Morphological segmentation of binary patterns

被引:430
作者
Soille, Pierre [1 ]
Vogt, Peter [2 ]
机构
[1] Joint Res Ctr, Spatial Data Infrastruct Unit, Inst Environm & Sustainabil, I-21027 Ispra, Italy
[2] Joint Res Ctr, Land Management & Nat Hazards Unit, Inst Environm & Sustainabil, I-21027 Ispra, Italy
关键词
Morphological image analysis; Pattern segmentation; Connectivity; LINEAR-TIME; ALGORITHM;
D O I
10.1016/j.patrec.2008.10.015
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper presents a method for segmenting binary patterns into seven mutually exclusive categories: core, islet, loop, bridge, perforation, edge, and branch. This is achieved by applying a series of morphological transformations such as erosions, geodesic dilations, reconstruction by dilation, anchored skeletonisation, etc. The proposed method depends on a single parameter only and can be used for characterising binary patterns with emphasis on connections between their parts as measured at varying analysis scales. This is illustrated on two examples related to land cover maps and circuit board defect detection. (c) 2008 Elsevier B.V. All rights reserved.
引用
收藏
页码:456 / 459
页数:4
相关论文
共 11 条