Study on steel plate scratch detection based on improved MSR and phase consistency

Abstract In the casting process of steel plate, due to influence rolling equipment and technology, defects such as cracks scratches appear on surface which affect performance plate even cause production accidents. this paper, an automatic detection method for scratch is proposed. Firstly, image decomposed by channel enhanced obtained improved MSR (Multi-Scale Retinex) enhancement algorithm. Then, phase consistency detected after Log Gabor wavelet transform areas are threshold segmentation intersection three channels. Finally, position identified characteristics width length can be calculated. The results show that minimum error measurement only 2.28% in experimental environment 4.15% field environment, mean running time 0.2826 s 0.3193 environment. It verifies proposed effective practical.