Target classification from HR sonar images

Olga Lopera, Yves Dupont

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdragepeer review

Samenvatting

This paper presents two integrated techniques for target classification from high-resolution (HR) sonar images. Both recognition procedures start with a despeckling algorithm based on the anisotropic diffusion filter. As a second step, a fuzzymorpho-based segmentation procedure is applied to the filtered images, which partitions the image into highlights and shadow areas. A number of geometrical features are extracted from these areas, and are then used to classify targets using two techniques: (i) a Markov Chain Monte Carlo (MCMC) approach and (ii) a Decision Tree Classifier (DTC). A comparison of both recognition techniques is drawn, and classification performance is estimated by ROC curves. Very promising results are obtained.

Originele taal-2Engels
TitelOCEANS 2013 MTS/IEEE Bergen
SubtitelThe Challenges of the Northern Dimension
DOI's
StatusGepubliceerd - 2013
EvenementOCEANS 2013 MTS/IEEE Bergen: The Challenges of the Northern Dimension - Bergen, Noorwegen
Duur: 10 jun. 201313 jun. 2013

Publicatie series

NaamOCEANS 2013 MTS/IEEE Bergen: The Challenges of the Northern Dimension

Congres

CongresOCEANS 2013 MTS/IEEE Bergen: The Challenges of the Northern Dimension
Land/RegioNoorwegen
StadBergen
Periode10/06/1313/06/13

Vingerafdruk

Duik in de onderzoeksthema's van 'Target classification from HR sonar images'. Samen vormen ze een unieke vingerafdruk.

Citeer dit