Fractal dimension estimation for texture images: A parallel approach

被引：71

作者：

Biswas, MK

Ghose, T

Guha, S

Biswas, PK ^{[1
]}

机构：

[1] Indian Inst Technol, Dept Elect & Elect Commun Engn, Kharagpur 721302, W Bengal, India

[2] Indian Inst Technol, Dept Math, Kharagpur 721302, W Bengal, India

来源：

PATTERN RECOGNITION LETTERS | 1998年 / 19卷 / 3-4期

关键词：

fractal dimension; Differential Box Counting; parallel algorithm; SIMD army processor;

D O I：

10.1016/S0167-8655(98)00002-6

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Fractal dimension is an important parameter that can be used in various applications, such as, estimation of roughness in an image, texture segmentation, surface roughness estimation and many others. A number of techniques for fractal dimension computation in the digital domain have been reported in the literature. A parallel implementation of the Differential Box Counting technique is reported in this paper. The accuracy and computational complexity of the parallel implementation are also discussed. (C) 1998 Elsevier Science B.V. All rights reserved.

引用

页码：309 / 313

页数：5

共 11 条

[1] TEXTURE SEGMENTATION USING FRACTAL DIMENSION [J].