An adaptive technique for handwritten Tamil character recognition
| Autor: | D.A.A.C. Ratnaweera, K. Sarveswaran |
|---|---|
| Rok vydání: | 2007 |
| Předmět: |
Artificial neural network
Intelligent character recognition Computer science Speech recognition media_common.quotation_subject Feature extraction Kohonen self organizing map language.human_language Intelligent word recognition Adaptability ComputingMethodologies_PATTERNRECOGNITION Tamil language Character recognition media_common |
| Zdroj: | 2007 International Conference on Intelligent and Advanced Systems. |
| DOI: | 10.1109/icias.2007.4658365 |
| Popis: | Handwritten character recognition is playing a vital role in many areas of modern world. Even though considerable research work has been done in handwritten character recognition, comparatively fewer efforts have been made on handwritten Tamil character recognition. This paper proposes an adaptable method for recognizing handwritten Tamil characters. The adaptability is achieved using a type of artificial neural network called Kohonen self organizing maps (KSOM). In addition, a fine-tuning method, that uses global features, is admitted to fine tune the results. A demo for proposed concept is presented here, which is developed for a subset of Tamil alphabet. The demo gives a recognition accuracy of 86%. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|