基于肤色分割及特征定位的人脸检测算法研究

针对复杂彩色图像提出一种快速有效的人脸检测算法。首先通过一种新的色彩均衡技术消除由光照引起的色彩偏移,然后采用HSI空间与归一化RGB空间相结合的混合肤色模型,提取出原图像中的类肤色区域,最后根据人脸的几何特征定位眼睛、嘴巴和人脸区域。实验结果表明,该算法能较为准确地定位彩色图像中的正面或小角度偏转的人脸,具有较快的检测速度。

Face Detection Based on Skin Segmentation and Features Location

A face detection algorithm for the color image with complex background is presented,which is effective and feasible.Firstly,a new technique of color balance is used to remove the color bias produced by light condition.Secondly,a mixed model in both HSI and normalized RGB color space is constructed,which can be used to acquire possible skin regions in the original image.After skin color segmentation,the geometrical shape information of face is used to locate eyes mouth the face region.Experimental results show that the proposed approach can detect positive or rotated face with little angle in a given image,and at the same time,a faster speed of the face detection is achieved.