Detection of ambiguous patterns in a SOM based recognition system: Application to handwritten numeral classification

This work presents a system for pattern recognition that combines a self-organising unsupervised technique (via a Kohonen-type SOM) with a bayesian strategy in order to classify input patterns from a given probability distribution and, at the same time, detect ambiguous cases and explain answers. We...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Seijas, L.M., Segura, E.C., Universitat Bielefeld; SFB 673
Formato:	CONF
Materias:	Bayesian statistics Pattern recognition Self-organising maps Ambiguous patterns Concordia University Handwritten digit Handwritten numeral Recognition systems Self-organising Unsupervised techniques Conformal mapping Probability distributions Self organizing maps Pattern recognition systems
Acceso en línea:	http://hdl.handle.net/20.500.12110/paper_NIS12828_v_n_p_Seijas
Aporte de:	Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires

id	todo:paper_NIS12828_v_n_p_Seijas
record_format	dspace
spelling	todo:paper_NIS12828_v_n_p_Seijas2023-10-03T16:45:50Z Detection of ambiguous patterns in a SOM based recognition system: Application to handwritten numeral classification Seijas, L.M. Segura, E.C. Universitat Bielefeld; SFB 673 Bayesian statistics Pattern recognition Self-organising maps Ambiguous patterns Bayesian statistics Concordia University Handwritten digit Handwritten numeral Recognition systems Self-organising Unsupervised techniques Conformal mapping Pattern recognition Probability distributions Self organizing maps Pattern recognition systems This work presents a system for pattern recognition that combines a self-organising unsupervised technique (via a Kohonen-type SOM) with a bayesian strategy in order to classify input patterns from a given probability distribution and, at the same time, detect ambiguous cases and explain answers. We apply the system to the recognition of handwritten digits. This proposal is intended as an improvement of a model previously introduced by our group, consisting basically of a hybrid unsupervised, self-organising model, followed by a supervised stage. Experiments were carried out on the handwritten digit database of Concordia University, which is generally accepted as one of the standards in most of the literature in the field. Fil:Seijas, L.M. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina. Fil:Segura, E.C. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina. CONF info:eu-repo/semantics/openAccess http://creativecommons.org/licenses/by/2.5/ar http://hdl.handle.net/20.500.12110/paper_NIS12828_v_n_p_Seijas
institution	Universidad de Buenos Aires
institution_str	I-28
repository_str	R-134
collection	Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA)
topic	Bayesian statistics Pattern recognition Self-organising maps Ambiguous patterns Bayesian statistics Concordia University Handwritten digit Handwritten numeral Recognition systems Self-organising Unsupervised techniques Conformal mapping Pattern recognition Probability distributions Self organizing maps Pattern recognition systems
spellingShingle	Bayesian statistics Pattern recognition Self-organising maps Ambiguous patterns Bayesian statistics Concordia University Handwritten digit Handwritten numeral Recognition systems Self-organising Unsupervised techniques Conformal mapping Pattern recognition Probability distributions Self organizing maps Pattern recognition systems Seijas, L.M. Segura, E.C. Universitat Bielefeld; SFB 673 Detection of ambiguous patterns in a SOM based recognition system: Application to handwritten numeral classification
topic_facet	Bayesian statistics Pattern recognition Self-organising maps Ambiguous patterns Bayesian statistics Concordia University Handwritten digit Handwritten numeral Recognition systems Self-organising Unsupervised techniques Conformal mapping Pattern recognition Probability distributions Self organizing maps Pattern recognition systems
description	This work presents a system for pattern recognition that combines a self-organising unsupervised technique (via a Kohonen-type SOM) with a bayesian strategy in order to classify input patterns from a given probability distribution and, at the same time, detect ambiguous cases and explain answers. We apply the system to the recognition of handwritten digits. This proposal is intended as an improvement of a model previously introduced by our group, consisting basically of a hybrid unsupervised, self-organising model, followed by a supervised stage. Experiments were carried out on the handwritten digit database of Concordia University, which is generally accepted as one of the standards in most of the literature in the field.
format	CONF
author	Seijas, L.M. Segura, E.C. Universitat Bielefeld; SFB 673
author_facet	Seijas, L.M. Segura, E.C. Universitat Bielefeld; SFB 673
author_sort	Seijas, L.M.
title	Detection of ambiguous patterns in a SOM based recognition system: Application to handwritten numeral classification
title_short	Detection of ambiguous patterns in a SOM based recognition system: Application to handwritten numeral classification
title_full	Detection of ambiguous patterns in a SOM based recognition system: Application to handwritten numeral classification
title_fullStr	Detection of ambiguous patterns in a SOM based recognition system: Application to handwritten numeral classification
title_full_unstemmed	Detection of ambiguous patterns in a SOM based recognition system: Application to handwritten numeral classification
title_sort	detection of ambiguous patterns in a som based recognition system: application to handwritten numeral classification
url	http://hdl.handle.net/20.500.12110/paper_NIS12828_v_n_p_Seijas
work_keys_str_mv	AT seijaslm detectionofambiguouspatternsinasombasedrecognitionsystemapplicationtohandwrittennumeralclassification AT seguraec detectionofambiguouspatternsinasombasedrecognitionsystemapplicationtohandwrittennumeralclassification AT universitatbielefeldsfb673 detectionofambiguouspatternsinasombasedrecognitionsystemapplicationtohandwrittennumeralclassification
_version_	1807316097951596544

Detection of ambiguous patterns in a SOM based recognition system: Application to handwritten numeral classification

Ejemplares similares