Efficient Learning for Test Feature Classifier by Overlap Index List

Abstract

SUMMARYThis paper presents a novel low cost learning algorithm for a test feature classifier by use of an overlap index list (OIL). In general, classifiers need a large amount of training data to achieve the high performance, which results in long computation times. The proposed algorithm using OIL can maintain search and check elemental combinatorial features from lower dimensions up to higher ones. Classification problems in real industrial inspection lines have been solved by the proposed algorithm, and large reductions in computation time have been obtained.