广东工业大学学报 ›› 2008, Vol. 25 ›› Issue (2): 69-73.

• 综合研究 • 上一篇    下一篇

基于动态椭圆滤波器的人脸定位

  

  1. 广东工业大学信息学院数字图像技术研究所; 广东工业大学信息学院数字图像技术研究所 广东广州510006; 广东广州510006;
  • 出版日期:2008-06-01 发布日期:2008-06-01

Face Location Based on Dynamic Ellipse Filters

  1. Institute of Digital Image Technology,Faculty of Information Engineering,Guangdong University of Technology,Guangzhou 510006,China
  • Online:2008-06-01 Published:2008-06-01

摘要: 针对传统肤色分割算法无法将人脸从类似肤色的背景分割出来这一个问题,提出一种基于动态椭圆滤波器的人脸定位算法.该算法利用积分投影法预测人脸可能的范围和位置,并结合椭圆的紧密度概念调整滤波器参数,构造最优滤波器,把肤色分割后的连通区域中非人脸的类肤色区域过滤掉,有效地减少类肤色区域带来的干扰,再结合模板匹配方法对人脸进行定位.

关键词: 人脸定位; 积分投影; 椭圆滤波; 模板匹配;

Abstract: Aiming at the problem that traditional skin-segment algorithm could not efficiently divide a face from its background whose color was similar to that of the face,a reliable face location algorithm based on dynamic ellipse filters,was presented.A possible position and the size of a face were predicted through integration projection.In combination with Density of the ellipse,the filter would be adjusted to make itself best so that it could remove the region that was connected to the face effective...

Key words: face location; integration projection; ellipse filter; matching;

[1] 蒋旻,刘华咏,贺贵明.  基于模型的人脸图像配准[J]. 计算机应用研究. 2007(02)

[2] 赵丽红,刘纪红,孙宇舸,徐心和.  基于外观的人脸检测方法[J]. 计算机应用研究. 2006(02)

[3] 刘文卿,朱俊青,王林泉.  基于积分投影的快速人脸定位[J]. 计算机辅助工程. 2003(02) 

[4] Henry Chang,Ulises Robles.Face Detection. http://www-cs-students.stanford.edu/~robles/ee368/main.html . 2000

[5] Yu B,Yuan B Z.A more Efficient Branch and Bound Algorithm for Feature Selection. Pattern Recognition . 1993
No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!