What's Mine is Yours: Pretrained CNNs for Limited Training Sonar ATR

TL;DR

This paper explores the use of transfer learning with pretrained CNNs to improve sonar automatic target recognition, demonstrating effective feature extraction and multiple instance detection despite limited sonar data.

Contribution

It introduces a transfer learning approach for sonar ATR, showing how pretrained CNNs can be effectively adapted for sonar imagery with limited data.

Findings

Transfer learning with CNNs yields impressive sonar ATR results.

A flexible CNN feature-extraction strategy is effective.

Proposed method enables multiple instance detection in sonar data.

Abstract

Finding mines in Sonar imagery is a significant problem with a great deal of relevance for seafaring military and commercial endeavors. Unfortunately, the lack of enormous Sonar image data sets has prevented automatic target recognition (ATR) algorithms from some of the same advances seen in other computer vision fields. Namely, the boom in convolutional neural nets (CNNs) which have been able to achieve incredible results - even surpassing human actors - has not been an easily feasible route for many practitioners of Sonar ATR. We demonstrate the power of one avenue to incorporating CNNs into Sonar ATR: transfer learning. We first show how well a straightforward, flexible CNN feature-extraction strategy can be used to obtain impressive if not state-of-the-art results. Secondly, we propose a way to utilize the powerful transfer learning approach towards multiple instance target detection and identification within a provided synthetic aperture Sonar data set.