Research On The 3D Vision Inspection Technology Of The Packaging Glue In Mobile Phone Based On The Structure Line Scanning

With the advent of the mobile & internet era,the smartphone has become an indispensable part of daily life.With the rapid development of the mobile phone industry,automated production and testing have become an urgent demand for the industry.Due to the special of the mobile phone modules in packaging and dispensing,during the packaging and production process of display screens and mobile phone shells,the glue will inevitably have defects such as glue breakage and overflow.If the problem is not found in time,it will bring irreversible trouble.Common defects of the packaging glue under the screen of the mobile phone can be divided into three kinds: no glue in the glue channel,dispensing too little results in lack of glue,too much glue leads to partial glue.These defects can be qualitatively judged by measuring the height and width of the cross-sectional profile of the glue.Due to the traditional image processing in the detection of the glue in the mobile phone shell under the complex background,it will encounter imaging challenges,as well as insufficient detection accuracy,missed detection,and wrong detection,and other problems.To overcome the defects such as cracking and overflowing of glue in the packaging process of mobile phones,a measuring method and algorithm for mobile phone packaging glue based on line-structure laser sensor(LMI Gocator2510)and HALCON software were proposed.In view of the above situation,this paper used LMI Gocator2510 line structured light sensor SDK and LEADSHINE motion control card SDK secondary development kit to integrate into a system which is composed of mechanical design,motion control,and image acquisition to perform image processing and image acquisition tasks,and achieve the online dispensing and real-time inspection of packaging glue.By using image preprocessing methods such as threshold segmentation,boundary contour extraction,edge point fitting,and ROI region generation are used to extract the characteristics of the glue area in the depth map of the phone shell.Separating the glue shell before dispensing and after dispensing to achieve the effective separation of no-glue depth maps area and glue depth maps area.All pixel values of the corresponding row coordinates in the segmented area were intercepted respectively,and then the contour line based depth map was reconstructed,and finally the depth contours by preprocessing the data to simplify the profile.Moreover,the least square method based on the contour information was used to fit the glue benchmark.Combining with the glue benchmark,the glue height after glue dispensing was measured.Finally,the width of the glue and various defects were judged based on the contour position after the glue dispensing.The experimental results show that the algorithm is capable to accurately measure the width and height of the cross-sectional profile of the glue in the mobile phone adhesive,which is packaged at micron-scale measurement,and further to judge the lack of glue and overflow defects accurately by setting the threshold data.At the same time,the measurement results of this method have stable data concentration.In order to verify the accuracy of the algorithm measurement,this paper also used the standard block measurement experiment to verify the detection accuracy of the measurement algorithm by dynamically measuring the height of the standard block.Through experiments,it is found that the repeated measurement accuracy of the algorithm can reach 0.01 mm,the measurement results of the algorithm glue width and height within a reasonable range and the accuracy can reach 99.74%.