Wavelet-based printed circuit board inspection algorithm

Article type: Research Article

Authors: Ibrahim, Zuwairie^a | Rahman Al-Attas, Syed Abdul^{b; *}

Affiliations: [a] Institute of Applied DNA Computing, Meiji University, 1-1-1 Higashi-mita, Tama-ku, Kawasaki, Kanagawa-ken, 214-8571, Japan. Tel.: +81 44 934 7377 Fax: +81 44 934 7909; E-mail: zuwairie@isc.meiji.ac.jp | [b] Department of Microelectronics and Computer Engineering, Faculty of Electrical Engineering, Universiti Teknologi Malaysia, 81310 Skudai, Johor Darul Takzim, Malaysia

Correspondence: [*] Corresponding author. Tel.: +6 07 553 5238; Fax: +6 07 556 6272; E-mail: syed@fke.utm.my.

Abstract: One of the backbones in electronic manufacturing industry is the printed circuit board (PCB) manufacturing. Due to the human limited resources and speed requirements, manual inspection is ineffective to inspect every printed circuit board. Hence, this paper presents an efficient algorithm for an automated visual PCB inspection system that detects and locates any defect found on PCBs. The detection mechanism utilizes the wavelet-based image difference algorithm. In order to locate the defective areas, the coarse resolution defect localization algorithm is proposed which is applied to the coarse resolution of the differenced image. This algorithm will then map the defective areas found in the coarse differenced image to the fine resolution of the tested image. Based on experimental results, the proposed method has achieved a reduction up to 82.11% of the computation time in comparison to the traditional image difference operation without sacrificing the accuracy of the defect detection . This tremendous amount of saving has been made possible by the use of Haar wavelet transform. Consequently, such amount of reduction will benefit the industries as the automatic inspection for each PCB can now be realized at high speed.

DOI: 10.3233/ICA-2005-12206

Journal: Integrated Computer-Aided Engineering, vol. 12, no. 2, pp. 201-213, 2005