IC element solder joint detection method based on robust principal component analysis

Abstract

The invention discloses an IC (integrated circuit) element solder joint detection method based on robust principal component analysis. The method includes: collecting the test pictures of a plurality of to-be-detected IC element solder joints to form a test set; respectively vectorizing each test picture of the test set into column vectors, merging all the obtained column vectors into an IC solder joint test matrix; conducting optimization decomposition treatment to obtain a low rank matrix and a sparse matrix; conducting reverse quantization on each column vector of the sparse matrix to obtain a corresponding picture, and then performing binarization segmentation on each picture to obtain a corresponding binary image; acquiring a frequency distribution diagram obtained according to training sample training, then combining the binary images to calculate the defect degree of each test picture; comparing the defect degree of the test picture with a defect degree threshold of the training sample so as to obtain an IC element solder joint detection result. The method provided by the invention can accurately, effectively and quickly detect IC element solder joint defects, and can be widely applied to the IC element solder joint detection industry.

Description

technical field [0001] The invention relates to the application field of digital image processing, in particular to a method for detecting solder joints of IC components based on robust principal component analysis. Background technique [0002] Glossary: [0003] RPCA: full name Robust principle component analysis, robust principal component analysis; [0004] PCB: full name Printed Circuit Board, printed circuit board; [0005] IC: Integrated Circuit, integrated circuit; [0006] Printed circuit board defect inspection is a hot topic in the application of automatic optical inspection (AOI), which has received more and more attention in recent years. At present, the main detection method is to detect the images of PCB components through CCD and then process them to realize defect detection. In actual use, the situation encountered by printed circuit boards is very complicated, and the images of PCB components collected by CCD often have varying degrees of changes and ir...

AI Technical Summary

Problems solved by technology

However, the current statistical modeling methods of this kind are based on pixels. In target detection, pixel-based methods often consider the background part too simply, because this method mainly considers the background to be static. Or it changes slowly, and in the qualified IC pin picture samples, each picture is very different, that is, the background changes drastically

This will seriously inhibit the effect of training and detection, resulting in unsatisfactory model training results, resulting in inaccurate detection results

[0009] In general, the current detection methods cannot effectively, accurately and quickly detect the solder joint defects of IC components

Images