利用灰色理论进行图像边缘检测  

Image Edge Detection Based on Gray System Theory

作　　者：胡鹏[1] 傅仲良[1] 陈楠[1] 

机构地区：[1]武汉大学遥感信息工程学院,武汉市珞喻路129号430079

出　　处：《武汉大学学报（信息科学版）》2006年第5期411-414,共4页Geomatics and Information Science of Wuhan University

基　　金：国家自然科学基金资助项目(60175022)

摘　　要：提出了一种基于灰色系统理论的边缘检测算法。该算法根据边缘点在像素邻域数列上表现出的与非边缘点的不同特征,分别选择典型非边缘像素的邻域数列和待检测图像所有像素的邻域数列作为参考数列和比较数列,利用斜率关联度计算两者之间的灰关联度,据此区分是否为边缘像素。实验表明,该算法对于椒盐噪声和随机噪声有较好的抗噪能力,提取边缘效果较好。This paper puts forward a novel algorithm for image edge detection based on gray correlation degree. First, this algorithm decides employing typical non-edge pixel neighboring sequence of gray level as referential sequence, simultaneously regards all pixel neighboring gray level sequence of disposed image as compared sequence, then computes gray correlation of referential sequence and compared sequence using slope correlation degree, finally judges whether a pixel is edge point based on the comparision of the computed gray correlation and the threshold value. To test this new algorithm the authors make experiments and the result shows this algorithm can gain more precise edge pixel than some traditional methods, futermore it can tolerate some noise as salt and stochastic.

关 键 词：边缘检测 灰色系统理论 灰关联分析 斜率关联度 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置] P237.4[自动化与计算机技术—控制科学与工程]