楚雄师范学院学报 ›› 2023, Vol. 38 ›› Issue (3): 135-140.

• 计算机科学 • 上一篇    下一篇

基于Sobel算子和灰色关联分析的图像边缘检测

薛文格, 邝天福   

  1. 楚雄师范学院 数学与计算机科学学院,云南 楚雄 675000
  • 收稿日期:2022-04-12 出版日期:2023-05-20 发布日期:2023-08-07
  • 作者简介:薛文格(1982–),女,讲师,研究方向为模式识别与图像处理。E-mail: xwg@cxtc.edu.cn,Tel:15808785801

Image Edge Detection Based on Sobel Operator and Gray Correlation Analysis

XUE Wenge, KUANG Tianfu   

  1. School of Mathematics and Computer Science, Chuxiong Normal University, Chuxiong, Yunnan Province 675000
  • Received:2022-04-12 Online:2023-05-20 Published:2023-08-07

摘要: 针对灰色关联分析方法提取的边缘存在不连续、噪声敏感性等问题,本文将Sobel算子与灰色关联分析方法相结合,提出了一种基于邓氏关联度图像边缘检测的改进型算法,其融合了Sobel八方向算子技术,同时采用中值法对灰色关联度中的参考序列和比较序列实现单位一致性,并对传统的灰色邓氏关联度模型做了改进,用其求得的关联度提取图像边缘。仿真实验表明,本文算法提取的边缘连续、清晰,并在一定程度上抑制了噪声的干扰,提高了算法的效率。

关键词: 灰色关联分析, 邓氏关联度, Sobel算子, 边缘检测, 参考序列, 比较序列

Abstract: Edge extracted with the gray correlation analysis method have edge discontinuities, noise sensitivity and other problems. This paper combines Sobel operator with the method of gray correlation analysis to detect edge and presents an improved edge detection algorithm based on Deng's correlation degree that combines with Sobel eight direction operators technology and, at the same time, uses median method to achieve unit consistency for the reference sequence and comparison sequence of gray correlation degree, so as to improve the traditional gray Deng's correlation degree, with which correlation degree can be obtained to extract image edges. The simulation results show that the edges extracted with the proposed algorithm are continuous, clear and suppressing the interference of noise to a certain extent, thus improving efficiency of the algorithm.

Key words: gray correlation analysis, Deng's correlation degree, Sobel operator, edge detection, reference sequence, comparison sequence

中图分类号: