Research On Shaft Sleeve Dimensional And Coating Surface Inspection System Based On Machine Vision

Machine vision inspecting technology has been widely applied in various of fields for its non-contact feature and high accuracy.Because of the general existence of corrosion in the petrochemical industry,especially for shaft sleeve and other metal parts,it is necessary to apply a certain thickness of anti-corrosion coating on shaft sleeve.Nowadays the shaft sleeve dimension and its coating surface defect were inspected separately,which were inefficient.Therefore,research of machine vision inspection was carried out and focused on the work of geometric dimension and surface defect inspection.The main content of this paper was as followed:Based on some relevant literatures,various of inspection methods were compared on their advantages and disadvantages.Machine vision technology was finally adopted.According to the study of machine vision detection technology in terms of theory and practical application,a scheme was proposed to detect the geometric dimensioning and coating of shaft sleeves.According to the requirements of the detection objects,hardware module selection was made and the experimental motion control system has been established.The image processing algorithms of machine vision inspection were analyzed,which includes image preprocessing algorithm and edge detection algorithm.Considering the characteristics of the detection object and the analysis of Zernike moment sub-pixel edge detection,a new algorithm for sub-pixel edge detection based on Sobel-Zernike moments is proposed,which was more efficient and robust.Based on the analysis of linear and circular feature detection and the advantages and disadvantages of Hough transform and Least square method,a new feature detection algorithm which combines Hough transform and Least square method was proposed.The experimental results show that the proposed method has higher accuracy and robustness.Based on the study of the commonly used image segmentation methods and surface coating methods,defect detection on the coating surface of the shaft sleeve was realized.With the powerful digital image processing function library of MATLAB software,the HMI for the shaft sleeve image processing was designed and the development of the design was analyzed in detail.The accuracy of the detection system is evaluated by means of experiments for the geometric dimensions of the shaft sleeve and the surface of the coating.