Classification of small boats in infrared images for maritime surveillance

Abstract

Autonomous round-the-clock observation of wide critical maritime areas can be a powerful support for border protection agencies to avoid criminal acts like illegal immigration, piracy or drug trafficking. These criminal acts are often accomplished by using small boats to decrease the probability of being uncovered. In this paper, we present an image exploitation approach to detect and classify maritime objects in infrared image sequences recorded from an autonomous platform. We focus on high robustness and generality with respect to variations of boat appearance, image quality, and environmental condition. A fusion of three different detection algorithms is performed to create reliable alarm hypotheses. In the following, a set of well-investigated features is extracted from the alarm hypotheses and evaluated using a two-stage-classification with support vector machines (SVMs) in order to distinguish between three object classes: clutter, irrelevant objects and suspicious boats. On the given image data we achieve a rate of 97% correct classifications

Maritime piracy situation modelling with dynamic Bayesian networks

“…An approach to detect pirates through satellite communication monitoring has been proposed [54]. Other approaches intend to detect pirate vessels by classifying small craft in imagery [55], [56].…”

A unified model for context-based behavioural modelling and classification

“…A satellite communication monitoring system for detecting maritime pirates has been proposed (Baldini et al, 2010). Other methods that detect pirate vessels by small craft classification in various forms of imagery have been proposed (Teutsch and Kruger, 2010), (Sanderson et al, 1999). A computer vision method of identifying and tracking vessels, such as pirate vessels, is presented in (Szpak and Tapamo, 2011).…”

“…Classification using low resolution radar has also been considered in (Gibbins et al, 1999). Several approaches to vessel classification using infra-red imagery have been published (Teutsch and Kruger, 2010;Alves et al, 2004;Li and Wang, 2008). Vessel classification using land based and ship based optical imagery in harbours has been proposed (Zia-ur Rahman, 2008;Lam, 2013).…”

Maritime surveillance with integrated systems

“…Those services facilitate resource planning, sensor and vehicle coordination, sensor data exploitation, and information dissemination. Additional specialized services such as the automatic detection and classification of maritime objects in IR images [3] or the fusion of uncertain and inaccurate data [4] can be integrated supplementary. Two basic support services, namely rule-based event response and local aerial reconnaissance support, are described in the following paragraphs.…”