Title :
Contextual modeling of hand written Chinese character for recognition. I. A comparative study
Author :
Xiong, Yan ; Chan, Chorkin
Author_Institution :
Dept. of Comput. Sci., Hong Kong Univ., Hong Kong
Abstract :
A handwritten Chinese character off-line recognizer based on contextual modeling of 2D images has been constructed. Each character is modeled by a collection of regions and their contextual relations. A relaxation training algorithm of the model is investigated. In addition, a mixture-region training algorithm considering each pixel as belonging to a mixture of regions instead of any particular one is studied
Keywords :
character recognition; handwriting recognition; image recognition; 2D images; comparative study; contextual modeling; contextual relations; hand written Chinese character recognition; mixture-region training algorithm; off-line recognizer; pixel; relaxation training algorithm; Character recognition; Context modeling; Digital signal processing; Image recognition; Image segmentation; Pixel; Stochastic processes;
Conference_Titel :
Digital Signal Processing Proceedings, 1997. DSP 97., 1997 13th International Conference on
Conference_Location :
Santorini
Print_ISBN :
0-7803-4137-6
DOI :
10.1109/ICDSP.1997.628557
