Surface defects detection of steel plate based on visual attention mechanism

Abstract

For the image of present surface defects in steel plates, a surface defect detection method based on human visual attention mechanism is proposed. Based on the physiological structure and function of human visual system, this paper establishes a steel surface defect detection model via Gaussian filter, Gabor filter and the calculation such as cross-scale reduction, cross-scale addition and normalization, from which we can get the feature maps, the saliency map and so on. The experimental results show that the method can not only detect the defect area accurately, but also can meet the requirements of online real-time detection.