Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

DisBeaNet: A Deep Neural Network to augment Unmanned Surface Vessels for maritime situational awareness

Srikanth Vemula, Eulises Franco, Michael Frye

TL;DR

The paper tackles the challenge of maritime situational awareness for unmanned surface vessels in contested environments where radar emission is undesirable. It introduces DisBeaNet, a monocular-camera pipeline that combines a modified YOLOv3 detector with a seven-input neural network to estimate distance and bearing and produce geo-referenced vessel tracks. The approach is trained on a custom maritime dataset (~3000 frames) and achieves high accuracy (approx. 94%) with real-time processing, enabling calculation of latitude and longitude from camera pose, distance, and bearing. It offers a low-cost, passive sensing alternative to radar-based perception for USV traffic avoidance, with future work aiming to improve lat/long precision and system robustness.

Abstract

Intelligent detection and tracking of the vessels on the sea play a significant role in conducting traffic avoidance in unmanned surface vessels(USV). Current traffic avoidance software relies mainly on Automated Identification System (AIS) and radar to track other vessels to avoid collisions and acts as a typical perception system to detect targets. However, in a contested environment, emitting radar energy also presents the vulnerability to detection by adversaries. Deactivating these Radiofrequency transmitting sources will increase the threat of detection and degrade the USV's ability to monitor shipping traffic in the vicinity. Therefore, an intelligent visual perception system based on an onboard camera with passive sensing capabilities that aims to assist USV in addressing this problem is presented in this paper. This paper will present a novel low-cost vision perception system for detecting and tracking vessels in the maritime environment. This novel low-cost vision perception system is introduced using the deep learning framework. A neural network, DisBeaNet, can detect vessels, track, and estimate the vessel's distance and bearing from the monocular camera. The outputs obtained from this neural network are used to determine the latitude and longitude of the identified vessel.

DisBeaNet: A Deep Neural Network to augment Unmanned Surface Vessels for maritime situational awareness

TL;DR

The paper tackles the challenge of maritime situational awareness for unmanned surface vessels in contested environments where radar emission is undesirable. It introduces DisBeaNet, a monocular-camera pipeline that combines a modified YOLOv3 detector with a seven-input neural network to estimate distance and bearing and produce geo-referenced vessel tracks. The approach is trained on a custom maritime dataset (~3000 frames) and achieves high accuracy (approx. 94%) with real-time processing, enabling calculation of latitude and longitude from camera pose, distance, and bearing. It offers a low-cost, passive sensing alternative to radar-based perception for USV traffic avoidance, with future work aiming to improve lat/long precision and system robustness.

Abstract

Intelligent detection and tracking of the vessels on the sea play a significant role in conducting traffic avoidance in unmanned surface vessels(USV). Current traffic avoidance software relies mainly on Automated Identification System (AIS) and radar to track other vessels to avoid collisions and acts as a typical perception system to detect targets. However, in a contested environment, emitting radar energy also presents the vulnerability to detection by adversaries. Deactivating these Radiofrequency transmitting sources will increase the threat of detection and degrade the USV's ability to monitor shipping traffic in the vicinity. Therefore, an intelligent visual perception system based on an onboard camera with passive sensing capabilities that aims to assist USV in addressing this problem is presented in this paper. This paper will present a novel low-cost vision perception system for detecting and tracking vessels in the maritime environment. This novel low-cost vision perception system is introduced using the deep learning framework. A neural network, DisBeaNet, can detect vessels, track, and estimate the vessel's distance and bearing from the monocular camera. The outputs obtained from this neural network are used to determine the latitude and longitude of the identified vessel.
Paper Structure (7 sections, 1 equation, 9 figures)

This paper contains 7 sections, 1 equation, 9 figures.

Table of Contents

  1. Introduction
  2. Overview of the System
  3. Real-time Vessel Detection and Tracking
  4. DisBeaNet: A Deep Neural Network to augment Unmanned Surface Vessels in predicting Distance and Bearing using Monocular Camera
  5. Geo-Referenced Vessel Predicting System
  6. Results
  7. Conclusion and Future Work

Figures (9)

  • Figure 1: The DisBeaNet -based system used for distance and bearing estimation of the detected vessel for predicting the Geo-referencing tracks of the vessel from a monocular camera
  • Figure 2: DisBeaNet Neural Network for estimating Distance and bearing of the Vessel Detected
  • Figure 3: DisBeaNet Neural Network results for estimating distance & Bearing of the detected vessel
  • Figure 4: Vessel Geo-Referenced Predicting System
  • Figure 5: Pseudo code for predicted latitude and longitude
  • ...and 4 more figures