首页 | 官方网站   微博 | 高级检索  
     

应用几何矩和边缘检测的手势识别算法
引用本文:何阳清,葛元,王林泉.应用几何矩和边缘检测的手势识别算法[J].计算机工程,2005,31(15):165-166,174.
作者姓名:何阳清  葛元  王林泉
作者单位:上海海事大学计算机系,上海,200135
基金项目:上海市自然科学基金资助项目(02ZD14053)
摘    要:提出了一种结合几何矩和边缘检测的手势识别算法。手势图像经过二值化处理后,提取手势图像的几何矩特征,取出几何矩特征7个特征分量中的4个分量,形成手势的几何矩特征向量。在灰度图基础上直接检测图像的边缘,利用直方图表示图像的边界方向特征。最后,通过设定两个特征的权重来计算图像间的距离,对30个字母手势进行识别,识别率为90%。

关 键 词:手势识别  几何矩  边缘检测  边界方向直方图
文章编号:1000-3428(2005)15-0165-02
收稿时间:2004-06-26
修稿时间:2004-06-26

Gesture Recognition Algorithm Based on Invariant Moment and Edge Detection
He Yangqing,Ge Yuan,Wang Linquan.Gesture Recognition Algorithm Based on Invariant Moment and Edge Detection[J].Computer Engineering,2005,31(15):165-166,174.
Authors:He Yangqing  Ge Yuan  Wang Linquan
Abstract:A gesture recognition algorithm based on two features of invariant moment and edge detection is presented. After preprocessing, 4 of 7 invariant moments are extracted on binary image. By edge detection, histogram is formed to describe the edge information. Finally, the recognition of 30 letter gestures is performed by computing distance, in which different coefficients are set to these two features. The recognition rate of 90% is achieved.
Keywords:Gesture recognition  lnvariant moment  Edge detection  Histogram of edge direction
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号