Back to Aurora Vision Studio website

You are here: Start » Filter Reference » Computer Vision » Image Segmentation » SegmentImage_Color

SegmentImage_Color

Advanced

Module: FoundationPro

Segments an image basing on distance to model colors.

Applications

Detection of objects of undefined shape, but characterized by uniform color and good contrast to the background.
Name Type Range Description
Input value inImage Image Input image
Input value inRoi Region* Range of pixels to be processed
Input value inReferenceColors PixelArray Colors to compare pixels to
Input value inMaxDifference Real 0.0 - Maximal difference between pixel and reference color to be accepted
Input value inDifferenceMultipliers RealArray* Scales for maximum differences for each color
Input value inChromaAmount Real 0.0 - 1.0 Proportion of chromatic information in distance computation
Input value inForceDisjointRegions Bool Force output regions to be disjoint
Output value outRegions RegionArray Regions of pixels closest to colors

Description

The operation finds regions of similar colors in RGB color space within inImage.

First, for each pixel of the input image its distance to each color in inReferenceColors is computed (see ColorDistanceImage function).

Then, for each reference color c, the function computes region of pixels that are closer to c than its corresponding threshold. In case inDifferenceMultipliers is Nil, threshold is equal to inMaxDifference. If inDifferenceMultipliers is not Nil, corresponding threshold is the product of inMaxDifference and the element of inDifferenceMultipliers that corresponds to c.

If flag inForceDisjointRegions is set to True, the regions are forced to be disjoint. In case a pixel has distance less than threshold for more than one color, distances are divided by thresholds and the minimum of ratios is taken. In particular, if the function works in one-threshold mode, the operation described above becomes simply minimum of distances.

Examples

SegmentImage_Color run with colors taken from three points with inChromaAmount = 0.7 and inMaxDifference = 15

Hardware Acceleration

This operation supports automatic parallelization for multicore and multiprocessor systems.

Errors

This filter can throw an exception to report error. Read how to deal with errors in Error Handling.

List of possible exceptions:

Error type Description
DomainError Inconsistent array sizes in SegmentImage_Color.
DomainError Reference color array size too big in SegmentImage_Color.

Complexity Level

This filter is available on Advanced Complexity Level.

Filter Group

This filter is member of SegmentImage filter group.

See Also

  • ColorDistance – Compares two pixels using chromatic and non-chromatic information. Assumes RGB color space.
  • ColorDistanceImage – Compares each pixel with the specified color using chromatic and non-chromatic information.