An Automatic Optical Inspection Algorithm of Capacitor Based on Multi-angle Classification and Recognition

Abstract

Abstract There are many plug-in components in PCB circuit production. The AOI of these components has always been a difficult problem because of the large deviation of manual installation position of these components. In this paper, an AOI algorithm based on multi-angle classification and recognition is proposed for the plug-in polar capacitors. The algorithm combines traditional image comparison method with feature recognition method, and uses AdaBoost classifier based on Haar-like feature to recognize and classify multi-angle polar capacitors. Experimental results show that the algorithm proposed in this paper can effectively detect three kinds of defects: capacitor missing, capacitor polarity error and capacitor size mismatch.