An automated inspection system for printed circuit board is preferred to get the technological advances in PCB’s design and manufacturing, to reduce the inspection duration and manufacturing cost because human inspectors’ decisions areinefficient, tedious and costly. It is very useful in PCB manufacturing process whereby the manufactures can detect the defect. This is because the defect is detected in the early stage on the bare PCB rather than in later stage after the lamination of the PCB. This will save the cost of manufacturing process by
reducing the amount of PCB being thrown away. The PCB inspection system is then improved by incorporating a geometrical image registration, minimum thresholding technique and median filtering in order to solve alignment and uneven illumination problem. Finally, defect classification operation is employed in order to identify the source for four types of defects namely, missing hole, pin hole, short-circuit, and opencircuit.