This paper presents a novel hand segmentation method in complex environment based on color model and boundary cutting. Probability map is generated through skin color model to get a rough segmentation. For boundary cutting, a vote map based method is proposed to get precise hand segmentation. Experimental result is appraised by sensitivity and specificity which are 96.54% and 96.92% respectively.
Published in:
Machine Learning and Cybernetics (ICMLC), 2012 International Conference on
(Volume:3
)
Date of Conference: 15-17 July 2012