A novel face segmentation algorithm
Wei Fang, Pengfei Li · 2002
This paper addresses a novel method to automatically segment out the face from a head-and-shoulders image with a complex background scene. The method involves a fast, reliable, and effective algorithm, which exploits the distribution characteristics of human skin color to perform color analysis. Then based on the color-analysis result, the algorithm employs a novel set of regularization processes including density regulation, luminance regulation and shape extraction, to reinforce pixels that are more likely to belong to the facial regions and eliminate those that are not. The performance of the face segmentation algorithm is illustrated by some simulation results carried out on a head-and-shoulders test image.