Surface wear detection of automotive cermet composite brake pads based on machine vision

To improve the accuracy and speed of surface wear detection for automotive metal-ceramic composite brake pads, a machine vision-based surface wear detection method for automotive metal-ceramic composite brake pads is studied. A CCD industrial camera is used to capture images of automotive metal-ceramic composite brake pads, and improved Retinex algorithm to enhance image texture features. Based on the principle of maximum entropy, a reasonable threshold is set to segment and extract the target area of the enhanced brake pad image. Using the target area of the brake pad image as input and the surface wear of the brake pad as output, a fuzzy neural network is used to construct a brake pad surface wear detection model. The experimental results indicate that the detection method studied can accurately detect the surface wear samples of brake pads, and the detection time is less than 500 ms. © 2024 Inderscience Enterprises Ltd.