A new HMM for on-line character recognition using pen-direction and pen-coordinate features
| Autor: | Seiichi Uchida, Hiroaki Sakoe, Yoshinori Katayama |
|---|---|
| Rok vydání: | 2008 |
| Předmět: |
business.industry
Computer science Transition (fiction) Speech recognition Feature extraction ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Pattern recognition Markov model ComputingMethodologies_PATTERNRECOGNITION Line segment Handwriting recognition Feature (computer vision) Line (geometry) Feature (machine learning) Artificial intelligence business Hidden Markov model ComputingMethodologies_COMPUTERGRAPHICS |
| Zdroj: | ICPR Scopus-Elsevier |
| ISSN: | 1051-4651 |
| DOI: | 10.1109/icpr.2008.4761449 |
| Popis: | A new hidden Markov model (HMM) is proposed for on-line character recognition using two typical features, pen-direction feature and pen-coordinate feature. These two features are quite different in their stationarity; pen-direction feature is stationary within every line segment of a stroke whereas pen-coordinate feature is not. In the proposed HMM, these contrasting features are used in a separative and selective way. Specifically speaking, pen-direction feature is out putted repeatedly at intra-state transition whereas pen-coordinate feature is out putted once at inter-state transition. The superiority of the proposed HMM over the conventional HMMs was shown through single-stroke and multi-stroke character recognition experiments. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|