Automatic search for break points in printed circuit board using intelligent visual system based on Dsp algorithms

• Main Author: Syed Abu Bakar, Syed Abdul Rahman
• Format: Monograph
• Language: English
• Published: Universiti Teknologi Malaysia 2003
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/id/eprint/2930/1/72396.pdf; http://eprints.utm.my/id/eprint/2930/
• Institution: Universiti Teknologi Malaysia

Printed circuit board (PCB). Since the existing technology is going towards full digital implementation, it is envisioned that the manufacturing of PCB will ever growing. In conjunction with this development, Malaysia has taken an important step. Currently there are 37-listed PCB manufacturing companies nation-wide. As important as producing the PCB is to produce a zero-defect PCB. This is to ensure a high quality PCB that translates to reliable and quality digital end products. Initially, the bare PCBs (PCB without components attached to it) were inspected randomly using manual inspection system, which involves human operators. This technique is quite costly since it is highly error-prone due to human error. A more sophisticated way of doing the inspection is the use of in-house circuit testing (ICT) technique. This technique uses a very expensive machine that checks the conductivity of the PCB using probes. However, the limitation of this technique is it can only detect defects that are based on either shorts or open. The objective of this project thus is to provide an alternative inexpensive and comprehensive defect detection technique. The basic technique of the proposed technique is to detect the defect based on the digital image of the PCB using image processing techniques. Defect detection using image processing itself is not a new field as it has been applied in the industries. However, what makes this proposed project a novel one is the implementation of wavelet transform that has assisted in detecting as many as 14 defects that could occur on the PCB. This project has successfully implemented this wavelet-based defect detection and has successfully detected 100% defects on synthetic PCB images and more than 95% accuracy using real PCB images. This project was exhibited in the Expo Science and Technology 2002 in Kuala Lumpur and has won a bronze medal.