Title :
A chain coding approach for real-time recognition of on-line handwritten characters
Author_Institution :
Dept. of Electron. Eng., Hong Kong Polytech., Kowloon, Hong Kong
Abstract :
A novel chain coding approach is proposed for real-time recognition of isolated on-line handwritten English characters. A new type of generalized chain code (GCC) is proposed for lossless encoding of the variably-spaced handwritten data points produced naturally by an electronic writing tablet due to the fixed sampling rate and the variable writing speed. Distinctive features for each character are then extracted based on the segmented cumulative differential normalized GCC values and structural descriptors. Experiments showed successful real-time recognition of isolated on-line handwritten lowercase English characters on a 486 class PC
Keywords :
character recognition; feature extraction; handwriting recognition; image coding; image sampling; image segmentation; microcomputer applications; 486 class PC; chain coding; distinctive feature extraction; electronic writing tablet; experiments; fixed sampling rate; generalized chain code; lossless encoding; lowercase English characters; online handwritten English characters; real-time recognition; segmented cumulative differential normalized GCC values; structural descriptors; variable writing speed; variably spaced handwritten data points; Character recognition; Data mining; Encoding; Feature extraction; Frequency; Handwriting recognition; Hidden Markov models; Sampling methods; Shape; Writing;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1996. ICASSP-96. Conference Proceedings., 1996 IEEE International Conference on
Conference_Location :
Atlanta, GA
Print_ISBN :
0-7803-3192-3
DOI :
10.1109/ICASSP.1996.550614
