Body parts detection in gesture recognition using color information

Abstract

Skin detection plays an important role in a wide range of image processing applications ranging from Hand and face detection, tracking in gesture analysis. This work deals with skin color identification algorithm using color segmentation to detect human hands and face in color images. For color segmentation both Single Gaussian Model (SGM) and Gaussian Mixture Model (GMM) are used on different images. Experimental results on images presenting variations in lighting condition and background, and variation in age of the person, demonstrate the efficiency of described skin-segmentation algorithm. This work evolves and compares four GMM's with respect to the SGM.