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

dc.creatorMorandeira, Natalia Soledad
dc.creatorGrimson, Rafael
dc.creatorKandus, Patricia
dc.date2019-08-28T22:32:38Z
dc.date2019-08-28T22:32:38Z
dc.date2016-01
dc.date.accessioned2021-10-07T00:21:08Z
dc.date.available2021-10-07T00:21:08Z
dc.identifierMorandeira, 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
dc.identifier2150-704X
dc.identifierhttp://ri.unsam.edu.ar/xmlui/handle/123456789/779
dc.identifierhttp://hdl.handle.net/123456789/779
dc.identifier.urihttp://repositorioslatinoamericanos.uchile.cl/handle/2250/3309377
dc.descriptionThe 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.
dc.descriptionFil: Morandeira, Natalia Soledad. Universidad Nacional de San Martín. Instituto de Investigación e Ingeniería Ambiental. Consejo Nacional de Investigaciones Científicas y Técnicas. Argentina
dc.descriptionFil: Grimson, Rafael. Universidad Nacional de San Martín. Escuela de Ciencia y Tecnología. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
dc.descriptionFil: Kandus, Patricia. Universidad Nacional de San Martín. Instituto de Investigación e Ingeniería Ambiental. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
dc.formatapplication/pdf
dc.format10 p.
dc.formatapplication/pdf
dc.languageeng
dc.publisherTaylor & Francis
dc.rightsinfo:eu-repo/semantics/restrictedAccess
dc.sourceRemote Sensing Letters. 2016; 7(2): 150-159
dc.sourcehttp://dx.doi.org/10.1080/2150704X.2015.1117153
dc.subjectACTIVE MICROWAVE
dc.subjectCLASSIFICATION
dc.subjectSEGMENTATION
dc.subjectSPECKLE FILTERING
dc.subjectCIENCIAS EXACTAS Y NATURALES
dc.subjectCiencias de la Tierra, del Agua y de la Atmósfera
dc.titleAssessment of SAR speckle filters in the context of object-based image analysis
dc.typeinfo:eu-repo/semantics/publishedVersion
dc.typeinfo:eu-repo/semantics/article
dc.typeinfo:ar-repo/semantics/artículo


Este ítem pertenece a la siguiente institución