Edge detection by Helmholtz principle

被引：155

作者：

Desolneux, A ^{[1
]}

Moisan, L ^{[1
]}

Morel, JM ^{[1
]}

机构：

[1] ENS Cachan, CMLA, F-94235 Cachan, France

来源：

JOURNAL OF MATHEMATICAL IMAGING AND VISION | 2001年 / 14卷 / 03期

关键词：

image analysis; perception; Helmholtz principle; edge detection; large deviations;

D O I：

10.1023/A:1011290230196

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We apply to edge detection a recently introduced method for computing geometric structures in a digital image, without any a priori information. According to a basic principle of perception due to Helmholtz, an observed geometric structure is perceptually "meaningful" if its number of occurences would be very small in a random situation: in this context, geometric structures are characterized as large deviations from randomness. This leads us to define and compute edges and boundaries (closed edges) in an image by a parameter-free method. Maximal detectable boundaries and edges are defined, computed, and the results compared with the ones obtained by classical algorithms.

引用

页码：271 / 284

页数：14

共 34 条

[1] AUTOMATIC THRESHOLDING OF GRAY-LEVEL PICTURES USING TWO-DIMENSIONAL ENTROPY [J].