A Novel Algorithm for Skin-tone Color Filtering in Color Images

Human face detection is the first step for a fully automated face recognition system. It is also crucial to video surveillance systems, human computer interface, and image/video retrieval systems. Skin color is an important cue for face detection. In this paper, we propose a novel skin-tone color filtering algorithm for detecting faces in color images with complex illumination and background. After a novel adaptive lighting compensation applied, a skin color filter is used to detect skin regions. Our algorithm is shown to be more robust and efficient by experiments when compared to existing skin-tone color filtering algorithms.