An efficient face detection method in color images

Abstract

In this paper, an efficient algorithm for detecting human faces in color images is proposed. The first step of our algorithm is to segment the possible skin-like regions in an image by using color information, and then the gray level clustering is applied to eye location in skin color regions. The possible face regions are formed by pairing two possible eye candidates in a skin color region. In order to improve the reliability and accuracy, both Symmetry verification and template matching methods are used to locate the faces in those regions. Experimental results show that this algorithm can detect single or multiple faces from images having simple or complex backgrounds. © Springer-Verlag Berlin Heidelberg 2005.