrees. A value of 0 represents the input data.

The Saturation parameter enables you to adjust the chrominance level by keying a value in the range of -1 to 1 into the data entry field. A value of 0 represents the input data value, -1 represents complete desaturation, and 1 creates maximum saturation.

The Lightness parameter enables you to adjust relative intensity, or brightness, by keying a value in the range of -1 to 1 into the data entry field. A value of 0 represents the input data value (i.e., no change), -1 will result in a completely black image, and 1 will result in a completely white image.

Curve Editor Layer

The Curve Editor enables you to modify the color attributes of an image by editing curves that represent color channels in a graph display.

The horizontal axis of the graph represents the input values, and the vertical axis represents the output values to which they are mapped. You can hide or display any channel, and you can modify channels separately or together. You can bring up a Keyframe Editor for each channel to animate values over time.

Display Depth

When working with 16-bit (or float) imagery, you can choose to display the data in the graph in Default view (0-65535), Cineon view (0-4095), 14-bit (0-16383), 12-bit (0-4095), or 10-bit (0-1023) by clicking the appropriate Display Depth button: D, C, 14, 12, or 10.

When working in 8-bit space, only the Default view (0-255) is available.

Editing Curves

To work with one channel at a time, select that channel by clicking its icon, and then deselect any other selected channels.

To select a point, click on it. To select overlapping points on different curves, hold down the Alt key and drag a bounding box around the points in the graph.

To add a point, simply click on the curve and a point will be created. To delete a point, click on the point to select it and then press the Delete key on the keyboard.

Tip:
The Plus icon (+) enables you to bring up an identical Curve Editor panel in a resizable window. Increasing the size of the graph can make it easier to select and manipulate points and curves.

Input and Output Parameters

The Input and Output parameters display the current values to which the selected point is mapped. As an alternative to dragging a point in the graph to reposition it, you can key specific values into the data entry fields to remap values.

Slope In and Out Parameters

The Slope In and Slope Out parameters display the current value of the slope of the curve at the selected point. If the Equal Slopes or Different Slopes function has been selected from the Slopes menu, the fields become active and you can key specific values into the data entry fields to define the slopes.

Function Menu

The Function menu enables you to select Linear, Continuous, Equal Slopes, or Different Slopes options for the selected point. The menu displays the type of curve used for the currently selected point. If no point is selected, the menu displays a row of dashes.

Monochrome Layer

The Monochrome layer desaturates an image to the degree you specify. This value, which is calculated using luminance, can be animated over time.

The monochrome tuner range is 0-1, with 0 representing no change in the image and 1 representing a totally monochromatic (grayscale) image.

You can use the Luminance popup menu to specify whether Film, NTSC, or PAL luminance should be used in the calculation. For more information about the formulas used to compute luminance, refer to the Channel Swap Node description in chapter 13, "Tool Nodes ."

Channel Swap Layer

This layer enables you to swap channel values or to fill any channel with black, white, or luminance.

The left-hand column represents the channel data available for use, and the right-hand column represents the output channels. The connecting lines drawn between the columns indicate what channel data is being used for each output channel.

To replace the existing data in a channel, simply click on a channel in one column, and then click on a channel in the other column. A line is drawn automatically to connect them. For example, you could click on Black in the left column and then on the alpha channel in the right column to replace the existing alpha channel with an entirely transparent alpha.

If you choose to fill a channel with luminance, use the Luminance popup menu to specify whether Film, NTSC, or PAL luminance should be used in the calculation.

Contrast Stretch Layer

The Contrast Stretch layer enables you to increase or decrease the contrast of an image or image channel. You start by specifying a low value and a high value, which define the range of colors in the input image that will be remapped to the output colorspace.

By default, the selected range is remapped to the entire colorspace. For example, working in 8 bits per channel, the selected range will be remapped to 0-255 and any values in the input that fall outside of the selected range will be clipped.

However, you can specify a new low and high value in the output colorspace to use for the remapping. This will remap the selected input range to the new low and high, and any input values falling outside the range will not be clipped but will gradually fall off in a linear distribution to the minimum or maximum value of the output colorspace.

By specifying a smaller range for the output than the input, you can effectively decrease the contrast of the image.

Adjusting the Input Range

To specify the range of the input to use, be sure that the input column is selected by clicking the RGB cube in the leftmost column--it should be selected by default. (To adjust a channel or channels individually, select the Red, Green, Blue, Alpha, or Other cube in the leftmost column.)

With the input column selected, the low and high values can be adjusted numerically or visually, by referencing the histogram data or clicking in a portion of the image.

To specify the low and high values numerically, key specific values into the data entry fields, or drag the associated tuner dial to change the values. When referencing the frequency distribution data, you can drag the vertical bars (dashed lines) in the histogram to redefine the range. The values in the data entry fields will update accordingly.

The input low (black) and high (white) values can also be selected visually, by clicking a "pick from" button. The cursor will change to an eye dropper tool, which you can use to select a white or black pixel in the monitor image. You can select the average of an area of pixels by adjusting the Sample Size Selector (refer to "Using the Color Picker " if you need more information).

Tip:
If you are not satisfied with the result of a "Pick From" operation, you can disable the layer and the Pick buttons will remain active. This enables you to see the "before" image in the monitor again and use it pick a new black or white point. When you enable the layer again, Chalice will recalculate using the new pick value.

Adjusting the Output Range

The same methods described above for the input range can also be used to adjust the output range. First, you must select the output column by clicking the RGB (or individual channel) cube in the column next to the input column. Then specify the range.

Note:
You can use an A/B wipe to access another image and then use this reference image to pick pixels that represent the black and white point for the output range. For the output range, be sure use the "pick to" buttons instead of the "pick from" buttons.

Invert Values Layer

The Invert Values layer inverts the color values of any or all channels in an image, effectively creating a negative of the image, by subtracting the current pixel value by the maximum pixel value.

The panel enables you to specify which channels should be inverted. By default, all available channels are inverted. To deselect a channel, click on its icon in the panel.

Stops Layer

The Stops layer enables you to control the color values at any channel in an image using photographic stops.

Increasing a channel by one stop will double the color values at each pixel. This effect is cumulative; bringing a channel up two stops will increase the amount of light by a factor of four. Similarly, bringing a channel down one stop will reduce the color values by a factor of two.

Printer Lights Layer

The Printer Lights Layer enables you to control the color values at any channel in an image using printer lights.

The Lights per Stop parameter specifies how many Printer Lights must be added to a channel to double the color values. The default values of 25 represent no change to the image. At eight lights per stop, an addition of eight printer lights will double the color values. At four lights per stop, an increment of eight printer lights will increase the color values fourfold.




Table of Contents | Index



Copyright 1996-1999 by Silicon Grail Corporation