Discrimination of True Defect and Indefinite Defect with Visual Inspection Using SVM
This paper proposes a new approach to discriminate the true defect and the indefinite defect with visual distinction of the electronic board. Some classification approaches have been proposed for the limited kinds of defects and there may be some incorrect recognitions for the defect which is difficult with the visual distinction. This paper proposes the approach to reduce the incorrect recognition ratio for the defects with difficult discrimination using the margin of SVM. Real electronic board image data are tested and evaluated with the proposed approach.
KeywordsFeature Selection Reference Image Defect Region Margin Region Defect Class
Unable to display preview. Download preview PDF.
- 1.Roh, B., et al.: A Neural Network Approach to Defect Classification on Printed Circuit Boards. Journal of the Japan Society of Precision Engineering 67(10), 1621–1626 (2001) (in Japanese)Google Scholar
- 2.Tanaka, T., Hotta, S., Iga, T., Nakamura, T.: Automatic Image Filter Creation System: To Use for a Defect Classification System. IEICE Technical Report 106(448), 195–198 (2007)Google Scholar
- 4.Kondo, K., et al.: Defect Classification Using Random Feature Selection and Bagging. The Journal of the Institute of Image Electronics Engineers of Japan 38(1), 9–15 (2009) (in Japanese)Google Scholar