Color Clustering
Color clustering determines the dominant colors in an image or video frame by clustering similar colors. The task returns the color at the center of each cluster, and the proportion of the pixels that belong to each cluster. If you configure a color dictionary, the task clusters colors around the colors that are defined in the dictionary, and returns a name (such as "light blue" or "red") for each cluster.
Configuration Parameter | Description |
---|---|
ColorDictionary | A dictionary file that associates names with RGB color values. |
ColorSpace | The color space to use when providing analysis results. |
ColorThreshold | Any color cluster with a score that does not meet the specified threshold is omitted from the results. |
CombineNamedColors | A Boolean value that specifies whether to combine clusters that have the same color name. |
Input | The image track to process. |
MaxInputQueueLength | Can be used to place a limit on latency. |
NumParallel | The maximum number of video frames to analyze simultaneously. |
Region | Specifies a region of the frame to process, instead of the entire frame. |
SampleInterval | The interval at which frames are selected to be analyzed. |
SegmentDuration | The maximum duration of a record in the SegmentedResult track. |
Type | The analysis engine to use. Set this parameter to ColorCluster . |
Output Tracks
The following table describes the tracks that are generated by this engine. The Output column indicates whether the information contained in the track is included by default in the output created by an output task (when you don't set the Input
parameter for the output task).
Output track | Description | Output |
---|---|---|
Data
|
Contains one record, describing the analysis results, for each input record. | No |
Result
|
Contains a record, describing the analysis results, for each event.
|
Yes |
SegmentedResult
|
(Available only when processing video). The same as the |
Yes |
Start
|
The same as the |
No |
End
|
The same as the |
No |
ClusteredImage
|
Contains one record for each event. Each record contains the source image, cropped to the analyzed region and modified so that it only contains colors that match the center of a color cluster. If the analyzed region is not rectangular any pixels outside the region are transparent (or black if you use an image format that does not support transparency). | No |
For more information see Color Analysis Results or use the action GetExampleRecord.