Surface Defect Inspection in Images Using Statistical Patches Fusion and Deeply Learned Features

Defect detection in images is a challenging task due to the existence of tiny and noisy patterns on surface images. To tackle this challenge, defect approach proposed paper using statistical data fusion. First, breaks large image that contains multiple separate defects into smaller overlapping patches detect each patch, conventional convolutional neural network approach. Then, fusion maintain spatial coherence cracks aggregate information extracted from enhance overall performance robustness system. The evaluated three benchmark datasets demonstrate its superior terms both individual patch inspection whole inspection.