Research on Fault Detection System of Vertical Shaft Hoisting Cage Shoes Based on Machine Vision

Aiming at the problems that the damage of the roller cage shoes in the shaft hoisting system cannot be found in time, the machine vision method is used to detect the thickness of the rubber wheel, and the real-time operation of the roller cage shoes is diagnosed to ensure the safe and stable operation of the skip. First of all, Retinex algorithm is used for image enhancement, and a combination of threshold segmentation and connected domain is used to extract the roller cage shoes images under complex background. Secondly, Zernike moments are used to extract sub-pixel edges to improve detection accuracy and RANSAC is used to perform ellipse fitting to reduce the influence of interference points. Finally, the irregular-shaped rubber wheel is sampled and measured, and the measured values are averaged to obtain the thickness value of the rubber wheel. The experimental results show that the average error of the measurement accuracy of the system is about 0.02mm, which meets the needs of the project. At the same time, the method avoids the danger of the inspectors during on-site inspection, and ensures the safety of the inspectors.