纺织学报 ›› 2011, Vol. 32 ›› Issue (8): 142-146.

• 管理与信息化 • 上一篇    下一篇

织物图像增强矩阵特征模型的建立

刘哲   

  1. 天津工业大学,中原工学院
  • 收稿日期:2010-08-02 修回日期:2011-04-13 出版日期:2011-08-15 发布日期:2011-08-15
  • 通讯作者: 刘哲 E-mail:xyliuzhe@163.com
  • 基金资助:

    河南省重大科技攻关项目(082102210026);省级

Establishment of Enhanced Matrix Feature Model for Fabric Image

  • Received:2010-08-02 Revised:2011-04-13 Online:2011-08-15 Published:2011-08-15

摘要: 针对目前缺乏有效显现织物特征的成熟模型,使织物疵点识别效果不佳的现状,本文提出了一种新织物图像特征模型—“增强矩阵特征模型”。该特征模型以图像的灰度值为基础,引入一种新的增强矩阵。该矩阵由根据织物图像梯度变化生成的矩阵算子组成,可对像素灰度值进行变换计算以放大或缩小图像局部特征,使图像的特征显现更加层次分明。通过采用MATLAB7.0编写程序验证该特征模型,发现该模型对织物疵点特征的显现效果明显,使疵点区域的特征变异得到明显增强,为织物疵点识别提供了一个新的理论依据。

Abstract: Aiming at the present situation of lacking of mature model for fabric feature and not better recognition effect, this paper proposes a new feature model of fabric image, called as “enhanced matrix feature model”. Firstly, this model introduces a new enhanced matrix based on image gray. Secondly, this matrix is made of matrix operators generated by gradient variation of fabric image, which can do transform calculation for pixel value to amply or reduce local characteristic of image to make image feature obvious. Finally, we compile some program to verify this model using MATLAB7.0, and we find this model is suitable for defect feature recognition of fabric and recognition effect is better. With this model, feature change in defect regions is enhanced obviously. Therefore, a new theoretical basis of defect recognition of fabric is provided in this paper.

中图分类号: 

  • TS 103.7
[1] 史伟民 陈春松 沈加海 彭来湖. 电脑横机自动起底控制系统设计[J]. 纺织学报, 2013, 34(3): 127-131.
[2] 刘哲. 基于图像分析的织物外观质量综合评价[J]. 纺织学报, 2012, 33(11): 61-65.
[3] 汪秀琛, 李晓久. 基于复合条件的织物纹理模糊识别[J]. 纺织学报, 2012, 33(8): 55-58.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!