纺织学报 ›› 2008, Vol. 29 ›› Issue (10): 43-46.

• 纺织工程 • 上一篇    下一篇

基于灰度共生矩阵的织物组织结构差异分析

常利利;马俊;邓中民;蒋蕾   

  1. 武汉科技学院
  • 收稿日期:2007-10-07 修回日期:2008-06-04 出版日期:2008-10-15 发布日期:2008-10-15

Study on identification of fabric texture based on gray-level co-occurrence matrix

CHANG Lili;MA Jun;DENG Zhongmin;JIANG Lei   

  1. Wuhan University of Science and Engineering;Wuhan;Hubei 430073;China
  • Received:2007-10-07 Revised:2008-06-04 Online:2008-10-15 Published:2008-10-15

摘要: 应用灰度共生矩阵分析不同织物组织结构的纹理特征差异。首先将图像的灰度级由256级降为16级以减少计算量;再通过计算图像的灰度共生矩阵选取由灰度共生矩阵衍生的能量、对比度、相关、熵和逆差矩5种特征值作为描述织物纹理的特征参数,并与实际织物纹理特征相比较。分析结果表明,该方法能够较好地提取平纹、斜纹和缎纹织物的纹理特征,可进一步实现织物纹理的分类。

Abstract: By using gray-level co-occurrence matrix(GLCM),a method for analyzing the fabric texture is developed.Firstly,we reduce the amount of calculations by compressing the image grayscales from 256 to 16;Then calculate the GLCM,we select the GLCM parameters ″energy″,″contrast″,″correlation″, ″entropy″ and(″inverse) difference moment″ to describe the textile texture,and compare the analysis results with the actual fabric texture.It shows that the algorithm could extract and analyze the texture of plain,twill and satin weave efficiently,and fabric texture classification can be further realized.

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!