Research On Fast Location And Detection Method Of PCB Solder Joint Based On Linear CCD

PCB industry plays a very key role in the economic development of Our country.In the PCB production and processing of many links,the quality inspection of solder joints is needed.In recent years,in order to improve the automation and efficiency of PCB solder spot quality inspection,the automatic optical inspection technology of PCB solder spot based on CCD camera image has attracted more and more attention.In order to solve the problem of PCB solder spot location and defect detection in dynamic environment,a fast PCB solder spot location and detection method based on linear CCD is studied in this paper.The main work completed is as follows:(1)The structure and working principle of linear CCD are introduced,and the existing defect detection methods are summarized.Aiming at the problems of inaccurate positioning and defect classification in solder spot detection,the general idea and process of PCB solder spot detection based on linear CCD are proposed.(2)Aiming at the problems of the rapidity and accuracy of solder joint positioning in dynamic environment,this paper proposes a PCB solder joint positioning method based on SURF algorithm image feature point detection and matching.By matching PCB image to be detected with template image,and calculating the homography matrix between the two to achieve accurate location of solder joint coordinates.Combined with the characteristics of linear CCD imaging,image rolling Mosaic is adopted to improve the speed of image matching.(3)In view of the problem of defect identification accuracy with fewer defect samples,a PCB solder spot defect detection method based on multi-classification SVM is proposed.Gray scale and geometric features were extracted from solder joint images,and feature selection was carried out by the method of maximum correlation and minimum redundancy.Then,a multi-classification SVM classifier was constructed based on various solder joint features to realize the classification and recognition of six solder joints.In this paper,linear CCD is used as PCB image acquisition equipment,which can obtain higher resolution images at a lower cost.Moreover,the working mode of linear CCD scanning imaging can be well applied to industrial assembly line production,and the speed of image matching and detection efficiency can also be improved through image rolling stitching.The SVM based classifier can guarantee the high accuracy of solder spot defect detection under the condition of fewer defect samples.Through experimental testing,the welding spot positioning and defect detection method designed in this paper can achieve ideal results in the detection rate and accuracy,which proves the effectiveness of the method.