To solve the problem that the elevator traction wheel slippage is difficult to detect quantitatively, a slippage detection method is proposed based on machine vision. The slip between the traction wheel and the wire rope will occur during the round-trip operation of the elevator, the displacement distance between the traction wheel and the wire rope in the circumferential direction is obtained through the image signal processing algorithm and related data analysis. First, the ROI (region of interest) of the collected original image is selected to reduce redundant information. Then, a nonlinear geometric transformation is carried out to transform the image into the target image with an equal object distance. Finally, the centroid method is used to obtain the slippage between the traction wheel and the wire rope. The field test results show that the absolute error of the system developed in this paper is 0.74 mm and the relative error is 2%, the extending uncertainty of the slip detection results is (33.8 ± 0.69) mm, the confidence probability is p = 0.95, and the degree of freedom is v = 8, which can meet accuracy requirements of elevator maintenance.

中文翻译：

---

基于机器视觉的电梯牵引轮滑动检测方法研究

针对电梯牵引轮打滑难以定量检测的问题，提出了一种基于机器视觉的打滑检测方法。电梯在往返运行过程中会发生牵引轮与钢丝绳之间的打滑，通过图像信号处理算法和相关数据分析得到牵引轮与钢丝绳在圆周方向上的位移距离。首先，选择采集到的原始图像的ROI（感兴趣区域）以减少冗余信息。然后进行非线性几何变换，将图像变换为等物距的目标图像。最后采用质心法求出牵引轮与钢丝绳之间的滑移量。