Automatic Recognition of Fabric Weave Patterns Using Texture Orientation Features
A novel approach is proposed for detecting fabric texture orientations and recognizing weave patterns. Wavelet transform is suited for fabric image decomposition and Radon transform is fit for line detection in fabric texture.Their excellent performances are combined to detect texture orientations in this study. Since different weave patterns have their own regular orientations in the original image and sub-band images composed by Wavelet transform, these orientations features are extracted, and then used as SOM inputs to achieve an automatic recognition of fabric weaves.The contribution of this algorithm is that it not only identifies basic fabric weaves but also accurately classifies some derivative twill weaves such as angular twills and pointed twills. The experimental results show that the proposed method is feasible and effective.
Wavelet Transform Radon Transform Fabric weave pattern SOM
Jianqiang Shen Zhaofeng Geng Xuan Zou Yongbin Pan
College of Information Science and Technology,Donghua University Shanghai, 200051, China;Shanghai -H College of Information Science and Technology,Donghua University Shanghai, 200051, China Shanghai -Hamburg Joint College, University of Shanghai for Science and Technology Shanghai, 200031,
国际会议
杭州
英文
494-498
2006-10-12（万方平台首次上网日期，不代表论文的发表时间）