基于综合肤色检测和二值形态学的人脸定位研究

Face Location and Detection Algorithm Based on Skin Color and Binary Morphology

针对人脸定位检测中存在的速度慢、精度低及噪声干扰问题,提出了一种基于综合肤色检测和二值形态学的人脸定位检测算法.该算法将YCrCb(明亮度-色调-饱和度)模型与HSV(色相-饱和度-色调)模型用于人脸综合肤色检测,在YCrCb与HSV空间中根据待检测图像每点的颜色值进行人体区域或背景区域的判断;然后,将检测图像转换为二值图像,对图像进行形态学处理;最后,选用人脸几何特征对筛选后连通区域进行判别,实现人脸的准确定位和检测.实验结果表明,该算法对于简单、中等、复杂三种情况下人脸图像的检测正确率分别达到了99%、92%和85%.另外,由于二值形态学消噪算法的使用不仅提高了人脸检测的准确率,而且加快了检测速度.

The face location and detection algorithm based on skin color and binary morphology was proposed in this paper .The HSV （Hue-Saturation-Value） model and YCrCb （Brightness-Value-Saturation） model are applied to the face skin detection in this detection algorithm ,and these models can detect and judge every pixel of image which belongs to human body region or background region according to the color value .Simultaneously ,the image can be converted into the binary image by the threshold method ,and be processed by the binary morphology algorithm . Finally ,the connected region which is screened can be decided by the geometric features of human face ,and the face can be located and detected accurately .Hence ,the face location and detection algorithm based on skin color and binary morphology consists of three steps ,which are the skin color detection ,the morphological operation and the geometric features of distinguish .The experimental results show the algorithm can improve the accuracy of face detection and speed up the recognition speed of face location and detection .