摘要
提出一种复杂背景下灰度图像的人眼定位算法.首先对基于均衡处理后的图像估计出初始阈值;其次采用阈值递增法逐次二值化图像,并对二值图像中的黑块逐一标记;然后用人眼位置判定准则确定可能的人眼区域,用相似度准则从中标记出两眼黑块;最后,针对该算法进行了仿真研究.
This paper presents an improved method of locating human eyes in gray image with complex background. Firstly, the initial threshold is obtained from the balanced-processing image. Secondly, the black blocks in the binary image derived from the balanced-processing image are marked one by one with the method of the progressive increase of the threshold. Thirdly, the possible eye＇s areas are determined according to the rule of eye＇s location. And the two marked black blocks represent the eye＇s areas by making use of similarity, Finally, simulations are performed on the algorithm.
出处
《河南大学学报（自然科学版）》
CAS
北大核心
2006年第4期83-86,共4页
Journal of Henan University:Natural Science
基金
河南省科委自然科学基金项目(0523020600)
河南省高校杰出科研人才创新工程项目(2005KYCX012)
关键词
人眼定位
直方图
相似度
eye location
histogram
similarity measure
