###
工程科学与技术:2014,46(2):66-71
←前一篇   |   后一篇→
本文二维码信息
码上扫一扫!
适应手脸遮挡手语视频的手势检测算法
(1.兰州理工大学 电气工程与信息工程学院;2.兰州交通大学 机电工程学院)
Hand Detection Algorithm During Hand over Face Occlusion in Sign Language Video
(1.School of Electrical and Info. Eng.,Lanzhou Univ. of Technol.;2.School of Mechatronic Eng.,Lanzhou Jiaotong Univ.)
摘要
图/表
参考文献
相似文献
本文已被:浏览 1770次   下载 1
投稿时间:2013-07-04    修订日期:2013-10-09
中文摘要: 针对手脸遮挡条件下的手语视频手势检测问题,提出一种基于力场(force field)转换的手势检测算法。首先分别计算手脸遮挡帧和纯脸部帧的力场图像,然后将力场图像分块并统计各分块直方图特征,再将相同空间位置的分块直方图对应相减,得到各分块直方图灰度分量差,最后将各分块直方图灰度分量差与灰度阈值进行比较获得手部位置。实验证明该算法能够实时进行手脸遮挡条件下的手势检测。
中文关键词: 手势检测  手语视频  力场转换  直方图
Abstract:Based on image force field transformation, a novel algorithm was proposed to track the hand during hand over face occlusion in sign language video.First, the frames with a hand occluding the face and those with only a face were transformed to force field images.Then the force field images were partitioned into sub-images and the histograms of each sub-image were calculated.For each sub-image, the histogram of frame with only a face was subtracted from the frame with a hand occluding the face to get the difference histogram.Finally, for each sub-image the difference histogram was compared to threshold to get the position of the hand.Experimental results showed that the proposed algorithm is capable of real-time tracking of hand.
文章编号:201300715     中图分类号:    文献标志码:
基金项目:国家自然科学基金资助项目(61302116;61365003);教育部博士点基金资助项目(20116201110002);甘肃省自然科学基金资助项目(1212RJYA026;1212RJZA050)
作者简介:
引用文本:
陈晓雷,张爱华,林冬梅,杨欣翥.适应手脸遮挡手语视频的手势检测算法[J].工程科学与技术,2014,46(2):66-71.
Chen Xiaolei,Zhang Aihua,Lin Dongmei,Yang Xinzhu.Hand Detection Algorithm During Hand over Face Occlusion in Sign Language Video[J].Advanced Engineering Sciences,2014,46(2):66-71.