Abstract: Gray-level threshold segmentation based on pixels distribution of high precision magnetic mark image defects detection method.In view of the magnetic mark image using histogram equalization method to image enhancement processing,to provide support for high precision image defects detection,based on the minimum error method to obtain the best segmentation threshold,complete the best segmentation of image,through to the various segmentation connected area of the scar after judging,realize to the magnetic mark the defects of image recognition,finish high precision magnetic mark image defects detection.The experimental results show that the improved algorithm for high precision of the magnetic mark image defect detection,can simplify the testing process,greatly improve the efficiency and accuracy of detection.