Visible to the public Hunting for Naval Mines with Deep Neural Networks

TitleHunting for Naval Mines with Deep Neural Networks
Publication TypeConference Paper
Year of Publication2017
AuthorsGebhardt, D., Parikh, K., Dzieciuch, I., Walton, M., Hoang, N. A. V.
Conference NameOCEANS 2017 - Anchorage
ISBN Number978-0-6929-4690-9
Keywordsautonomous underwater vehicles, autonomous unmanned underwater vehicle, Brain modeling, calculation requirements, computation requirements, Computational modeling, data distribution training, Data models, deep neural network methods, deep neural networks, detection efficacy, DNN depth, DNN models, explosive naval mines, explosives, feature extraction, image recognition, image recognition tasks, learning (artificial intelligence), memory requirements, Metrics, minelike objects, modest DNN model, naval engineering computing, neural nets, privacy, pubcrawl, remotely operated vehicles, robot vision, sea faring vessels, side-scan sonar imagery, Sonar, sonar imaging, support vector machine, Support vector machines, threat vectors, Throughput, trainable parameter count, Training, Training data, Underwater vehicles, visualization technique, Weapons

Explosive naval mines pose a threat to ocean and sea faring vessels, both military and civilian. This work applies deep neural network (DNN) methods to the problem of detecting minelike objects (MLO) on the seafloor in side-scan sonar imagery. We explored how the DNN depth, memory requirements, calculation requirements, and training data distribution affect detection efficacy. A visualization technique (class activation map) was incorporated that aids a user in interpreting the model's behavior. We found that modest DNN model sizes yielded better accuracy (98%) than very simple DNN models (93%) and a support vector machine (78%). The largest DNN models achieved textless;1% efficacy increase at a cost of a 17x increase of trainable parameter count and computation requirements. In contrast to DNNs popularized for many-class image recognition tasks, the models for this task require far fewer computational resources (0.3% of parameters), and are suitable for embedded use within an autonomous unmanned underwater vehicle.

Citation Keygebhardt_hunting_2017