Robust Defect Detection Using Improved Higher Order Wavelet Descriptors and Support Vector Machines for Visual Inspection Systems.

This paper aims at investigating a novel solution to the problem of defect detection from textile images using the Discrete Wavelet Transform (DWT) Analysis, involving multiple wavelet bases, and the Support Vector Machines (SVM) classification approach, that can find applications in the design of robust quality control systems for the textile industry. The suggested solution focuses on detecting defects in textile manufacturing applications from the corresponding images DWT and higher order vector quantization related properties of the associated wavelet coefficients. More specifically, a novel methodology is investigated for discriminating defects by applying a supervised neural classification technique, namely SVM, to innovative multidimensional multi-wavelet based feature vectors. These vectors are extracted from the multi-wavelet bases K-Level 2-D DWT (Discrete Wavelet Transform) transformed original image using higher order Vector Quantization techniques and correlation analysis applied to these wavelet domain quantization vectors. The results of the proposed methodology are illustrated in defective textile images where the defective areas are recognized with higher accuracy than the one obtained by applying two rival defect detection methodologies. The first one of them uses all the wavelet coefficients derived from the k-Level 2-D DWT and involves SVM again in the classification stage, while the second one uses again all the wavelet coefficients derived from the k-Level 2-D DWT but involves a Multilayer Perceptron (MLP) neural network in the classification stage. The promising results herein shown outline the importance of judicious selection and processing of 2-D DWT wavelet coefficients for industrial pattern recognition applications as well as the generalization performance benefits obtained by involving SVM neural networks instead of other ANN models.