【摘要】 提出一种基于肤色信息的正面人脸检测方法。设计了基于颜色的皮肤分类器,对图像中的皮肤区域和非皮肤区域进行分割;并对分割出的皮肤区域进行预处理,初步确定候选人脸;再进一步将候选脸与模板脸进行匹配最终确定是否为人脸,并给出检测到的人脸的位置信息。对各类型图像进行大量实验表明,本方法具有较好的准确性和实用性。更多还原

【Abstract】 In this paper developed and implemented the method for detecting frontal human faces color-based information in images where they appear. Designed a color-based skin classifier to segment human skin regions from non-skin regions. Skin regions are segmented to determine which regions can possibly be determining a frontal human face. Farther on use of template matching to determine if a given skin region represents a frontal human face or not, if it exist while given the coordinates of the face. Experiments with variety of images show that the method proposed here is efficient and useful.更多还原