基于多层小波和共生矩阵的纹理表面缺损检测

被引:7
作者
韩彦芳
施鹏飞
机构
[1] 上海交通大学图像处理与模式识别研究所
关键词
图像识别; 小波变换; 共生矩阵; 纹理分类; 缺损检测;
D O I
10.16183/j.cnki.jsjtu.2006.03.014
中图分类号
TP274.4 [];
学科分类号
摘要
提出一种利用多层小波和共生矩阵进行纹理表面缺损检测的有效方法.该方法首先将缺损图像在不同水平上进行小波分解,得到一系列低频子图像和高频细节子图像;然后计算和分析各水平上高频细节子图像的共生矩阵特征;最后选择低频子图像进行小波合成得到无纹理图像进行检测.实验证明,该方法能够快速准确地进行纹理缺损检测.
引用
收藏
页码:425 / 430
页数:6
相关论文
共 9 条
[1]  
Texture classificationusing wavelet transform. Arivazhagan S,Ganesan L. Pattern Recognition Letters . 2003
[2]  
Orthogonal bases of compactly suppor-ted wavelets. Daubechies I. Communications in Pure Applied Mathematics . 1988
[3]  
Pattern recognition[P]. DECO GUSTAVO;SCHUERMANN BERND.中国专利:US6980689B1,2005-12-27
[4]  
Color image retrievalusing multispectral random field texture model andcolor content features. Khotanzad A,Hernandez O J. Pattern Recognition . 2003
[5]  
Fabric defect detection by Fou-rier analysis. Chan C,Pang K H. IEEE Transactions on Industry Applications . 2000
[6]  
Texture classifica-tion based on Markov modeling in wavelet featurespace. Shirazi M N,Noda H,Takao N. Image and Vision Computing . 2000
[7]  
Texturefeatures for image classification. Haralick R M,Shanmugam K,Dinstein I. IEEE Trans SystMan Cybern . 1973
[8]  
Discriminative trainingapproaches to fabric defect classification based onwavelet transform. Yang X,Pang G,Yung N. Pattern Recognition . 2004
[9]  
Feature analysis for scaledand rotated texture segmentation. Manian V,Vasques R. Computers and Industrial Engineering . 1997