基于肤色平滑度的人脸分割  

Face Segmentation Based on the Smoothness of Skin Color

作　　者：陈令[1] 汪亚明[1] 

机构地区：[1]浙江工程学院计算机视觉与模式识别研究中心,浙江杭州310033

出　　处：《测试技术学报》2004年第1期60-64,共5页Journal of Test and Measurement Technology

摘　　要：　人脸检测具有非常重要的现实意义.在进行人脸检测时,人们通常将注意力集中在眼、鼻、口等具有显著特征的核心区域,而忽略人脸面部、额头肤色是比较的平滑的这一特征.针对这点,本文提出肤色平滑度的概念,并利用这一概念和HSV颜色模型对图像进行分割.实验表明,本文的基于肤色平滑度的分割方法,能方便快速有效地分割出处于各种旋转角度、各种光照条件、各种颜色背景下的人脸.Face Detection is of great practical significance. In the process of face detection, people tend to concentrate on the core area with outstanding characteristics such as the eyes, the nose and the mouth, but ignore the fact that the color of the cheek and the forehead is fairly smooth in comparison. This paper brings forward the concept of the Smoothing of Skin Color. By this concept and together with HSV color model, we can successfully segment the face image. The experimentation shows the method that this paper adopted could conveniently, rapidly and effectively segment the faces from any rotation angle, in any light conditions and under any color background.

关 键 词：人脸图像分割 HSV颜色模型 肤色平滑度 肤色模型 图像检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]