Automatic Classification of Woven Fabric Structure by Using Learning Vector Quantization

Abstract Current procedure for classification of woven fabric structure in the textile industry is performed by human,which wastes manpower and time.This paper proposes an automatic woven fabric image extraction and classification method.To speed up the system running time,2-D wavelet transform for fabric images is used firstly.Extracting the feature parameters of gray level co-occurrence matrix is a typical method for fabric texture image.Finally, identification and classification of the woven fabric structures exactly by using the learning vector quantization neural network.The experimental results show that focusing on learning vector quantization improves the classification performance as well as increases the computational efficiency.