节点文献

基于多尺度数学形态学的边缘检测

Multi-Scale Edge Detection Based on Mathematical M orphology

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 周维忠赵海洋孙国基冯心海

【Author】 Zhou Weizhong1) Zhao Haiyang1) Sun Guoji 2) Feng Xinhai1) 1) Department of Electronic Engineering,Fo shan Institute of Science and Technology Foshan,528000 (2) Systems Engineering Institute, Xi′an Jiaotong University Xi′an,7 10049)

【机构】 佛山科学技术学院电子系!佛山528000西安交通大学系统工程研究所!西安710049

【摘要】 提出了一种基于多尺度数学形态学的图像边缘检测方法 ,利用数学形态学在描述灰度图像方面的优势 ,分析了在不同尺度下边缘提取的特点 ,采用非极大值运算方法提取边缘点 ,并与其他形态学边缘检测法进行了比较 ,给出了在含噪条件下边缘提取的实验结果。

【Abstract】 A multi-scale edge detection method based on the mathe matical morpho-logy is proposed. Multiscale primitives are defined to be in agr ee ment with the spatial size of image objects to implement the multi-scale analy sis. This paper analyses features of edge detection on different sizes. Making u se o f advantages of morphological method to describe grayscale image, gray-scale pr ism structuring element and an extensive morphological operator are defi ned. Structural elements of different sizes are used to extract features at diff erent scales,and the edge points are extracted by using non-maximal operation technique, and it is compared with other morphology edge detectors. Experimental res ults obtained by applying the edge detectors on a real image with and without no ise are presented.

  • 【文献出处】 数据采集与处理 ,JOURNAL OF DATA ACQUISITION & PROCESSING , 编辑部邮箱 ,2000年03期
  • 【分类号】TP391.4
  • 【被引频次】77
  • 【下载频次】414
节点文献中: 

本文链接的文献网络图示:

本文的引文网络