High-Speed and Accurate Method for the Gear Surface Integrity Detection Based on Visual Imaging

In view of the defects of tooth profile defect and end face defect in the production process of injection gears due to the influence of technology, equipment and environment, there is an urgent need for rapid and accurate detection methods of gear surface integrity. In this paper, digital image processing technology is used to propose a rapid and accurate target matching detection method for the gear surface integrity based on normalized cross-correlation coefficients: First, the first derivative of the Gaussian function is used to construct a sub-pixel edge detector optimized based on the Canny operator to accurately extract the edge information of the gear to determine the target area; then, calculate the normalized correlation coefficient between the target area image and the template image, and realize the judgment of the surface integrity of the injection gear according to the matching coefficient threshold. The experimental results verify the feasibility and reliability of the algorithm. Its accuracy rate reaches 99.33%, and the detection time does not exceed 0.2s, which has a good industrial detection application value.