Saliency-Based Detection for Maritime Object Tracking

This paper presents a new method for object detection and tracking based on visual saliency as a way of mitigating against challenges present in maritime environments. Object detection is based on adaptive hysteresis thresholding of a saliency map generated with a modified version of the Boolean Map Saliency (BMS) approach. We show that the modification reduces false positives by suppressing detection of wakes and surface glint. Tracking is performed by matching detections frame to frame and smoothing trajectories with a Kalman filter. The proposed approach is evaluated on the PETS 2016 challenge dataset on detecting and tracking boats around a vessel at sea.

1. Introduction

Maritime piracy continues to place a huge economic and human cost on commercial shipping around the world [1]. The most effective protection for ships is a proper lookout to maximise early warning of a potential attack, allowing time for the crew to prepare accordingly [3]. Radar and crew members with binoculars represent the state of the art technology available to commercial fleets. However, the navigation radar available on ships does not perform well with small, fast-moving objects [24] such as the ‘skiffs’ used by pirates, and crew members become fatigued after maintaining a lookout for a long period.