Research On Automobile Vehicle Cup-holders Assembly Defect Detection System Based On Machine Vision

As one of the automobile interior accessories,vehicle cup-holders provide the condition of drinking water for the driver.Although vehicle cup-holders seem unremarkable,they also affect the drivers’ driving comfort and safety.Therefore,it is essential to carry out quality inspection on the parts of the cup-holder.In the process of industrial production,the assembly defect inspection of vehicle cup-holders mainly depends on workers’ visual inspection.Due to the influence of human subjective factors and the inevitable visual fatigue of human eyes,the assembly defect inspection efficiency of vehicle cup-holders is low and the false detection rate is high,which leads to the decline of product quality.Based on the above reasons,this paper develops an assembly detection system based on machine vision technology to identify and detect the defects of vehicle cup-holders.Aiming at the structure of the vehicle cup-holder,this paper designed a lighting and image acquisition scheme suitable for assembly defect detection of vehicle cup-holders.According to the analysis of the possible assembly defects of the parts(foams,screws,clasps,non-woven fabrics,iron sheets and keys)of vehicle cup-holders,an assembly detection algorithm was designed to detect the defects of the vehicle cup-holder.The main work of this paper is as follows:1.Through the analysis of the structure of vehicle cup-holder and the defect types of it,the lighting scheme,image acquisition scheme and the overall mechanical structure of the system that are suitable for vehicle cup-holders detection were designed.By selecting the components of the system(industrial camera,lens,light source and ipc),a semi-closed assembly defect detection system was built.2.In this paper,the common assembly defects of vehicle cup-holder are analyzed and the corresponding assembly defect detection algorithm is designed.Specifically speaking,the Scharr operator is used for image processing of foam,and the threshold segmentation,Hole Filling algorithm and morphological operation(corrosion and expansion)are used to distinguish foam from the background.The contour detection algorithm based on Suzuki85 algorithm is designed by analyzing the detection requirements of iron sheet.The edge direction gradient template matching algorithm based on image pyramid is used to detect whether the key is pressed or not and to detect installation errors.Through the analysis of the screws installation process,a detection algorithm based on the screw centroid position is designed.The core is to use Suzuki85 algorithm to get the screw contour,and then use the contour distance to get the centroid position.3.The efficiency and accuracy of visual inspection system are analyzed and discussed by setting a contrast experiment between visual inspection and manual visual inspection.Through setting up the false alarm experiment,this paper analyzes the causes,and offers a corresponding solution.Finally,this paper analyzes and summarizes the research direction of how to further improve the efficiency and accuracy of the system algorithm and how to expand the system into a complete automatic device.