Research On Surface Defect Detection System Of Mobile Phone Shell Based On Machine Vision

With the increasingly rich functions of smart phones and the improvement of people's spending power,smart phones have gradually become an indispensable part of people's lives.As one of the important components of mobile phones,the appearance quality of mobile phone shells seriously affects consumers' judgment of mobile phone quality.Due to the many production steps of mobile phone shells,defects are likely to occur on the surface during the production process,so defect detection steps must be added to the production process to ensure product quality.At present,the inspection methods in the manufacturing process of mobile phone shells are mainly based on manual visual inspection,which not only has the problems of high cost,high false detection rate,and inconsistent inspection standards,but also the detection speed is slow,which is difficult to meet production needs.In recent years,inspection technology based on machine vision has perfectly compensated for the shortcomings of manual visual inspection due to its advantages of high stability,high speed,and high precision,and has been gradually applied to industrial production.Based on machine vision technology,this article takes the metal frosted mobile phone shell on the automated production line of precision electronic manufacturing robots as the object.Aiming at the typical defects such as scratches,scratches,sand spots and dirt on the surface during the manufacturing process,a set of mobile phone shells is designed.The surface defect detection system realizes the precise detection and classification function of metal frosted mobile phone shell defects on the automated production line of precision electronic manufacturing robots.The main research contents are as follows:First,analyze the specific research objects of this article,determine the types of defects and their detection standards,and then draw up the overall framework of the surface defect detection system according to the detection requirements,introduce in detail the core parts of the system such as mechanical structure,optical imaging,and electrical control,and analyze in detail The impact of different types of light sources,lighting methods,camera and lens types on image quality.Secondly,the mobile phone shell image preprocessing algorithm was studied,and the segmentation effects of three different types of background segmentation algorithms were compared,and a cluster-based background segmentation method was implemented,and the background was optimized by combining the least squares line fitting method.The segmentation effect solves the problem of unsmooth segmentation of the surface edge of the mobile phone shell.The Hough circle detection method and the pixel gray projection method are used to divide the different areas of the mobile phone shell image.A method based on the combination of morphological filtering and Gaussian filtering is used to filter out the frosted texture noise on the surface of the mobile phone shell.Thirdly,the mobile phone shell defect detection and classification algorithm is studied,the classic threshold segmentation algorithm is analyzed and compared,and a maximum between-class variance method based on entropy weighting is proposed to threshold the defect,and then the segmented image is performed Connected domain analysis to achieve precise segmentation of defects under frosted texture background.Finally,statistics and analysis of the characteristics of different types of defects to train the SVM classifier to identify and classify the defects.Finally,according to the research content and requirements,the mobile phone shell surface defect detection software platform is designed,and the design process of the overall design of the software system,interface development,and work flow are introduced in detail.