ColorClassifier

Module used to classify one or several objects of a particular color in a scene. The input is a color image in rgI format which can be obained from the ColorTransform module. The output is a table of target coordinates.

Example XML definition

A simple example

  <module
      class = "ColorClassifier"
      name = "ColorClassifier"
      color = "145 degrees"
      width = "30 degrees"
      saturation_min = "0.05"
      saturation_max = "0.35"
      compensation = "yes"
  />

Parameters

Name	Description	Type	Default
class	The class name for the module; must be "ColorClassifier".	string
name	The name of this instance of the module.	string
compensation	Compensate for the illumination using a gray world assumption	bool	no
diagnostics	Draw pixel locations in the diagnostic output COLORSPACE. This is slower but useful during set-up.	bool	yes
color	The target hue (in degrees). red=135, green=315; blue = 225	float	225
width	The target width (in degrees).	float	20
saturation_min	The minimum saturation of the target color.	float	0.05
saturation_min	The maximum saturation of the target color.	float	0.50

Module Connections

Inputs

Name	Description
R	The normalized red channel
G	The normalized green channel
I	The intensity channel

Outputs

Name	Description
OUTPUT	Image with the detected points.
COLORSPACE_R	Image with all pixels plotted in the rg color space. Red channel.
COLORSPACE_G	Image with all pixels plotted in the rg color space. Green channel.
COLORSPACE_B	Image with all pixels plotted in the rg color space. Blue channel.

Author

Christian Balkenius
christian.balkenius@lucs.lu.se
Lund University Cognitive Science

Files

ColorClassifier.h
ColorClassifier.cc
ColorClassifier.ikc