Application of back-propagation Neural Network Fuzzy Clustering in textile texture automatic recognition system

• Published in: 2008 International Conference on Wavelet Analysis and Pattern Recognition Vol. 1; pp. 46 - 49
• Main Authors: Te-Li Su, Fu-Chen Kung, Yu-Lin Kuo
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2008
• Subjects: Artificial neural networks; Automatic inspection system; Fabrics; Fuzzy clustering; HSV color space; Image color analysis; Neural network; Pattern recognition; Transforms; Wavelet analysis; Wavelet transform; Wavelet transforms
• ISBN: 9781424422388; 1424422388
• ISSN: 2158-5695
• DOI: 10.1109/ICWAPR.2008.4635748

This study proposed to use wavelet transfer to acquire image hue and value as image features, and use Back-propagation Neural Network Fuzzy Clustering analysis to recognize type of textile texture. Firstly, the RGB color space of original color image is transferred to HSV color space; secondly, wavelet transfer is applied to obtain vertical, horizontal and diagonal images of hue and value, and compute its wavelet energy to take them as texture features of this image. Finally, the back-propagation neural network is adopted to make fuzzy clustering analysis of this image texture feature. As indicated by experimental result, this system can recognize accurately plain weave, twill weave and satin weave textures in woven fabric, single and double textures in knitted fabric, and nonwoven texture in nonwoven fabric. Among 300 test samples in total where there are 50 samples each kind of fabric texture, the general recognition rate amounts to 97.67%. Therefore, this study succeeded in building the automatic computer visual inspection system to recognize textile texture type, which can greatly improve and avoid current low efficiency, non-objective judgment and labor waste due to human inspection.