A new approach to segmentation of remote sensing images with hidden markov models

In this work, we present a new segmentation algorithm for remote sensing images based on two-dimensional Hidden Markov Models (2D-HMM). In contrast to most 2D-HMM approaches, we do not use Viterbi Training, instead we propose to propagate the state probabilities through the image. Therefore, we deno...

Descripción completa

Detalles Bibliográficos
Autores principales: Baumgartner, Josef, Scavuzzo, Marcelo, Rodríguez Rivero, Cristian, Pucheta, Julián
Formato: conferenceObject
Lenguaje:Inglés
Publicado: 2022
Materias:
Algorithm
2D-HMM
Complete Enumeration Propagation
Path Constrained Viterbi Training
Acceso en línea:http://hdl.handle.net/11086/29331
Aporte de:
Repositorio Digital Universitario (UNC) de Universidad Nacional de Córdoba
id I10-R141-11086-29331
record_format dspace
institution Universidad Nacional de Córdoba
institution_str I-10
repository_str R-141
collection Repositorio Digital Universitario (UNC)
language Inglés
topic Algorithm
2D-HMM
Complete Enumeration Propagation
Path Constrained Viterbi Training
spellingShingle Algorithm
2D-HMM
Complete Enumeration Propagation
Path Constrained Viterbi Training
Baumgartner, Josef
Scavuzzo, Marcelo
Rodríguez Rivero, Cristian
Pucheta, Julián
A new approach to segmentation of remote sensing images with hidden markov models
topic_facet Algorithm
2D-HMM
Complete Enumeration Propagation
Path Constrained Viterbi Training
description In this work, we present a new segmentation algorithm for remote sensing images based on two-dimensional Hidden Markov Models (2D-HMM). In contrast to most 2D-HMM approaches, we do not use Viterbi Training, instead we propose to propagate the state probabilities through the image. Therefore, we denote our algorithm Complete Enumeration Propagation (CEP). To evaluate the performance of CEP, we compare it to the standard 2D-HMM approach called Path Constrained Viterbi Training (PCVT). As both algorithms are not restricted to a certain emission probability, we evaluate the performance of seven probability functions, namely Gamma, Generalized Extreme Value, inverse Gaussian, Logistic, Nakagami, Normal and Weibull. The experimental results show that our approach outperforms PCVT and other benchmark algorithms. Furthermore, we show that the choice of the probability distribution is crucial for many segmentation tasks.
format conferenceObject
author Baumgartner, Josef
Scavuzzo, Marcelo
Rodríguez Rivero, Cristian
Pucheta, Julián
author_facet Baumgartner, Josef
Scavuzzo, Marcelo
Rodríguez Rivero, Cristian
Pucheta, Julián
author_sort Baumgartner, Josef
title A new approach to segmentation of remote sensing images with hidden markov models
title_short A new approach to segmentation of remote sensing images with hidden markov models
title_full A new approach to segmentation of remote sensing images with hidden markov models
title_fullStr A new approach to segmentation of remote sensing images with hidden markov models
title_full_unstemmed A new approach to segmentation of remote sensing images with hidden markov models
title_sort new approach to segmentation of remote sensing images with hidden markov models
publishDate 2022
url http://hdl.handle.net/11086/29331
work_keys_str_mv AT baumgartnerjosef anewapproachtosegmentationofremotesensingimageswithhiddenmarkovmodels
AT scavuzzomarcelo anewapproachtosegmentationofremotesensingimageswithhiddenmarkovmodels
AT rodriguezriverocristian anewapproachtosegmentationofremotesensingimageswithhiddenmarkovmodels
AT puchetajulian anewapproachtosegmentationofremotesensingimageswithhiddenmarkovmodels
AT baumgartnerjosef newapproachtosegmentationofremotesensingimageswithhiddenmarkovmodels
AT scavuzzomarcelo newapproachtosegmentationofremotesensingimageswithhiddenmarkovmodels
AT rodriguezriverocristian newapproachtosegmentationofremotesensingimageswithhiddenmarkovmodels
AT puchetajulian newapproachtosegmentationofremotesensingimageswithhiddenmarkovmodels
bdutipo_str Repositorios
_version_ 1764820392094990339

Ejemplares similares