首页 | 本学科首页   官方微博 | 高级检索  
     检索      

最小核值相似区低层次图像处理算法的改进及应用
引用本文:陆宏伟,于起峰.最小核值相似区低层次图像处理算法的改进及应用[J].应用光学,2000,21(1):32-36.
作者姓名:陆宏伟  于起峰
作者单位:国防科技大学,航天技术系,湖南,长沙,410073
摘    要:首先介绍一种能有效地进行边缘、角点检测和滤波等低层次图像处理的最小核值相似区算法,然后提出自适应阈值的选取方法,局部区域灰度重心判据对其算法的改进使得边缘检测算法抗噪能力更强。针对序列图像的具体应用,用改进的边缘检测算法能准确、快速地从噪声图像中得到较准确的边缘信息,用滤波算法对序列图像作预处理,可使互相关跟踪结果更可靠、更准确。

关 键 词:SUSAN算法  边缘检测  角点检测  平滑滤波
收稿时间:1999/4/11
修稿时间:1999-04-11

THE IMPROVEMENT OF SUSAN LOW LEVEL IMAGE PROCESSING AND IT'S APPLICATIONS
LU Hong-wei,YU Qi-feng.THE IMPROVEMENT OF SUSAN LOW LEVEL IMAGE PROCESSING AND IT'S APPLICATIONS[J].Journal of Applied Optics,2000,21(1):32-36.
Authors:LU Hong-wei  YU Qi-feng
Abstract:Low level image processing is a key essential task in computervision.A new approach to low level image processing,SUSAN algorithm,is estimated and recommended in this paper.Then we made some improvement of the approach in edge detection.An adaptive threshold is developed to replace the interactive threshold in SUSAN algorithm.A new criterion for removal of fake edge points is proposed.Finally an edge finder with subpixel accuracy based on SUSAN algorithm is developed.Experimental results show the success of the improvement of SUSAN approach.
Keywords:SUSAN algorithm  edge detection  corner detection  smoothing filter  
本文献已被 CNKI 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