Research And Application Of SOP Chip Pin Defect Detection Based On Machine Vision

Chip packaging can protect the internal structure of the chip,support the chip,connect the chip's electrodes with external circuits,and improve the reliability of the chip.Because of the rapid development of chip packaging in the direction of miniaturization and high performance,pin defect detection after chip packaging is the necessary step to determine the correct chip packaging.SOP chip is one of the mainstream chip packaging forms.Because the pin size of SOP chip is small,the shape of SOP chip is curved and the surface of SOP chip reflects mirror,so the requirement for professional visual light source is high.This paper focuses on the research of SOP chip pin defect detection algorithm,which overcomes the traditional detection algorithm's strict requirements on detection environment.At the same time,the detection system of pin length,width and pitch of SOP chip is established,and the algorithm studied before is applied to it.The main works of this thesis are as follows:(1)The influence of light intensity in pin detection is studied.The influence of image gradient and image binarization on light intensity is analyzed.At the same time,the characteristics of gray histogram of chip images with different exposure are explored,and the evaluation method of camera exposure for chip detection is proposed.(2)By comparing the commonly used edge detection algorithms,we choose the algorithm suitable for SOP image edge detection.This paper focuses on the determination of the threshold of the edge detection algorithm.An adaptive threshold is added to the traditional edge detection algorithm,which enables the algorithm to adapt to the interference of light intensity change when image binarization,and enhance the stability of the binarization effect in the process of image preprocessing.In the process of improvement,the histogram equalization algorithm is optimized,and it is introduced into the threshold determination algorithm,and the suitable image exposure range of the Tianjin algorithm is extended.(3)The detection and location algorithm for SOP chip is studied,and the overall position of SOP chip is positioned by image moments.At the same time,it combines the method of connected domain analysis to improve the process of chip pin appearance detection.At the same time,it solves the problem of chip anti-dummy interference in the detection process.(4)Design and implement the pin defect detection system of SOP chip.According to the actual needs of the detection,the overall design and selection of hardware equipment are carried out,and the previous research on the evaluation of exposure,improved image preprocessing algorithm,chip pin location algorithm are added to the detection system for application.The detection results are analyzed from two angles of system error and random error,and the detection system is evaluated.