If you accept suggestions, I would recommend you to use an RGB-D camera (like Kinect) it will make your life a lot easier on hand segmentation.
You can take a look at this paper:
"Both-hand Gesture Recognition Based on KOMSM with Volume Subspaces for Robot Teleoperation". Peris et al. IEEE-CYBER 2012.
What is done there basically is to use OpenNI library to track the user's skeleton and use the location of the hand joints to accurately segment the hands and input them to a classifier for gesture recognition.
Good luck!
Asked: 2012-08-07 09:19:31 -0600
Seen: 1,388 times
Last updated: Aug 09 '12
Please post some images of examples.
Also, can you please be a little bit more specific? If I understood, you have a mask for the whole hand which you want to segment into palm and non-palm regions, is that it?