关键词: 机器视觉技术, 电梯曳引轮, 磨损检测, 双边滤波, 畸变模型, 双阈值化

Abstract: The wear detection method of elevator traction wheel based on machine vision technology was studied, and the non-contact accurate detection of the wear of elevator traction wheel was realized by using machine vision technology. The traction wheel image was collected by the combination of CCD camera and high power and high resolution optical lens. The distortion model was used to correct the distortion in the process of image acquisition. The distortion corrected image was preprocessed through bilateral filtering, gradient calculation and double threshold processing. By using the corrected image, the gap between the traction rope and the traction wheel groove of the elevator traction wheel was obtained by contour positioning and edge fitting method, and the final detection results of the traction wheel wear were obtained by calculating the shielding compensation of the gap between the traction wheel rope groove. The experimental results show that the proposed method can accurately detect the wear of traction wheels, and the test results are consistent with the actual results. The safety of traction elevators can be improved by using the wear test results.

Key words: machine vision technology, elevator traction wheel, wear detection, bilateral filtering, distortion model, double thresholding