Classification of Sonar Snapshot Images for Autonomous Underwater Vehicle  


Project Lead Organisation:
University of Wollongong

Collaborating Organisations:
Western Sydney University
Macquarie University
DST Group
Solutions from Silicon Pty Ltd

DIN Funding:


Project objective:

The University of Wollongong has lead a collaboration with academia, industry and DST Group to develop and investigate the application of convolutional neural networks and advanced machine learning for autonomous underwater mine detection and recognition using sonar systems.


While deep learning can produce state-of-the-art classification performances in several application domains, it often relies on a large amount of labeled training data, which are difficult to obtain for application.


This project has increased the understanding of the potential for deep learning to benefit automatic classification of snapshot images containing mine like objects (MLOs), non-mine like objects (NMLOs) or False Alarms as detected by automatic target detection software applied to sonar images. The experimental results indicate the feasibility of the proposed techniques, with a classification accuracy of 98.3%.

 These outcomes outline the potential to incorporate these approaches into automatic target detection software that has been commercialized by DST Group in collaboration with the Solutions from Silicon.

Our Partner Universities

Australian National University logo
Charles Sturt University logo
Macquarie university logo
University of New South Wales logo
University of Newcastle logo
University of Sydney logo
University of Wollongong logo
University of Technology Sydney logo
Western Sydney University logo