I am using a code that is applying a TopHat morphology operation and then it is converting the image to gray scale. I know what does TopHat on gray images, but what approach is implemented if the image is RGB (or BGR)?
i know this is not an answer to your question but i want to point out morphology operation on color image is promising on some detection cases.
when i was thinking about a solution to this question i met your question and it guided to me to promising solution.
the question : how to detect pimples
result image by morphology operation :
it is easier to detect red crosses.
What have you done there? TopHat on each channel or what? My question was: Does topHat applies on each channel or on float values, or how? I understand the idea of doing it on binary images, but color ones are not binary...
the result image from Morphology_1.cpp you can test TopHat Morphology_2.cpp
Asked: 2014-10-21 06:50:22 -0600
Seen: 786 times
Last updated: Oct 05 '15
http://answers.opencv.org/question/66125 maybe related.