A Parametric Discriminative Approach for Skin color Detection by Training Weak Learners on Normalized Chrominance and Luminance

Abstract

This paper presents a novel approach for the detection of skin color in image or video, captured through ordinary web camera. The TSL color space is used due to its specialty in distinguishing among the skin and Non-skin color. To label the skin colors, a classifier based on adaboost algorithm has been trained. To validate the performance of the classifier, a database of skin colors was developed using different color tones ranging from fair to deep. General TermsSkin color detection, TSL color space, Adaboost color classifier, Skin color tones mixing.