| 多尺度自适应加权形态边缘检测方法 |
| |
| 引用本文： | 杨述斌,彭复员,张增常.多尺度自适应加权形态边缘检测方法[J].华中科技大学学报(自然科学版),2002,30(10):87-89. |
| |
| 作者姓名： | 杨述斌 彭复员 张增常 |
| |
| 作者单位： | 1. 华中科技大学,电子与信息工程系
2. 襄樊学院物理系 |
| |
| 基金项目： | 国防科技预研基金资助项目 |
| |
| 摘 要： | 针对常用的多尺度边缘检测方法因卷积运算造成模糊图像边缘，提出一种多尺度自适应加权形态边缘检测方法，利用大小不同的结构元素提取图像边缘特征，在大尺度下抑制噪声，可靠地识别边缘，在小尺度下定位，再由粗到细跟踪边缘，得到边缘的位置，再由各尺度结构元的抗噪性确定加权值的大小，然后加权处理得到最后的边缘结果，实验表明，用多尺度自适应加权形态边缘检测算法可得到较理想的图像边缘。
|
| 关 键 词： | 图像处理 多尺度结构 特征提取 加权合成 多尺度自适应加权形态边缘检测 |
| 文章编号： | 1671-4512(2002)10-0087-03 |
| 修稿时间： | 2002年4月23日 |
Morphological edge detector with multiscale adaptive weighting |
| |
| Yang Shubin Peng Fuyuan Zhang Zengchang Master, Dept. of Electronics & Information Eng.,Huazhong Univ. of Sci. Tech.,Wuhan ,China..Morphological edge detector with multiscale adaptive weighting[J].JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE,2002,30(10):87-89. |
| |
| Authors: | Yang Shubin Peng Fuyuan Zhang Zengchang Master Dept of Electronics & Information Eng Huazhong Univ of Sci Tech Wuhan China |
| |
| Affiliation: | Yang Shubin Peng Fuyuan Zhang Zengchang Master, Dept. of Electronics & Information Eng.,Huazhong Univ. of Sci. Tech.,Wuhan 430074,China. |
| |
| Abstract: | Common multiscale edge detection blurs the edge of image because of convolving algorithm. This paper is intended to present a multiscale adaptive weight adding algorithm of morphological edge detecting. Different image edge can be detected by different scale elements. Image noise can retained by using big scale element; at the same time edge position is gained by using multiscale morphological edge detecting operators and tracked accurately by using small element. Then different weights are ascertained by different scale element. The edges gained with different scale elements are processed by weighting method. And thus the ultima edge is taken hold of. The experiment proves that it is a good effective edge detection algorithm. |
| |
| Keywords: | image processing edge detection multiscale analysis adaptive algorithm morphological edge detection |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
