摘要
通过对复杂背景下的人脸图片进行自适应光照补偿,建立高斯肤色模型,并利用改进的Otsu算法进行人脸肤色检测.实验证明,该方法效率高,对背景、尺寸、旋转角度等细节具有很好的适应性,在人脸实时检测中具有很好的应用价值.
The detection and location of skin is a very important part of face recognition. First, the method compensates for the illumination automatically on the image and then it establishes a face Gauss model and makes use of the improved Ostu algorithm to carry out skin checking, the aim of which is to search for a quick, efficient and steady faces detection method. The experimental results show that the proposed approach is an efficient and robust method under different backgrounds, sizes and rotation angles.
出处
《湖州师范学院学报》
2007年第1期50-52,共3页
Journal of Huzhou University
关键词
人脸检测
肤色高斯模型
光照补偿
Otsu阀值分割
face detection
skin Gauss Model
illumination compensation
Otsu threshold selection
作者简介
徐静云（1980-），男，江西广丰人，湖州师范学院信息工程学院助教，研究方向：图像处理与模式识别．
