Mine detection and classification in side scan sonar

Scott Reed, Yvan Petillot, Judith Bell

Research output: Contribution to journalArticle

Abstract

Autonomous underwater vehicle (AUV) technology has matured sufficiently to allow large areas of seafloor to be covered quickly. This has led to an increased interest in automated mine countermeasures (MCM) techniques. The majority of current models require training, and their success rate is often dependent on the similarity between the testing data and the data used to train the system. They also often produce a black box solution to the problem. Therefore, while the correct result may be obtained, it is often very difficult to determine why the model has produced such a result. The approach detailed in this article is a model-based alternative to traditional supervised models.

Original languageEnglish
Pages (from-to)35-39
Number of pages5
JournalSea Technology
Volume45
Issue number11
Publication statusPublished - Nov 2004

Fingerprint Dive into the research topics of 'Mine detection and classification in side scan sonar'. Together they form a unique fingerprint.

  • Cite this