Abstract:  Aiming at the problem that the existing measurement method of main girder deflection of gantry crane was not safe, convenient and economical, a machine vision measurement method based on improved Canny edge detection algorithm was proposed. Firstly, the shape images of the main girder before and after loading the rated load of the crane were obtained, and the images were preprocessed by cutting and segmentation. Then, the improved Canny algorithm was used to extract the edge features from the mid-span of the main girder after preprocessing. By measuring the relative displacement of the extracted edge in the image, the actual lower deflection value of the crane was obtained according to the scale conversion. Finally, a crane lower deflection measurement system based on this method was developed on LabVIEW platform. The test results show that the system can accurately identify and extract the target edge features of the main girder of the crane under various working conditions, and the error of the calculated lower deflection value is within 0.5%. The research results show that the measurement method can meet the requirements of the engineering measurement of the lower deflection of lifting machinery, and provides a reliable method for crane inspection engineering application.

Key words:  lower deflection measurement; machine vision; Canny algorithm; image processing

ZHANG Bai, YUAN Zhe-yuan, YAN Bing-sheng, et al. Crane deflection measurement method based on improved Canny algorithm［J］.Journal of Mechanical & Electrical Engineering, 2021,38(9):1191-1196.