タイ数字の筆記速度と重心の軌跡に着目したオンライン筆者認識(<特集>通信(コミュニケーション)工学特集)
スポンサーリンク
概要
- 論文の詳細を見る
We propose an on-line writer recognition method for Thai numeral based on barycenter trajectory and handwriting velocity. A barycenter of triangle composed of the origin and the two adjacent points of pen-point positions(BCTP) is used effectively to extract the feature of handwriting instead of the pen-point position. Using BCTP, the velocity of handwriting process is estimated. We incorporate the velocity of handwriting process and the BCTP trajectoiy in this paper. Writer is recognized by using dynamic programming(DP) matching. Writer recognition experiments were performed by using 4,100 data collected by 10 Thai writers for one year. As the results, the average of Type I(false rejection) error rate and Type II(false acceptance) error rate were 9.15% and 4.72%, respectively.
- 東海大学の論文
- 2003-09-30
著者
-
Matsuura Takenobu
Department Of Communications Engineering
-
Matsuura Takenobu
Department Of Communications Engineering Tokai University
-
THUMWARIN Pitak
Course of Electrical Engineering, Tokai University
-
Thumwarin Pitak
Course Of Electrical Engineering Tokai University
関連論文
- On Stochastic System Representation of Handwriting Process and Its Application to Writer Recognition
- A Seal Imprint Verification Method with Rotation Invariance(ABSTRACTS OF PROCEEDINGS OF THE SCHOOL OF INFORMATION TECHNOLOGY AND ELECTRONICS SERIES J TOKAI UNIVERISYT : 2002-2003)
- Signature Verification by FIR Filter
- A Method for Solving Stochastic Differential Equation with Random Coefficients
- タイ数字の筆記速度と重心の軌跡に着目したオンライン筆者認識(通信(コミュニケーション)工学特集)
- On-Line Signature Verification Based on Angular Direction of Pen-point Movement
- Oil-Line Writer Recognition for Thai Based on Impulse Response of FIR System Characterizing Handwriting Motion
- Writer Identification Based on the Impulse Response of Writing Motion System
- On-Line Writer Recognition for Thai Numeral Based on Barycenter Trajectory and Handwriting Velocity(PROCEEDINGS OF THE SCHOOL OF INFORMATION TECHNOLOGY AND ELECTRONICS OF TOKAI UNIVERSITY)
- Analysis and Identification of Mean Square Periodic Stochastic Systems