Optical character recognition using transfer learning decision forests

In this paper, we present a novel method for transfer learning which uses decision forests, and we apply it to recognize characters. We introduce two extensions into the decision forest framework in order to transfer knowledge from the source tasks to a given target task. We show that both of them a...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autor principal: Mejail, Marta Estela
Publicado: 2014
Materias:
decision forests
OCR
transfer learning
Character recognition
Image processing
Learning algorithms
Optical character recognition
Optical data processing
Decision forest
State of the art
Transfer learning
Forestry
Acceso en línea:https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_97814799_v_n_p4309_Goussies
http://hdl.handle.net/20.500.12110/paper_97814799_v_n_p4309_Goussies
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
id paper:paper_97814799_v_n_p4309_Goussies
record_format dspace
spelling paper:paper_97814799_v_n_p4309_Goussies2023-06-08T16:37:47Z Optical character recognition using transfer learning decision forests Mejail, Marta Estela decision forests OCR transfer learning Character recognition Image processing Learning algorithms Optical character recognition Optical data processing Decision forest State of the art Transfer learning Forestry In this paper, we present a novel method for transfer learning which uses decision forests, and we apply it to recognize characters. We introduce two extensions into the decision forest framework in order to transfer knowledge from the source tasks to a given target task. We show that both of them are important to achieve higher recognition rates. Our experiments demonstrate improvements over traditional decision forests in the MNIST dataset. They also compare favorably against other state-of-the-art classifiers. © 2014 IEEE. Fil:Mejail, M.E. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina. 2014 https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_97814799_v_n_p4309_Goussies http://hdl.handle.net/20.500.12110/paper_97814799_v_n_p4309_Goussies
institution Universidad de Buenos Aires
institution_str I-28
repository_str R-134
collection Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA)
topic decision forests
OCR
transfer learning
Character recognition
Image processing
Learning algorithms
Optical character recognition
Optical data processing
Decision forest
State of the art
Transfer learning
Forestry
spellingShingle decision forests
OCR
transfer learning
Character recognition
Image processing
Learning algorithms
Optical character recognition
Optical data processing
Decision forest
State of the art
Transfer learning
Forestry
Mejail, Marta Estela
Optical character recognition using transfer learning decision forests
topic_facet decision forests
OCR
transfer learning
Character recognition
Image processing
Learning algorithms
Optical character recognition
Optical data processing
Decision forest
State of the art
Transfer learning
Forestry
description In this paper, we present a novel method for transfer learning which uses decision forests, and we apply it to recognize characters. We introduce two extensions into the decision forest framework in order to transfer knowledge from the source tasks to a given target task. We show that both of them are important to achieve higher recognition rates. Our experiments demonstrate improvements over traditional decision forests in the MNIST dataset. They also compare favorably against other state-of-the-art classifiers. © 2014 IEEE.
author Mejail, Marta Estela
author_facet Mejail, Marta Estela
author_sort Mejail, Marta Estela
title Optical character recognition using transfer learning decision forests
title_short Optical character recognition using transfer learning decision forests
title_full Optical character recognition using transfer learning decision forests
title_fullStr Optical character recognition using transfer learning decision forests
title_full_unstemmed Optical character recognition using transfer learning decision forests
title_sort optical character recognition using transfer learning decision forests
publishDate 2014
url https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_97814799_v_n_p4309_Goussies
http://hdl.handle.net/20.500.12110/paper_97814799_v_n_p4309_Goussies
work_keys_str_mv AT mejailmartaestela opticalcharacterrecognitionusingtransferlearningdecisionforests
_version_ 1768544341522907136

Ejemplares similares