An AOI algorithm for PCB based on feature extraction

With the development of the micro-electronic industry, electronical components assembled on the printed circuit board (PCB) become more and more microsize and its mounted density is increasing. It is out of date to depend on manual inspection to assure the joints quality. Instead, automated optical inspection (AOI) for solder joints based on the machine vision has become more and more important. In this paper, based on the image acquired from a 3-CCD color camera and a 3-color hemispherical LED arrays light resource (red, green, and blue), the place, shape, and logical features of solder joints of chip components are extracted. The place features are related to solder joint place. As for shape features, we divide solder joint into several regions and extract its shape features by its color, occupancy ratio of area, center of gravity and continuous pixels. The logical features come from their close relationships of different regions of shape features, color distributing and place features. On the basis of the features, an AOI algorithm is developed. The defects of lacking solder, surplus solder, no solder, pseudo joints, wrong component, damaged component, component absent, shift, tomb stone, wrong polarity, etc. can be identified properly by using the proposed algorithm. Finally, some experiment results are presented to show the validity of the algorithm.