Abstract
Automatic target recognition (ATR) of objects in side scan sonar imagery typically employs image processing techniques (e.g. segmentation, Fourier transform) to extract features describing the objects. The features are used to discriminate between sea floor clutter and targets (e.g. sea mines). These methods are typically developed for a specific sonar, and are computationally intensive. The present work used the Restricted Boltzmann Machine (RBM) to discriminate between images of targets and clutter, achieving a 90% probability of detection and a 15% probability of false alarm, which is comparable to the performance of a Support Vector Machine (SVM) and other state-of-the-art methods on the data. The RBM method uses raw image pixels and thus avoids the issue of manually selecting good representations (features) of the data.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Chapple, P.: Automated detection and classification in high-resolution sonar imagery for autonomous underwater vehicle operations. Technical report, Defence Science and Technology Organization (2008)
Fawcett, J., Crawford, A., Hopkin, D., Myers, V., Zerr, B.: Computer-aided detection of targets from the CITADEL trial Klein sonar data. Defence Research and Development Canada Atlantic TM 2006-115 (November 2006), pubs.drdc.gc.ca
Fawcett, J., Crawford, A., Hopkin, D., Couillard, M., Myers, V., Zerr, B.: Computer-aided classification of the Citadel Trial sidescan sonar images. Defence Research and Development Canada Atlantic TM 2007-162 (2007), pubs.drdc.gc.ca
Hinton, G.E., Osindero, S., Teh, Y.W.: A fast learning algorithm for deep belief nets. Neural Comput. 18(7), 1527–1554 (2006)
Fawcett, J., Couillard, M., Hopkin, D., Crawford, A., Myers, V., Zerr, B.: Computer-aided detection and classification of sidescan sonar images from the citadel trial. In: Proceedings of the Institute of Acoustics (2007)
Smolensky, P.: Information processing in dynamical systems: foundations of harmony theory, pp. 194–281 (1986)
Hinton, G.E.: Training products of experts by minimizing contrastive divergence. Neural Comput. 14(8), 1771–1800 (2002)
Nair, V., Hinton, G.E.: Implicit mixtures of restricted boltzmann machines. In: NIPS, pp. 1145–1152 (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Connors, W.A., Connor, P.C., Trappenberg, T. (2010). Detection of Mine-Like Objects Using Restricted Boltzmann Machines. In: Farzindar, A., Kešelj, V. (eds) Advances in Artificial Intelligence. Canadian AI 2010. Lecture Notes in Computer Science(), vol 6085. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-13059-5_47
Download citation
DOI: https://doi.org/10.1007/978-3-642-13059-5_47
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-13058-8
Online ISBN: 978-3-642-13059-5
eBook Packages: Computer ScienceComputer Science (R0)