Competitiveness improvement in multi-classifier systems by data equalization

One way of obtaining better recognition result is to have multi-classifier systems. The problem of multi-classifier systems is the lack of competitiveness which degrade the performance of the final output. A data equalization method is proposed to increase the competitiveness of the output activation values of the individual classifier in a multi-classifier system. Data equalization helps to redistribute the output activation values such that the average difference of the output activation values is smaller. The experimental results shows that the proposed method improves the accuracy rate of a combined classifier (CC) which aggregates the output activation values of the front-end classifiers.