首页 | 本学科首页   官方微博 | 高级检索  
     

一种改进非极大值抑制的Canny边缘检测算法
引用本文:李旭,王正勇,吴晓红,滕奇志. 一种改进非极大值抑制的Canny边缘检测算法[J]. 成都信息工程学院学报, 2011, 26(5): 564-569
作者姓名:李旭  王正勇  吴晓红  滕奇志
作者单位:四川大学电子信息学院图像信息研究所,四川成都,610064
基金项目:中央高校基本科研业务费资助项目(2009SCU11009);四川省科技支撑计划资助项目(10ZC0968)
摘    要:在传统Canny算法的非极大值抑制过程中,由于对比点的选择存在随意性而导致边缘检测不准确。针对此问题,提出了一种改进非极大值抑制过程的Canny边缘检测算法,利用目标像素点梯度方向周围4个像素点和邻域相关系数,在梯度方向上进行插值,代替传统算法中直接以梯度方向周围的邻近像素点作为对比点,实现新的非极大值抑制过程。通过实验证明:改进的边缘检测算法在边缘检测及虚假边缘抑制的性能方面有所提升。

关 键 词:计算机应用技术  数字图像处理  Canny算子  边缘检测  非极大值抑制

An Canny Edge Detection Algorithm Base on Improved Non-Maximum Suppression
LI Xu,WANG Zheng-yong,WU Xiao-hong,TENG Qi-zhi. An Canny Edge Detection Algorithm Base on Improved Non-Maximum Suppression[J]. Journal of Chengdu University of Information Technology, 2011, 26(5): 564-569
Authors:LI Xu  WANG Zheng-yong  WU Xiao-hong  TENG Qi-zhi
Affiliation:(School of Electronics and Information Engineering,Sichuan University,Chengdu 610064,China)
Abstract:In the non-maximum suppression process of Canny algorithm, the randomness in selecting contrastive points inaccuracy in edge detecting. To solve the problem, this paper proposes improved algorithm based on classical Canny algorithm. The improved algorithm interpolate points based on both four pixels around and the related parameter to replace the pixels near gradient direction as contrastive points, then the new non-maximum suppression process. The results of related experiments show that the improved algorithm has better performance on both edge detection and false edge suppression than the classical Canny algorithm.
Keywords:computer application technology  digital image processing  Canny algorithm  edge-detection  non-maximum suppression
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