Feature transformation with generalized learning vector quantization for hand-written Chinese character recognition

In this paper, the generalized learning vector quantization (GLVQ) algorith m is applied to design a handwritten Chinese character recognition system. The system proposed herein consists of two modules, feature transformation and recognizer. The feature transformation module is designed to extract di scriminative features to enhance the recognition performance. The initial f eature transformation matrix is obtained by using Fisher's linear discrimin ant (FLD) function. A template matching with minimum distance criterion rec ognizer is used and each character is represented by one reference template . These reference templates and the elements of the feature transformation matrix are trained by using the generalized learning vector quantization al gorithm. In the experiments, 540100 (5401 x 100) hand-written Chinese chara cter samples are used to build the recognition system and the other 540100( 5401 x 100) samples are used to do the open test. A good performance of 92. 18% accuracy is achieved by proposed system.