Solder joint inspection using eigensolder features

Abstract

Purpose The authors propose a solder joint recognition method based on eigenspace technology. Design/methodology/approach The original solder joint image is transformed into a small set of feature subspace called “eigensolder”, which is the eigenvector of the training set and can represent a solder joint well. Then, the eigensolder feature is extracted by projecting the new solder joint image into the subspace, and the Euclidean distance measure is used to classify the solder joint. Findings The experimental results show that the proposed method is superior to the traditional classification method in solder joint recognition, and it can achieve 96.43 per cent recognition rate using only 15 eigenvalue images. It is suitable for the classification with small samples. Originality/value Traditional classification method like neural network and statistical method cost long time. Here, Eigensolder method is used to extract feature. Eigensolder method is more efficient, as it uses the principal component analysis method to reduce the feature dimension of input image and only measure the distance to classify.

Automatic optical inspection system for IC solder joint based on local-to-global ensemble learning

“…To validate our proposed method, we compare it with some state-of-the-art inspection methods, which are SAM (Xie et al, 2009), SDL (Wu and Zhang, 2011), ViBe-based (Cai et al, 2015) and RPCA-based (Cai et al, 2017;Song et al, 2019;and Wu and Xu, 2018). The comparison results are shown in Table 2.…”

“…The reasons are that a single Gaussian model is not able to represent the distribution of the pixels in the image sequence and susceptible to shifts inherently in the structures of IC solder joints. Song et al (2019) and Wu and Xu (2018) both achieve large omission rates and relatively small error rates. Song et al (2019) used the binary sub-images segmented from the solder joint image for training the SVM.…”

“…This indicates that it cannot address the imbalanced problem, which will result in large omission rate. Wu and Xu(2018) used principal component analysis (PCA) to construct several eigenvectors for inspection, which implements the discrimination by means of the Euclidean distance. However, the defects of IC solder joints often occur in the small regions between the lead tips and the pads, and there are no distinct appearance differences between the qualified and unqualified samples.…”

“…However, the defects of IC solder joints often occur in the small regions between the lead tips and the pads, and there are no distinct appearance differences between the qualified and unqualified samples. Thus, Wu and Xu (2018) performs bad inspection in omission rate, as the PCA that is a global scheme for image representation cannot grasp these local and tiny appearance differences in small regions.…”

Electronic component detection based on image sample generation

“…Common defect detection methods for electronic components based on machine vision can be divided into the following categories (Xie et al, 2007), namely, feature-based methods, machine learning-based methods and deep learning-based methods that have emerged in recent years. Common featurebased methods include wavelet transform (Cho et al, 2010;Acciani et al, 2006) and principal component analysis (Wu et al, 2018;Matsushima et al, 2010). For machine learning-based methods, there are neural networks (Hao et al, 2013;Giaquinto et al, 2009), decision trees (Hongwei et al, 2011) and support vector machines (Song et al, 2019;Wu et al, 2013).…”