Texture recognition in synthetic aperture sonar images with scattering operators

Nicolas Erwan Valeyrie, Yan Pailhas, Christopher Capus, Keith Edgar Brown

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The proposed paper is about the description of textured areas in synthetic aperture sonar images. The texture descriptors come from the recursive interferometric representation of signals. This representation originates from a scattering operator that iterates over a wavelet transform and a modulus operator. This representation is invariant to translations and Lipschitz continuous relatively to deformations up to a log factor. A scattering metric can be derived from the scattering operator and used to quantify the similarity between signals. The scattering metric was used with a simple classifier to classify synthetic aperture sonar images with respect to the types of sea bottom.
Original languageEnglish
Title of host publication4th International Conference and Exhibition on Underwater Acoustic Measurements: Technologies & Results
ISBN (Electronic)978-960-98883-5-6
Publication statusPublished - Jun 2011
Event4th International Conference and Exhibition on Underwater Acoustic Measurements: Technologies and Results - Kos Island, Greece
Duration: 20 Jun 201124 Jun 2011

Conference

Conference4th International Conference and Exhibition on Underwater Acoustic Measurements
Abbreviated titleUAM 2011
CountryGreece
CityKos Island
Period20/06/1124/06/11

Fingerprint Dive into the research topics of 'Texture recognition in synthetic aperture sonar images with scattering operators'. Together they form a unique fingerprint.

Cite this