Composite Event Recognition for Maritime Monitoring

Pitsikalis, Manolis; Artikis, Alexander; Dreo, Richard; Ray, Cyril; Camossi, Elena; Jousselme, Anne-Laure

doi:10.1145/3328905.3329762

Cited by 41 publications

(25 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A vessel is considered underway when it is not aground, anchored, or has not been made fast to a dock, the shore, or some other stationary object. An underway vessel is not necessarily propelled or pushed by an instrument or a device, but it may be underway because of the wind or the sea current [52,53]. Figure 1c shows the trajectory of an underway vessel.…”

Section: Maritime Patternsmentioning

confidence: 99%

A Deep Learning Streaming Methodology for Trajectory Classification

Kontopoulos

Makris

Tserpes

2021

IJGI

View full text Add to dashboard Cite

Due to the vast amount of available tracking sensors in recent years, high-frequency and high-volume streams of data are generated every day. The maritime domain is no different as all larger vessels are obliged to be equipped with a vessel tracking system that transmits their location periodically. Consequently, automated methodologies able to extract meaningful information from high-frequency, large volumes of vessel tracking data need to be developed. The automatic identification of vessel mobility patterns from such data in real time is of utmost importance since it can reveal abnormal or illegal vessel activities in due time. Therefore, in this work, we present a novel approach that transforms streaming vessel trajectory patterns into images and employs deep learning algorithms to accurately classify vessel activities in near real time tackling the Big Data challenges of volume and velocity. Two real-world data sets collected from terrestrial, vessel-tracking receivers were used to evaluate the proposed methodology in terms of both classification and streaming execution performance. Experimental results demonstrated that the vessel activity classification performance can reach an accuracy of over 96% while achieving sub-second latencies in streaming execution performance.

show abstract

Section: Maritime Patternsmentioning

confidence: 99%

A Deep Learning Streaming Methodology for Trajectory Classification

Kontopoulos

Makris

Tserpes

2021

IJGI

View full text Add to dashboard Cite

show abstract

“…The unsupervised approaches that already exist in the literature, show that using some simplistic rules [2] that seek for repetitive changes in speed and orientation or using a spatial clustering of important trajectory points [13] is enough for distinguishing between a normal movement and a potential SAR maneuver. However, when we need to detect the type of maneuver and decide whether it is a Search one (e.g., a step-up pattern) or a Rescue one (e.g., a spiral) it is necessary to classify the maneuver to the nearest document pattern.…”

Section: Experimental Evaluation and Resultsmentioning

confidence: 99%

“…Motivated by the abundance of AIS data and the rich information that they contain, a lot of data mining applications have been developed by researchers on top of historic AIS data, ranging from the automatic detection of events [1,2] to the prediction of future vessel position [3] and estimation of the time of arrival at the port [4]. In a parallel line of research, several works on trajectory data analytics have contributed platforms for the visualisation of vessel trajectories [5,6], efficient handling of large volumes of such spatio-temporal data streams [7,8] and methodologies for the abstraction of AIS data collected for a period and a region [9].…”

Section: Introductionmentioning

confidence: 99%

“…The current work focuses on the classification of SAR maneuvers that may happen during a vessel's trip and thus cannot be directly compared to methods that classify the whole voyage as a SAR mission or not (e.g., References [14,15]) or methods that search for more composite naval events that examine more than one vessels in tandem (e.g., References [1,2]). However, it proposes a new classification approach for documented maneuvers such as the SAR ones, in the absence of annotated data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Framework for the Detection of Search and Rescue Patterns Using Shapelet Classification

et al. 2019

View full text Add to dashboard Cite

The problem of unmanned supervision of maritime areas has attracted the interest of researchers for the last few years, mainly thanks to the advances in vessel monitoring that the Automatic Identification System (AIS) has brought. Several frameworks and algorithms have been proposed for the management of vessel trajectory data, which focus on data compression, data clustering, classification and visualization, offering a wide variety of solutions from vessel monitoring to automatic detection of complex events. This work builds on our previous work in the topic of automatic detection of Search and Rescue (SAR) missions, by developing and evaluating a methodology for classifying the trajectories of vessels that possibly participate in such missions. The proposed solution takes advantage of a synthetic trajectory generator and a classifier that combines a genetic algorithm (GENDIS) for the extraction of informative shapelets from training data and a transformation to the shapelets’ feature space. Using the generator and several SAR patterns that are formally described in naval operations bibliography, it generates a synthetic dataset that is used to train the classifier. Evaluation on both synthetic and real data has very promising results and helped us to identify vessel SAR maneuvers without putting any effort into manual annotation.

show abstract

“…These Complex Events are defined by input events along with additional conditions upon their attributes, their relations within the stream or both. CEP on the maritime domain may include the detection of illegal activity, such as fishing in prohibited areas, or reporting on ship-to-ship transfer of goods [18], [19].…”

Section: Introductionmentioning

confidence: 99%

Experimental Comparison of Complex Event Processing Systems in the Maritime Domain

Troupiotis-Kapeliaris

Chatzikokolakis

Zissis

et al. 2020

2020 21st IEEE International Conference on Mobile Data Management (MDM)

View full text Add to dashboard Cite

Complex Event Processing (CEP) 's main purpose is recognizing interesting phenomena upon streams of data. So its only natural that it would find applications in the maritime domain, where detecting vessel activity plays an important role in monitoring movement at sea. In this study we briefly examine the field of Complex Event Processing; we present two CEP implementations, one based on machine learning techniques and a rule-based system modeled with Event Calculus. Finally, we evaluate their ability in modeling activities that involve multiple vessels, by comparing their results on real-life examples.

show abstract

Composite Event Recognition for Maritime Monitoring

Cited by 41 publications

References 21 publications

A Deep Learning Streaming Methodology for Trajectory Classification

A Deep Learning Streaming Methodology for Trajectory Classification

A Framework for the Detection of Search and Rescue Patterns Using Shapelet Classification

Experimental Comparison of Complex Event Processing Systems in the Maritime Domain

Contact Info

Product

Resources

About