A distributed lightning fast maritime anomaly detection service

Chatzikokolakis, Konstantinos; Zissis, Dimitris; Vodas, Marios; Spiliopoulos, Giannis; Kontopoulos, Ioannis

doi:10.1109/oceanse.2019.8867269

Cited by 9 publications

(9 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although many methodologies have been developed for trajectory classification, fewer studies have focused on real-time stream processing of events in the maritime domain [10,31,32]. Lin et al [31] extracted features from AIS messages, which were then fed to a deep neural network for the prediction of the Estimated Time of Arrival (ETA) of vessels.…”

Section: Trajectory Classificationmentioning

confidence: 99%

“…Lin et al [31] extracted features from AIS messages, which were then fed to a deep neural network for the prediction of the Estimated Time of Arrival (ETA) of vessels. Chatzikokolakis et al [32] developed a real-time anomaly detection service, which was focused on identifying a wide range of events of interest in the maritime domain either through the use of machine learning techniques or with rule-based approaches. The authors in [10] presented an online feature extraction process for the classification of fishing trajectories in real time with the use of random forests.…”

Section: Trajectory Classificationmentioning

confidence: 99%

See 1 more Smart Citation

A Deep Learning Streaming Methodology for Trajectory Classification

Kontopoulos

Makris

Tserpes

2021

IJGI

Self Cite

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: Trajectory Classificationmentioning

confidence: 99%

Section: Trajectory Classificationmentioning

confidence: 99%

A Deep Learning Streaming Methodology for Trajectory Classification

Kontopoulos

Makris

Tserpes

2021

IJGI

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, given the requirement for massive vessel traffic, metocean, and other data sets to achieve this, a big data processing solution would be required (Abualhaol, Falcon, Abielmona, & Petriu, 2018; Lensu & Goerlandt, 2019). While some work has considered the application of architectures such as Apache Spark or Hadoop for maritime risk analysis (Chatzikokolakis, Zissis, Vodas, Spiliopoulos, & Kotopoulos, 2019; Filipiak et al., 2018; Zhang, Meng, & Fwa, 2019), further research is required to integrate machine learning processes with big data infrastructure for maritime risk assessment.…”

Section: Discussion and Future Research Directionsmentioning

confidence: 99%

Spatial Modeling of Maritime Risk Using Machine Learning

2021

View full text Add to dashboard Cite

Managing navigational safety is a key responsibility of coastal states. Predicting and measuring these risks has a high complexity due to their infrequent occurrence, multitude of causes, and large study areas. As a result, maritime risk models are generally limited in scale to small regions, generalized across diverse environments, or rely on the use of expert judgement. Therefore, such an approach has limited scalability and may incorrectly characterize the risk. Within this article a novel method for undertaking spatial modeling of maritime risk is proposed through machine learning. This enables navigational safety to be characterized while leveraging the significant volumes of relevant data available. The method comprises two key components: aggregation of historical accident data, vessel traffic, and other exploratory features into a spatial grid; and the implementation of several classification algorithms that predicts annual accident occurrence for various vessel types. This approach is applied to characterize the risk of collisions and groundings in the United Kingdom. The results vary between hazard types and vessel types but show remarkable capability at characterizing maritime risk, with accuracies and area under curve scores in excess of 90% in most implementations. Furthermore, the ensemble tree‐based algorithms of XGBoost and Random Forest consistently outperformed other machine learning algorithms that were tested. The resultant potential risk maps provide decisionmakers with actionable intelligence in order to target risk mitigation measures in regions with the greatest requirement.

show abstract

“…They used HBase, which is a popular distributed database running on Hadoop Distributed Filesystem (HDFS). Also, Chatzikokolakis, Zissis, Vodas, Spiliopoulos, and Kontopoulos (2019) proposed a distributed architecture for detecting possible collisions, groundings and travel patterns deviations based on AIS. Their system follows the Lambda architecture paradigm, in which one part of data processing is executed in batches, while the other one in the streaming mode with the goal of detecting deviations of vessel behaviour in real-time.…”

Section: Application Of Big Data Technologies For Maritime Anomalies ...mentioning

confidence: 99%

Data analysis in the maritime domain

Stróżyna¹,

Abramowicz²,

Węcel³

et al. 2022

View full text Add to dashboard Cite

The book delivers a rich set of foundations, state-of-the-art knowledge, new approaches and methods for the purpose of anomalies detection, maritime traffic analysis as well as risk and reliability assessment. It addresses relevant research problems at the intersection of maritime transport in global economies, reliability and risk assessment, and information systems and data processing. The book provides a theoretical overview of available maritime data sources and approaches for maritime data analysis, as well as a set of novel tools and methods for maritime data retrieval, fusion, and analysis. The proposed methods are evaluated on real-life AIS data, covering the entire world and more than 200 thousands of vessels, illustrating how they may be used for anomaly detection and risk assessment. The primary audience of the book are researchers from the fields of computer science and maritime transport as well as logistics service providers, shipping companies and port authorities companies that need support in managing security, safety, and risk of maritime transport services in global economies by making use of large-scale data processing.

show abstract

A distributed lightning fast maritime anomaly detection service

Cited by 9 publications

References 13 publications

A Deep Learning Streaming Methodology for Trajectory Classification

A Deep Learning Streaming Methodology for Trajectory Classification

Spatial Modeling of Maritime Risk Using Machine Learning

Data analysis in the maritime domain

Contact Info

Product

Resources

About