ChangeChip: A Reference-Based Unsupervised Change Detection for PCB Defect Detection

被引:18
作者
Fridman, Yehonatan [1 ]
Rusanovsky, Matan [1 ]
Oren, Gal [2 ]
机构
[1] Ben Gurion Univ Negev, Israel Atom Energy Commiss, Beer Sheva, Israel
[2] Technion Israel Inst Technol, Nucl Res Ctr Negev, Beer Sheva, Israel
来源
PROCEEDINGS OF THE 2021 IEEE INTERNATIONAL CONFERENCE ON PHYSICAL ASSURANCE AND INSPECTION ON ELECTRONICS (PAINE) | 2021年
关键词
Change Detection; Unsupervised Machine Learning; PCA-Kmeans; PCBs; SMT Quality Control;
D O I
10.1109/PAINE54418.2021.9707699
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The usage of electronic devices increases, and becomes predominant in most aspects of life. Surface Mount Technology (SMT) is the most common industrial method for manufacturing electric devices in which electrical components are mounted directly onto the surface of a Printed Circuit Board (PCB). Although the expansion of electronic devices affects our lives in a productive way, failures or defects in the manufacturing procedure of those devices might also be counterproductive and even harmful in some cases. It is therefore desired and sometimes crucial to ensure zero-defect quality in electronic devices and their production. While traditional Image Processing (IP) techniques are not sufficient to produce a complete solution, other promising methods like Deep Learning (DL) might also be challenging for PCB inspection, mainly because such methods require big adequate datasets which are missing, not available or not updated in the rapidly growing field of PCBs. Thus, PCB inspection is conventionally performed manually by human experts. Unsupervised Learning (UL) methods may potentially be suitable for PCB inspection, having learning capabilities on the one hand, while not relying on large datasets on the other. In this paper, we introduce ChangeChip, an automated and integrated change detection system for defect detection in PCBs, from soldering defects to missing or misaligned electronic elements, based on Computer Vision (CV) and UL. We achieve good quality defect detection by applying an unsupervised change detection between images of a golden PCB (reference) and the inspected PCB under various setting. In this work, we also present CD-PCB, a synthesized labeled dataset of 20 pairs of PCB images for evaluation of defect detection algorithms.
引用
收藏
页数:8
相关论文
共 27 条
[1]  
Acciani G, 2005, IEEE IJCNN, P2324
[2]  
Botero U.J., 2020, HARDWARE TRUST ASSUR
[3]  
Botero Ulbert J, 2020, 2020 IEEE PHYS ASSUR
[4]   Unsupervised Change Detection in Satellite Images Using Principal Component Analysis and k-Means Clustering [J].
Celik, Turgay .
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2009, 6 (04) :772-776
[5]   Exact histogram specification [J].
Coltuc, D ;
Bolon, P ;
Chassery, JM .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2006, 15 (05) :1143-1152
[6]  
Derpanis K.G., 2010, Image Rochester NY, V4, P2
[7]  
Desai H, 1997, PROCEEDINGS: ELECTRICAL INSULATION CONFERENCE AND ELECTRICAL MANUFACTURING & COIL WINDING CONFERENCE, P677
[8]   A real-time approach for automatic defect detection from PCBs based on SURF features and morphological operations [J].
Hassanin, Abdel-Aziz I. M. ;
Abd El-Samie, Fathi E. ;
El Banby, Ghada M. .
MULTIMEDIA TOOLS AND APPLICATIONS, 2019, 78 (24) :34437-34457
[9]  
Huang W., 2019, ARXIV PREPRINT ARXIV
[10]  
Huihui Wu, 2010, 2010 IEEE International Conference on Mechatronics and Automation (ICMA), P1789, DOI 10.1109/ICMA.2010.5588029