基于共生矩阵和形态学的彩色图像边缘检测  

Color image edge detection based on co-occurrence matrix and morphology

作　　者：杜棋东 陈雪梅[2] 

机构地区：[1]广东工业大学计算机学院,广东广州510006 [2]广州铁路职业技术学院

出　　处：《计算机时代》2014年第6期1-3,共3页Computer Era

基　　金：广州铁路职业技术学院"十一五"重点课题(GTXYZ1106)

摘　　要：针对在RGB空间中难以有效区分颜色相似性的问题,提出了一种基于共生矩阵和HSI空间形态学的彩色图像边缘检测方法。采用二维共生矩阵直方图均衡化方法,解决图像过渡不自然的现象;针对HSI的分量利用形态学方法进行加权分析和图像融合,得到彩色图像边缘。通过MATLAB实验表明,该算法检测准确率高,自适应性好,有较好的检测效果。To solve the difficulty of effectively distinguishing color similarity in edge based on co-occurrence matrix and HIS space morphology is described. Equalization method is utilized to solve the problem that the image transition is weighted analysis and image fusion are used according to morphology method from MATLAB experiments demonstrate that the proposed algorithm has high RGB space, a Two-dimensional detection method of color image matrix histogram not natural. In addition, for the HIS component, to obtain color image edge. The results obtained cognition rate, good self-adaptability and better detection effect.

关 键 词：共生矩阵 HIS空间 形态学梯度 边缘检测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]