Segmentation of Wear Debris Based on Edge Detection and Contour Classification

High-frequency sampling of wear debris of equipment in lube oil was performed by OLVF, based on which on-line wear monitoring was realized. Existing researches are mainly concentrated on extraction of wear debris concentration and size from ferrograms under transmitted light, while less concentrated on the segmentation algorithm of ferrogram under reflected light. In this paper, an wear debris segmentation algorithm based on edge detection and contour classification was proposed to realize OLVF segmentation of wear debris under reflected light, which overcomes excessive brightness or darkness of wear debris that are often neglected by the traditional segmentation algorithm.