Machine Learning Algorithms for Multiple Autonomous Unmanned Vehicle Operations – A Fast Detection Algorithm

PDF

Authors
  1. Shao, H.
  2. Japkowicz, N.
Corporate Authors
Defence R&D Canada - Centre for Operational Research and Analysis, Ottawa ON (CAN);Nathalie Japkowicz Consulting Services, Hampstead QC (CAN)
Abstract
Autonomous Underwater Vehicles (AUVs) are planned to conduct Mine Countermeasure missions in the future. With the help of high resolution imagery produced by the sonar systems mounted on AUVs, mines and other objects of interest can be detected. In this work, the existing approaches for Mine-Like Object (MLO) detection are first reviewed, then, considering the limitation of the exiting works, a novel machine learning method is designed for MLO detection. The experimental result on real side scan sonar images show that the new learning method can provide reliable and fast MLOs detection.

Il y a un résumé en français ici.

Report Number
DRDC-CORA-CR-2013-059 — Contractor Report
Date of publication
01 Apr 2013
Number of Pages
35
DSTKIM No
CA037692
CANDIS No
537526
Format(s):
Electronic Document(PDF)

Permanent link

Document 1 of 1

Date modified: