Handwritten digit recognition using statistical and rule-based decision fusion

In this paper, the cooperation of two feature families for handwritten digit recognition using SVM (support vector machine) classifiers is examined. We investigate the advantages and weaknesses of various decision fusion schemes using statistical and rule-based reasoning. The obtained results show that it is difficult to exceed the recognition rate of a single classifier applied straightforwardly on both feature families as one set by rule based reasoning applied on the individual classifier decisions. However, the rule-based cooperation schemes enable an easy and efficient implementation of various rejection criteria. On the other hand, the statistical cooperation schemes offer a better possibility for fine tuning of the recognition versus the reliability tradeoff, which leads to recognition systems with high reliability that also keep high recognition rates.