Assessment of SAR speckle filters in the context of object-based image analysis

The initial step in most object-based classification methodologies is the application of a segmentation algorithm to define objects. In the context of synthetic aperture radar (SAR) image analysis, the presence of speckle noise might hamper the segmentation quality. The aim of this study is to asses...

Descripción completa

Detalles Bibliográficos
Autores principales: Morandeira, Natalia Soledad, Grimson, Rafael, Kandus, Patricia
Formato: publishedVersion Artículo
Lenguaje:Inglés
Publicado: Taylor & Francis 2016
Materias:
ACTIVE MICROWAVE
CLASSIFICATION
SEGMENTATION
SPECKLE FILTERING
CIENCIAS EXACTAS Y NATURALES
Ciencias de la Tierra, del Agua y de la Atmósfera
Acceso en línea:https://ri.unsam.edu.ar/handle/123456789/779
Aporte de:
Repositorio Institucional de la UNSAM de Universidad Nacional de General San Martín
id I78-R216-123456789-779
record_format dspace
spelling I78-R216-123456789-7792020-12-26T22:05:40Z Assessment of SAR speckle filters in the context of object-based image analysis Morandeira, Natalia Soledad Grimson, Rafael Kandus, Patricia ACTIVE MICROWAVE CLASSIFICATION SEGMENTATION SPECKLE FILTERING CIENCIAS EXACTAS Y NATURALES Ciencias de la Tierra, del Agua y de la Atmósfera info:eu-repo/semantics/publishedVersion The initial step in most object-based classification methodologies is the application of a segmentation algorithm to define objects. In the context of synthetic aperture radar (SAR) image analysis, the presence of speckle noise might hamper the segmentation quality. The aim of this study is to assess the segmentation performance of SAR images when no filter or different filters are applied before segmentation. In particular, the performance of the mean-shift segmentation algorithm combined with different adaptive and non-adaptive filters is assessed based on both synthetic and natural SAR images. Studied filters include the non-adaptive Boxcar filter and four adaptive filters: the well-known Refined Lee filter and three recently proposed non-local filters differing, in particular, in their dissimilarity criteria: the Hellinger and the Kullback-Leibler filters are based on stochastic distances, whereas the NL-SAR filter is based on the generalized likelihood ratio. Two measures were used for quality assessment: ?-index and ?-index. Over-segmentation was assessed by the ?-index, the ratio of the resulting number of segments to the number of connected components of the ground-truth classes. The accuracy of the best possible classification given on the segmentation result was assessed with ground truth information by maximizing the ?-index. A Monte Carlo experiment conducted on synthetic images shows that the quality measures significantly differ for the applied filters. Our results indicate that the use of an adaptive filter improves the performance of the segmentation. In particular, the combination of the mean-shift segmentation algorithm with the NLSAR filter gives the best results and the resulting process is less sensitive to variations in the mean-shift operational parameters than when applying other filters or no filter. The results obtained may help improve the reliability of land-cover classification analyses based on an object-based approach on SAR data. Fil: Morandeira, Natalia Soledad. Universidad Nacional de San Martín. Instituto de Investigación e Ingeniería Ambiental. Consejo Nacional de Investigaciones Científicas y Técnicas. Argentina Fil: Grimson, Rafael. Universidad Nacional de San Martín. Escuela de Ciencia y Tecnología. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina Fil: Kandus, Patricia. Universidad Nacional de San Martín. Instituto de Investigación e Ingeniería Ambiental. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina 2016-01 info:eu-repo/semantics/article info:ar-repo/semantics/artículo Morandeira, Natalia Soledad; Grimson, Rafael; Kandus, Patricia; Assessment of SAR speckle filters in the context of object-based image analysis; Taylor & Francis; Remote Sensing Letters; 7; 2; 1-2016; 150-159 2150-704X https://ri.unsam.edu.ar/handle/123456789/779 eng info:eu-repo/semantics/restrictedAccess application/pdf 10 p. application/pdf Taylor & Francis Remote Sensing Letters. 2016; 7(2): 150-159 http://dx.doi.org/10.1080/2150704X.2015.1117153
institution Universidad Nacional de General San Martín
institution_str I-78
repository_str R-216
collection Repositorio Institucional de la UNSAM
language Inglés
topic ACTIVE MICROWAVE
CLASSIFICATION
SEGMENTATION
SPECKLE FILTERING
CIENCIAS EXACTAS Y NATURALES
Ciencias de la Tierra, del Agua y de la Atmósfera
spellingShingle ACTIVE MICROWAVE
CLASSIFICATION
SEGMENTATION
SPECKLE FILTERING
CIENCIAS EXACTAS Y NATURALES
Ciencias de la Tierra, del Agua y de la Atmósfera
Morandeira, Natalia Soledad
Grimson, Rafael
Kandus, Patricia
Assessment of SAR speckle filters in the context of object-based image analysis
topic_facet ACTIVE MICROWAVE
CLASSIFICATION
SEGMENTATION
SPECKLE FILTERING
CIENCIAS EXACTAS Y NATURALES
Ciencias de la Tierra, del Agua y de la Atmósfera
description The initial step in most object-based classification methodologies is the application of a segmentation algorithm to define objects. In the context of synthetic aperture radar (SAR) image analysis, the presence of speckle noise might hamper the segmentation quality. The aim of this study is to assess the segmentation performance of SAR images when no filter or different filters are applied before segmentation. In particular, the performance of the mean-shift segmentation algorithm combined with different adaptive and non-adaptive filters is assessed based on both synthetic and natural SAR images. Studied filters include the non-adaptive Boxcar filter and four adaptive filters: the well-known Refined Lee filter and three recently proposed non-local filters differing, in particular, in their dissimilarity criteria: the Hellinger and the Kullback-Leibler filters are based on stochastic distances, whereas the NL-SAR filter is based on the generalized likelihood ratio. Two measures were used for quality assessment: ?-index and ?-index. Over-segmentation was assessed by the ?-index, the ratio of the resulting number of segments to the number of connected components of the ground-truth classes. The accuracy of the best possible classification given on the segmentation result was assessed with ground truth information by maximizing the ?-index. A Monte Carlo experiment conducted on synthetic images shows that the quality measures significantly differ for the applied filters. Our results indicate that the use of an adaptive filter improves the performance of the segmentation. In particular, the combination of the mean-shift segmentation algorithm with the NLSAR filter gives the best results and the resulting process is less sensitive to variations in the mean-shift operational parameters than when applying other filters or no filter. The results obtained may help improve the reliability of land-cover classification analyses based on an object-based approach on SAR data.
format publishedVersion
Artículo
Artículo
author Morandeira, Natalia Soledad
Grimson, Rafael
Kandus, Patricia
author_facet Morandeira, Natalia Soledad
Grimson, Rafael
Kandus, Patricia
author_sort Morandeira, Natalia Soledad
title Assessment of SAR speckle filters in the context of object-based image analysis
title_short Assessment of SAR speckle filters in the context of object-based image analysis
title_full Assessment of SAR speckle filters in the context of object-based image analysis
title_fullStr Assessment of SAR speckle filters in the context of object-based image analysis
title_full_unstemmed Assessment of SAR speckle filters in the context of object-based image analysis
title_sort assessment of sar speckle filters in the context of object-based image analysis
publisher Taylor & Francis
publishDate 2016
url https://ri.unsam.edu.ar/handle/123456789/779
work_keys_str_mv AT morandeiranataliasoledad assessmentofsarspecklefiltersinthecontextofobjectbasedimageanalysis
AT grimsonrafael assessmentofsarspecklefiltersinthecontextofobjectbasedimageanalysis
AT kanduspatricia assessmentofsarspecklefiltersinthecontextofobjectbasedimageanalysis
_version_ 1765660360671494144

Ejemplares similares