Computer-aided system for defect inspection in the PCB manufacturing process

This paper presents a visual inspection system aimed at the automatic detection and classification of bare-PCB manufacturing errors. The interest of this CAE system lies in a twofold approach. On the one hand, we propose a modification of the subtraction method based on reference images that allows higher performance in the process of defect detection. On the other hand, this method is combined with a particle classification algorithm based on two measures of light intensity. As a result of this strategy, a machine vision application has been implemented to assist people in etching, inspection and verification tasks of PCBs.