基于二阶局部纹理熵的边缘检测方法  

Edge-Detection Method Based on Regional Texture ENT2

作　　者：刘宸[1] 刘智[1] 刘长建[1] 徐奥[2] 

机构地区：[1]信息工程大学,河南郑州450001 [2]东北大学测绘遥感与数字矿山研究所,辽宁沈阳100819

出　　处：《测绘科学技术学报》2017年第1期74-78,共5页Journal of Geomatics Science and Technology

基　　金：信息工程大学校立新兴科研基金项目(2016609602)

摘　　要：针对传统的边缘检测方法将灰度变化较为平缓的区域误判为边缘的问题,提出了基于二阶局部纹理熵的边缘检测方法。利用了二阶局部纹理熵能表示局部区域灰度变化大小的性质,并采用灰度错位压缩法消除灰度值压缩造成的条纹现象。实验表明,本文提出的方法能有效地检测出各类边缘,解决了上述的误判问题;通过不同大小窗口的边缘检测效果比较,建议在采用基于二阶局部纹理熵的边缘检测方法时将窗口设置为3×3。In view of the misjudgment from area with gentle gray scale changes to edges,the edge-detection method based on regional texture ENT2 has been proposed in this paper.The characteristics of regional texture ENT2,reflecting the size of local gray change,was utilized in the proposed method.Gray-scale dislocation processing before gray-scale compression was proposed to eliminate the phenomenon of the stripes in result of gray-scale compression.The experiments with the different styles of edges and picture show that the proposed method has a great performance on edge-detection and solves the misjudgment problem.By comparison,3×3 window is advised to be adopted in edge-detection method based on regional texture ENT2.

关 键 词：边缘检测 二阶纹理熵 局部纹理熵 阶跃型边缘 线状型边缘 斜坡型边缘 屋顶型边缘 

分 类 号：P237[天文地球—摄影测量与遥感]