TY - GEN
T1 - On-line handwritten cursive character recognition system
AU - Toscano-Medina, Karina
AU - Nakano, Mariko
AU - Perez-Meana, Hector
AU - Yasuhara, Makoto
PY - 2005
Y1 - 2005
N2 - During the last two decades have been proposed many handwritten character recognition systems, however until now there are still many limitations, especially for the cursive handwritten characters. In this paper a new algorithm for cursive handwritten characters recognition based on the Spline functions is proposed, in which the inverse process of the handwritten character construction task will be used to recognize the character. From the samples got by using a digitizer board, the sequence of the most significant points (optimal knots) of the handwriting character will be obtained, and then the natural Spline function (Slalom method) and the steepest descent method will be used to interpolate and approximate the character shape. Using a training set consisting of the sequence of optimal knots, each character model will be constructed. Finally the unknown input character will be, compared with each model of all characters to get the similarity scores. The character model with higher similarity score will be considered as the recognized character of the input data. In me recognition stage, two-steps classification is realized detail analysis for some groups of similar characters. The global recognition rate of the proposed system is 94.5%.
AB - During the last two decades have been proposed many handwritten character recognition systems, however until now there are still many limitations, especially for the cursive handwritten characters. In this paper a new algorithm for cursive handwritten characters recognition based on the Spline functions is proposed, in which the inverse process of the handwritten character construction task will be used to recognize the character. From the samples got by using a digitizer board, the sequence of the most significant points (optimal knots) of the handwriting character will be obtained, and then the natural Spline function (Slalom method) and the steepest descent method will be used to interpolate and approximate the character shape. Using a training set consisting of the sequence of optimal knots, each character model will be constructed. Finally the unknown input character will be, compared with each model of all characters to get the similarity scores. The character model with higher similarity score will be considered as the recognized character of the input data. In me recognition stage, two-steps classification is realized detail analysis for some groups of similar characters. The global recognition rate of the proposed system is 94.5%.
KW - Handwritten Character Recognition
KW - Minimum Description Length
KW - On-line Recognition
KW - Salalom Method
KW - Spline Function
KW - Steepest Descent
UR - http://www.scopus.com/inward/record.url?scp=84867350588&partnerID=8YFLogxK
M3 - Contribución a la conferencia
AN - SCOPUS:84867350588
SN - 9806560590
SN - 9789806560598
T3 - WMSCI 2005 - The 9th World Multi-Conference on Systemics, Cybernetics and Informatics, Proceedings
SP - 141
EP - 145
BT - WMSCI 2005 - The 9th World Multi-Conference on Systemics, Cybernetics and Informatics, Proceedings
T2 - 9th World Multi-Conference on Systemics, Cybernetics and Informatics, WMSCI 2005
Y2 - 10 July 2005 through 13 July 2005
ER -