Adaptive Surface Ship-Wake Detection Based on Improved One-Class Support Vector Machine

DOI

10.2991/icmii-15.2015.112How to use a DOI?

Keywords

mathematic wake echo signal model; One-class support vector machine; Sequential minimal optimization; ship-wake detector.

Abstract

It is difficult to collect bubble-wake signals from different ocean environments caused by various types of ship. One-class support vector machine (OCSVM) can make decision based on incomplete information. This paper found an OCSVM detection model which use only reflected signals without a bubble wake to detect the surface ship-wake. In order to improve the training efficiency, a training algorithm based on Sequential Minimal Optimization (SMO) was introduced for OCSVM. Grid search method and Particle Swarm Optimization (PSO) algorithm are used to search optimal parameter. The simulation shows that the proposed detector can detect the ship weak well and it was robust with respect to noisy signals.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Cheng Wang
AU  - Tingfei Yang
AU  - Qiang Meng
PY  - 2015/10
DA  - 2015/10
TI  - Adaptive Surface Ship-Wake Detection Based on Improved One-Class Support Vector Machine
BT  - Proceedings of the 3rd International Conference on Mechatronics and Industrial Informatics
PB  - Atlantis Press
SP  - 656
EP  - 661
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmii-15.2015.112
DO  - 10.2991/icmii-15.2015.112
ID  - Wang2015/10
ER  -