Comparison of performance and results in optical recognition hand written numbers using radial basis functions and memetic differential system
The problem optical recognition of handwritten numbers has been approached by different methods, obtaining satisfactory results. In this paper, we propose fuzzy systems with memetic genetic algorithms. Results from this methodology are compared with artificial neuronal networks trained using semi-su...
- Autores:
-
Montes Castañeda, Bryan
Bello Santos, Omar David
Gómez Piragauta, Oscar Manuel
Orjuela-Cañón, Alvaro David
- Tipo de recurso:
- Article of journal
- Fecha de publicación:
- 2014
- Institución:
- Universidad Antonio Nariño
- Repositorio:
- Repositorio UAN
- Idioma:
- spa
- OAI Identifier:
- oai:repositorio.uan.edu.co:123456789/3944
- Acceso en línea:
- http://revistas.uan.edu.co/index.php/ingeuan/article/view/379
http://repositorio.uan.edu.co/handle/123456789/3944
- Palabra clave:
- Rights
- openAccess
- License
- Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0)
| Summary: | The problem optical recognition of handwritten numbers has been approached by different methods, obtaining satisfactory results. In this paper, we propose fuzzy systems with memetic genetic algorithms. Results from this methodology are compared with artificial neuronal networks trained using semi-supervised learning and radial base functions (RBF). It is possible to observe that this kind of neuronal networks offer advantages regarding error rates and time-to-results of the recognition system, compared with methods based in fuzzy systems. |
|---|