For effective recognition, we developed a recognition component, combining kinematic vessel streams with library of maritime patterns in close collaboration with domain contextual (geographical) knowledge for real-time vessel activity experts. We present a thorough evaluation of the system and the detection. To improve the accuracy of the system, we collaborated, patterns both in terms of predictive accuracy and computational in the context of this paper, with domain experts in order to construct efficiency, using real-world datasets of vessel position streams and effective patterns of maritime activity. Thus, we present a contextual geographical information.

We present a system for online monitoring of maritime activity over streaming positions from numerous vessels sailing at sea. It employs an online tracking module for detecting important changes in the evolving trajectory of each vessel across time, and thus can incrementally retain concise, yet reliable summaries of its recent movement. In addition, thanks to its complex event recognition module, this system can also offer instant notification to marine authorities regarding emergency situations, such as risk of collisions, suspicious moves in protected zones, or package picking at open sea. Not only did our extensive tests validate the performance, efficiency, and robustness of the system against scalable volumes of real-world and synthetically enlarged datasets, but its deployment against online feeds from vessels has also confirmed its capabilities for effective, real-time maritime surveillance.

The output of the former type of recognition is a set of activities taking place in a short period of time: 'short-term activities'. The output of the latter type of recognition is a set of'long-term activities', ie predefined temporal combinations of short-term activities. We focus on high-level recognition. We define a set of long-term activities of interest, such as'fighting' and'meeting', as temporal combinations of short-term activities -- eg, 'walking', 'running', and'inactive' (standing still) -- using a logic programming implementation of the Event Calculus [9]. More precisely, we employ the Event Calculus to express the temporal constraints on a set of short-term activities that, if satisfied, lead to the recognition of a long-term activity. We presented preliminary results on activity recognition from video content in [2].