Accession Number:

AD1120709

Title:

Automatic Target Classification Using Multiple Sidescan Sonar Images of Different Orientations

Descriptive Note:

[Technical Report, Memorandum Report]

Corporate Author:

NATO, SACLANT Undersea Research Centre

Personal Author(s):

• Zerr, B
• Stage, B
• Guerrero, A

Report Date:

1997-09-01

Pagination or Media Count:

86

Abstract:

In this report, the target classification performance of a multiple view side scan sonar is investigated. The classification statistics are estimated using model based automatic classifiers. The guidelines to the design of efficient classification algorithms are defined. The shadow is retained as the basic information for target classification. The input feature vector of the automatic classifier is the cross sectionor height profile of the target estimated from its shadow.

Descriptors:

• neural networks
• supervised machine learning
• target recognition
• detectors
• information science
• self organizing systems
• databases
• computer vision
• detection
• artificial intelligence software
• autonomous underwater vehicles
• computer languages
• network science
• pattern recognition
• warning systems
• target classification
• automata theory
• collision avoidance
• computers
• image processing

Subject Categories:

• Target Direction, Range and Position Finding
• Acoustic Detection and Detectors

Distribution Statement:

[A, Approved For Public Release]