A scalable pattern recognition method

This paper put forward a scalable pattern recognition system that has a knowledge-increasable ability for solving problems of complex or large quantities of classes. Since multiple classifiers work in parallel during the training process and dynamic grouping technique is adopted on recognition, performance on both scalability and recognition rate has improved greatly. Our method can not only accelerate calculation speed and improve recognition rate but also be extended easily and freely. It can be used in pattern recognition such as face, fingerprint and character. The constructive method of the system shows the generality. Experimental results prove its reasonableness and feasibility.