中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

重庆交通大学学报(自然科学版) ›› 2010, Vol. 29 ›› Issue (3): 488-492.

• • 上一篇    下一篇

基于改进YCrCb颜色空间的肤色分割

曹建秋,王华清,蓝章礼   

  1. 重庆交通大学信息科学与工程学院, 重庆 400074
  • 收稿日期:2009-12-08 出版日期:2010-06-15 发布日期:2015-01-22
  • 作者简介:曹建秋(1967—) , 男, 湖南益阳人, 副教授, 硕士, 主要研究方向数字图像处理。E-mail:caojq86@cquc.edu.cn。
  • 基金资助:
    重庆市自然科学基金项目( CSTC, 2006BB2409, 2007BB2438)

Skin Color Division Base on Modified YCrCb Color Space

CAO Jian-qiu,WANG Hua-qing,LAN Zhang-li   

  1. School of Information Science & Engineering,Chongqing Jiaotong University,Chongqing 400074,China
  • Received:2009-12-08 Online:2010-06-15 Published:2015-01-22

摘要: 人脸识别是计算机视觉、模式识别、生物特征识别、图像处理等多学科交叉的一门科学,肤色分割是人脸识别 研究中一个重要的步骤。针对基于YCrCb颜色空间的肤色分割在强光下分割不准确的现状,提出了一种基于改进 的YCrCb空间的肤色分割方法,本文利用改进YCrCb颜色空间提高了强光下肤色分割的准确性和适应性。实验证 明该算法减弱了光线强弱对人脸肤色分割的影响,更加准确的找到人脸区域,误检率较低。

关键词: YCrCb, 颜色空间, 肤色分割

Abstract: Face recognition is a crossed subject of computer vision, pattern identification, biology character identification, and digital image processing. Skin color division is one of the most important steps of face recognition. Aiming at the current condition that the skin color division based on YCrCb color space can't distinguish the face accurately in the strong light, a modified division method of YCrCb color space is put forward to improve the adaptability and accuracy of YCrCb in strong light. The experiment verifies the conclusion that this new method can reduce the influence of light luminance on the human face segmentation operation, therefore it will be more accurate to find the face region and the error ratio is low.

Key words: YCrCb, color space, skin color division

中图分类号: