SURFACE DEFECT INSPECTION OF COLD-ROLLED STRIPS WITH FEATURES BASED ON ADAPTIVE WAVELET PACKETS

The defects in the cold rolled strips have textural characteristics, w hich are nonuniform due to its irregularities and deformities in geome trical appearance. In order to handle the textural characteristics of images with defects, this paper proposes a surface inspection method b ased on textural Feature extraction using the wavelet transform. The w avelet transform is employed to extract local features from textural i mages with defects both in the frequency and in the spatial domain. To extract features effectively, an adaptive wavelet packet scheme is de veloped, in which the optimum number of features are produced automati cally through subband coding gain. The energies For all subbands of th e optimal quadtree of the adaptive wavelet packet algorithm and four e ntropy features in the level one LL subband, which correspond to the l ocal features in the spatial domain, are extracted. A neural network i s used to classify the defects of these features. Experiments with rea l image data show good training and generalization performances of the proposed method.