印刷電路內層板瑕疵檢測之研究

A Study of Defect Inspection on PCB Inner Layer

吳建宏(Chine-Hung Wu)

指導教授 ： 饒忻

中原大學/工學院/工業工程研究所/碩士(2001年)

https://doi.org/10.6840/cycu200100399

摘要

機器視覺在工業自動化生產的過程中扮演著一個重要的角色，在印刷電路板的製程中利用機器視覺檢測出電路板瑕疵是現今最典型的應用實例。但大多數的電路板視覺檢測設備多由國外進口且頗為昂貴，因此本研究主要針對內層印刷電路板作瑕疵，提出檢測方法並開發出檢測系統。
內層電路板的瑕疵共有14種，而這些瑕疵依據製程的不同又可區分為蝕刻製程瑕疵與鑽孔製程瑕疵兩類。本研究提出瑕疵邊緣偵測與電路結構掃描的方式可成功的檢測出蝕刻製程瑕疵，而特徵比對的方式則適用於鑽孔瑕疵的檢測。本研究將所提出的檢測法發展成一套電路板瑕疵檢測系統，並用實際的樣本驗證該系統的瑕疵檢測功能。
影像相減法為本研究所提出的瑕疵邊緣偵測法的前置處理作業，而一個好的二值化影像是影像相減法所不可獲缺的重要因素，本研究參照模糊理論所提出的影像處理方法，發展出一套語意式的判斷法則來提高二值化影像的品質。經實驗後發現本研究所提出的方法確實可以消除雜訊與環境的干擾並保留瑕疵的影像資料，有效地降低瑕疵檢測的誤判率。

關鍵字

印刷電路內層板 ； 瑕疵檢測 ； 模糊語意式方法

並列摘要

Machine vision plays a major role in automatic manufacturing industry. Using machine vision to detect PCB defect is one of most important applications. Unfortunately, most pieces of vision equipment are abroad and expensive, which is the major motivation of this research. This study develops a PCB inspection system to recognize PCB inner layer defects.
There are 14 defect types in PCB inner manufacturing; these defects can be divided into 2 groups according to different processes. This research proposes a defect outer tracing method and a pattern skeleton scanning method that can successfully detect defects after the etching process. In addition, using the pattern feature matching method, defects in the drilling process can be detected. The real PCB samples have been tested in the inspection system, and all defects on the sample image can be detected and recognized.
A fine binary image is a really important ingredient for the image subtraction operation, so this research proposes a fuzzy linguistic synthesis method for assuring the image binary operation. The method has an excellent performance to remove noise but preserve defect information. Therefore, false defects occurring at the PCB inspection system can be reduced effectively.

並列關鍵字

Fuzzy linguistic method ； Defect inspection ； PCB inner layer

參考文獻

[1] Madhav Moganti and Fikret Ercal, “Automatic PCB inspection Algorithms: A Survey”, Computer vision and image understanding, Vol.63, No.2, March, pp.287-313, 1996.

[3] Ito, M.; Nikaido, Y., “Recognition of pattern defects of printed circuit board using topological information”, Electronics Manufacturing Technology Symposium, 1991., Eleventh IEEE/CHMT International , 1991 , pp.202 –206

[4] Wen-Yen Wu, Mao-Jiun J. Wang, Chih-Ming Liu, “Automated inspection of printed circuit boards through machine vision”, Computers in Industry , 1996, Vol. 28, pp. 103-111.

[5] Ja H. Koo, “A structural matching for two dimensional visual pattern inspection”, Systems, Man, and Cybernetics, 1998 IEEE International Conference on, 1998, Vol.5, pp.4429-4434.

[6] Borba, J.F.; Facon, J., “A printed circuit board automated inspection system”, Circuits and Systems, 1995, Proceedings, Proceedings of the 38th Midwest Symposium on, Vol.1, 1996, pp.69-72.