Abstract: The image edge of steel strip surface defects cannot be detected exactly on-line at present. It is a serious problem to be solved. A wavelet-based maximum modulus algorithm is therefore proposed for edge protection, which can coordinate well the edge detection precision and denoising effect to extract the surface defect edges on multi-scale. Experimental results showed that the new algorithm can exclude well the fault detect edge from detection with the particulars of defect edge kept up, thus providing better edge detectivity for the subsequent online processing in surface defect detection, such as image auto-segmentation, defect identification, etc.