A particle swarm optimization approach for components placement inspection on printed circuit boards

The importance of the inspection has been magnified by the requirements of the modern manufacturing environment. In electronics mass-production manufacturing facilities, especially in the printed circuit board (PCB) industry, 100% quality assurance of all work-in-process and finished goods is required in order to reduce the scrap costs and re-work rate. One of the challenges for PCB inspection is in the use of a surface mount device (SMD) placement check. Missing, misaligned or wrongly rotated components are the critical causes of defects. To prevent the PCB from having these defects, inspection must be done before the solder reflow process commences, otherwise, everything will be too late. The research reported in this paper concentrates on automatic object searching techniques, in a grey-scale captured image, for locating multiple components on a PCB. The presented approach includes the normalized cross correlation (NCC) based multi-template matching (MTM) method. The searching process has been carried out by using the proposed accelerated species based particle swarm optimization (ASPSO) method and the genetic algorithm (GA) approach as a reference. The experimental results of the ASPSO-based MTM approaches are reported.