Automated visual inspection of surface mounted devices (SMD) requires the correct classification of an image as either “component present” or “component absent”. The inspection system must allow the classification to be fast and reliable, while also assuring that the training of the classifier is simple and not time consuming.

Problem Description:

The traditional sequential approach to classifying images, while uncomplicated to implement, presents some disadvantages, including an increase in false alarm (type I) errors and the need for a time-consuming training phase. The method presented in this paper seeks to reduce classification errors by using a vector approach for the classification of components. The classification vector is based on three simple features obtained from the image being analyzed: energy, one-dimensional correlation, and diffusion. The use of simple features gives the resulting classification characteristics that make it easy to train and implement.