Character recognition in cursive handwriting with the mechanism of selective attention
| Autor: | Taro Imagawa, Kunihiko Fukushima |
|---|---|
| Rok vydání: | 1993 |
| Předmět: |
Artificial neural network
Computer science Intelligent character recognition Mechanism (biology) Speech recognition Construct (python library) Theoretical Computer Science Task (project management) Character (mathematics) Computational Theory and Mathematics Hardware and Architecture Pattern recognition (psychology) Selective attention Information Systems |
| Zdroj: | Systems and Computers in Japan. 24:89-97 |
| ISSN: | 1520-684X 0882-1666 |
| Popis: | The recognition of connected characters in cursive handwriting is a difficult task with ordinary pattern-matching techniques, since the shape of the individual character is affected by its preceding and succeeding characters. One of the authors has proposed a neural network model called the selective attention model, which has the ability to recognize and extract individual patterns from a composite of a number of elementary patterns. However, when a large number of patterns are presented concurrently, this model doe not always work correctly. In this paper, we extend the idea of this model and construct a new system, which has the ability to recognize the connected characters in cursive handwriting. it has been verified by computer simulation that this system can be used to correctly recognize connected characters. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Plný text