Fabric defect classification using radial basis function network

In this paper, a new approach for fabric defect classification using radial basis function (RBF) network improved by Gaussian mixture model (GMM) is investigated. First, the gray level arrangement in the neighborhood of each pixel is extracted as the feature. This raw feature is subject to principal component analysis (PCA) which adopts the between class scatter matrix as the generation matrix to eliminate the variance within the same class. Second, the RBF network with Gaussian kernel is used as the classifier because of the nonlinear discrimination ability and support for multi-output. To train the classifier, GMM is introduced to cluster the feature set and precisely estimate the parameter in Gaussian RBF, in which each cluster strictly conforms to a multi-variance Gaussian distribution. Thus the parameter of each kernel function in RBF network can be acquired from a corresponding cluster. The proposed algorithm is experimented on fabric defect images with nine classes and achieves superior performance, which proves its utility in practice.