Research On High-precision Machine Vision Inspection Method Of Mobile Phone Surface Defects

At present,the penetration rate of mobile phones continues to increase,basically one person holds a mobile phone,and mobile phones have become a necessity in people's daily lives.As an important part of the mobile phone,the surface defects of the mobile phone affect the overall appearance of the mobile phone.Therefore,the detection of the surface defects of the mobile phone shell is an important part of the quality inspection of the mobile phone.In this paper,a mobile phone case made of metal is taken as the research object,and a high-precision method for detecting and identifying the surface defects of the mobile phone case is researched and designed for the types of defects such as scratches,scratches,pits,and particles on the surface..The main research contents and results are as follows:(1)Introduced the characteristics and technical difficulties of the research object,analyzed the mechanical structure,visual imaging,and electrical control of the design of the mobile phone shell surface d efect detection platform,selected the hardware,and designed the structure.(2)The mobile phone case positioning method is studied,and the positioning method flow is introduced.Aiming at the noise interference around the edge contour of the mobile phone shell,which affects the positioning error,an edge fitting based on Kalman filtering is used.Aiming at the orientation of the mobile phone case,an image rotation based on affine transformation is adopted,and the rotation angle is calculated and corrected according to the contour offset angle and the length of the offset side.Experiments show that the actual average positioning accuracy of the mobile phone shell positioning method in this paper reaches 0.15 mm,and the image pixel positioning accuracy reaches 2.02 pixel.Aiming at ROI region division,this paper proposes a mask processing method based on illumination model and column histogram.(3)The method for detecting the surface defects of the mobile phone case is studied.Aiming at the phenomenon of uneven illumination caused by unidirectional lighting,a new image enhancement algorithm combining retinex and logarithmic transformation is proposed to eliminate the effect of uneven illumination;for the complex texture of mobile phone shells,the def ects are small and there are many types Features,a defect detection method based on wavelet transform and surround suppression is proposed,and the image segmentation is performed.The experimental results show that the detection accuracy rate reaches 96.84%;for defect recognition classification,the classification method based on SVM support vector machine is used.The experimental results It shows that the recognition accuracy rate reaches96.00%.(4)Design high-precision visual inspection software for surface defects of mobile phone shells,and introduces the overall structure of the software system,software inspection process and system interface in detail.