Visual inspection scheme for use in optical solder joint inspection system

An optical solder joint inspection system(0SIIS) has been developed for the automatic visual inspection of soldered parts on the printed circuit boards. Its advantages over existing techniques include the detection of 3-0 shape of specular objects with high reliable and high speed. In this paper, we will propose a solder joint inspection scheme for a prototype of the OUIS. The inspection scheme is composed of two steps: feature extraction and classification. In the feature extraction step, by scanning a laser beam over the area of solder joints the system obtain's two orientation curves representing the quality of solderin,g condition, and then nine features are extractedfiom the curves. In the classijication step, a neural network classifies the solder joint according to the applicatio,n requirements by using the features. Experiments were performed for SOP(smal1 outline package)s and QFP(quaid flat package)s in insufficient, normal and excess soldering condition. Based upon observation of the experimental results, the proposed inspection scheme shows excellent consistency with visual inspection and a good accuracy o f classification performance of 94.2%.