OPTIMAL-DESIGN OF REFERENCE MODELS FOR LARGE-SET HANDWRITTEN CHARACTER-RECOGNITION

For the recognition of large-set handwritten characters, classificatio n methods based on pattern matching have been commonly used, and good reference models play an important role in achieving high performance in these methods. Learning Vector Quantization (LVQ) has been intensiv ely studied to generate good reference models in speech recognition si nce 1986. However, the design of reference models based on LVQ has sev eral drawbacks for the recognition of large-set handwritten characters . In this paper, to cope with these, we propose a new method for the o ptimal design of reference models using Simulated Annealing combined w ith an improved LVQ3 for the recognition of large-set handwritten char acters. Experimental results reveal that the proposed method is superi or to the conventional method based on averaging or other LVQ-based me thods.