Abstract
A stochastic model of stroke order variation is proposed and applied to the stroke-order free on-line Kanji character recognition. The proposed model is a hidden Markov model (HMM) with a special topology to represent all stroke order variations. A sequence of state transitions from the initial state to the final state of the model represents one stroke order and provides a probability of the stroke order. The distribution of the stroke order probability can be trained automatically by using an EM algorithm from a training set of on-line character patterns. Experimental results on large-scale test patterns showed that the proposed model could represent actual stroke order variations appropriately and improve recognition accuracy by penalizing incorrect stroke orders.
| Original language | English |
|---|---|
| Title of host publication | ICDAR2009 - 10th International Conference on Document Analysis and Recognition |
| Pages | 803-807 |
| Number of pages | 5 |
| DOIs | |
| Publication status | Published - 2009 |
| Event | ICDAR2009 - 10th International Conference on Document Analysis and Recognition - Barcelona, Spain Duration: Jul 26 2009 → Jul 29 2009 |
Other
| Other | ICDAR2009 - 10th International Conference on Document Analysis and Recognition |
|---|---|
| Country | Spain |
| City | Barcelona |
| Period | 7/26/09 → 7/29/09 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Computer Vision and Pattern Recognition
Cite this
Stochastic model of stroke order variation. / Katayama, Yoshinori; Uchida, Seiichi; Sakoe, Hiroaki.
ICDAR2009 - 10th International Conference on Document Analysis and Recognition. 2009. p. 803-807 5277515.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Stochastic model of stroke order variation
AU - Katayama, Yoshinori
AU - Uchida, Seiichi
AU - Sakoe, Hiroaki
PY - 2009
Y1 - 2009
N2 - A stochastic model of stroke order variation is proposed and applied to the stroke-order free on-line Kanji character recognition. The proposed model is a hidden Markov model (HMM) with a special topology to represent all stroke order variations. A sequence of state transitions from the initial state to the final state of the model represents one stroke order and provides a probability of the stroke order. The distribution of the stroke order probability can be trained automatically by using an EM algorithm from a training set of on-line character patterns. Experimental results on large-scale test patterns showed that the proposed model could represent actual stroke order variations appropriately and improve recognition accuracy by penalizing incorrect stroke orders.
AB - A stochastic model of stroke order variation is proposed and applied to the stroke-order free on-line Kanji character recognition. The proposed model is a hidden Markov model (HMM) with a special topology to represent all stroke order variations. A sequence of state transitions from the initial state to the final state of the model represents one stroke order and provides a probability of the stroke order. The distribution of the stroke order probability can be trained automatically by using an EM algorithm from a training set of on-line character patterns. Experimental results on large-scale test patterns showed that the proposed model could represent actual stroke order variations appropriately and improve recognition accuracy by penalizing incorrect stroke orders.
UR - http://www.scopus.com/inward/record.url?scp=71249145228&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=71249145228&partnerID=8YFLogxK
U2 - 10.1109/ICDAR.2009.146
DO - 10.1109/ICDAR.2009.146
M3 - Conference contribution
AN - SCOPUS:71249145228
SN - 9780769537252
SP - 803
EP - 807
BT - ICDAR2009 - 10th International Conference on Document Analysis and Recognition
ER -