IC Assembly Product Inspection Using Image Processing Techniques

Abstract:

In this study, we present an image identification and measurement system for examining and testing the packaged semiconductor specifications. Most semiconductor processes rely on measurement gauges or inspectors to examine the finished product specifications visually. Measurement gauges are costly; inspectors have to be trained professionally and their performance depends on the maturity of their skills, which requires enormous costs, time, and efforts. Therefore, an automatic identification and measurement system will not only reduce costs, but minimize the complexity of measurement, thereby upgrading the effectiveness of manpower significantly. Experiments were conducted using 1657 BMP images with 640 * 480 pixels taken with a semiconductor machine. These images are snapshots randomly taken from of a variety of the 7.2 cm * 4.8 cm Substrate circuit boards each includes 180 pieces of 30 * 6 IC packaged products. Promising results were derived where 1492 out of 1657 product images were successfully detected and measured. In addition to the 90.04% success rate for inspection, the process time is reduced significantly to about 1/6 as compared to human professionals.

• Periodical: Applied Mechanics and Materials (Volume 145)
• Pages: 209-213
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.145.209
• Online since: December 2011
• Authors: Hsi Chieh Lee, Shao Hsuan Chang
• Keywords: Automated Optical Inspection, Hough Transform Algorithm, Image Processing, Morphological Operator, Sobel Operator

References

[1] S. H. Oguz and L. Onural, An automated system for design-rule-based visual inspection of printed circuit boards, (1991). Proceedings. IEEE International Conference on Robotics and Automation, p.2696 – 2701.

DOI: 10.1109/robot.1991.132038

[2] M. Moganti and F. Ercal, Segmentation of printed circuit board images, Computer Vision and Image Understanding, Vol. 70, No. 1 (1998), 74–86.

DOI: 10.1006/cviu.1998.0594

[3] M. Sasai., Printed Circuit Board Inspection for Flexible Manufacture, 5th World Conference on NonDestructive Testing Proceedings, Italy (2000).

[4] S. Mashohor and J. R. Evans, and T. Arslan, Genetic Algorithms Based to Printed Circuit Board (PCB) Inspection System, IEEE International Symposium Consumer Electronics (2004).

DOI: 10.1109/isce.2004.1376000

[5] D. -M. Tsai and R. -H. Yang, An eigenvalue-based similarity measure and its application in defect detection, Image and Vision Computing, Vol. 23, No. 12 (2005), pp.1094-1101.

DOI: 10.1016/j.imavis.2005.07.014

[6] Greenberg, Method for printed circuit board inspection, United States Patent 6, 990, 227 (2006).

[7] H. Y. T. Ngan and G. K. H. Pang, Novel method for patterned fabric inspection using Bollinger bands, Opt. Eng. 45, 087202 , Sep 06 (2006).

DOI: 10.1117/1.2345189

[8] M. Moganti, F. Ercal, F., C. H. Dagli, and S. Tsunekawa, Automatic PCB inspection algorithms: A survey, Computer Vision Image Understanding, Vol. 63, No. 2 (1996), pp.287-313.

DOI: 10.1006/cviu.1996.0020