Flank wear measurement by a threshold independent method with sub-pixel accuracy

This paper presents an image processing procedure to detect and measure the tool flank wear area. Unlike the traditional thresholding-based methods, a rough-to-fine strategy is considered in this paper whereby a binary image is first obtained and used to find the candidate wear bottom edge points; then a threshold-independent edge detection method based on moment invariance is employed for more robust determination of the wear edge with sub-pixel accuracy. To shorten computation time, a critical area is initially defined and the subsequent procedure is confined to processing this area as the region of interest. Images from three types of inserts, A30N, AC325 and ACZ350 under different cutting conditions are captured with the similar illumination conditions after milling. The measured results obtained with the proposed method from these images are compared with those obtained by direct manual measurement with a toolmaker's microscope and a method based totally on binary image contour detection. The proposed method is shown to be effective and suitable for the unmanned measurement of flank wear.