A new efficient SVM-based edge detection method

被引：43

作者：

Zheng, S ^{[1
]}

Liu, H ^{[1
]}

Tian, JW ^{[1
]}

机构：

[1] Huazhong Univ Sci & Technol, Inst Pattern Recognit & Artificial Intelligence, State Educ Commiss Key Lab Image Proc & Intellige, Wuhan 430074, Peoples R China

来源：

PATTERN RECOGNITION LETTERS | 2004年 / 25卷 / 10期

关键词：

edge detection; least squares support vector machine; Gaussian radial basis function kernel; gradient and zero crossing operators;

D O I：

10.1016/j.patrec.2004.03.009

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

An innovative edge detection algorithm, using both the gradients and the zero crossings to locate the edge positions, is presented in this paper. Based on the least squares support vector machine (LS-SVM) with Gaussian radial basis function kernel, a set of the new gradient operators and the corresponding second derivative operators are obtained. Computer experiments are carried out for extracting edge information from real images and sharp image edges are obtained from a variety of sample images. Some of the best results are attained from a number of standard test problems. The performance of the proposed algorithm is compared with many other existing methods, including Sobel and Canny detectors. The experimental results indicate that the proposed edge detector is near equal to the Canny in the performance and is fast in the speed. (C) 2004 Elsevier B.V. All rights reserved.

引用

页码：1143 / 1154

页数：12

共 37 条

[1] QUANTITATIVE DESIGN AND EVALUATION OF ENHANCEMENT-THRESHOLDING EDGE DETECTORS [J].