Context-based behaviour modelling and classification of marine vessels in an abalone poaching situation

Dabrowski, Joel Janek; Villiers, J. P. de; Beyers, Conrad

doi:10.1016/j.engappai.2017.06.005

Cited by 7 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Trajectory-based techniques treat the entire traffic data sequence as a whole. Several research directions are analyzed in the literature related to the analysis of vessel trajectories: maritime traffic pattern mining, ship collision risk assessment [12], maritime anomaly detection [13,14,15], identification of the types of ships [16], and combating abalone poaching [17].…”

Section: Reviewmentioning

confidence: 99%

Real-Time Maritime Traffic Anomaly Detection Based on Sensors and History Data Embedding

Venskus

Treigys

Bernatavičienė

et al. 2019

Sensors

View full text Add to dashboard Cite

The automated identification system of vessel movements receives a huge amount of multivariate, heterogeneous sensor data, which should be analyzed to make a proper and timely decision on vessel movements. The large number of vessels makes it difficult and time-consuming to detect abnormalities, thus rapid response algorithms should be developed for a decision support system to identify abnormal movements of vessels in areas of heavy traffic. This paper extends the previous study on a self-organizing map application for processing of sensor stream data received by the maritime automated identification system. The more data about the vessel’s movement is registered and submitted to the algorithm, the higher the accuracy of the algorithm should be. However, the task cannot be guaranteed without using an effective retraining strategy with respect to precision and data processing time. In addition, retraining ensures the integration of the latest vessel movement data, which reflects the actual conditions and context. With a view to maintaining the quality of the results of the algorithm, data batching strategies for the neural network retraining to detect anomalies in streaming maritime traffic data were investigated. The effectiveness of strategies in terms of modeling precision and the data processing time were estimated on real sensor data. The obtained results show that the neural network retraining time can be shortened by half while the sensitivity and precision only change slightly.

show abstract

Section: Reviewmentioning

confidence: 99%

Real-Time Maritime Traffic Anomaly Detection Based on Sensors and History Data Embedding

Venskus

Treigys

Bernatavičienė

et al. 2019

Sensors

View full text Add to dashboard Cite

show abstract

“…One approach would be to include contextual information. This can be achieved in the SLDS using methods such as those described in [22,23,24]. Contextual information could include road signs, proximity to crossing areas, and traffic congestion levels.…”

Section: Discussionmentioning

confidence: 99%

“…The SLDS has been extended in various ways, such as introducing variables representing behavioural context information. Such models have been applied to maritime piracy applications [22,23] and abalone poaching applications [24]. Linderman et.…”

Section: Switching Linear Dynamical Systemmentioning

confidence: 99%

Deep Learning and Statistical Models for Time-Critical Pedestrian Behaviour Prediction

Dabrowski¹,

Villiers²,

Rahman³

et al. 2019

Communications in Computer and Information Science

Self Cite

View full text Add to dashboard Cite

The time it takes for a classifier to make an accurate prediction can be crucial in many behaviour recognition problems. For example, an autonomous vehicle should detect hazardous pedestrian behaviour early enough for it to take appropriate measures. In this context, we compare the switching linear dynamical system (SLDS) and a three-layered bi-directional long short-term memory (LSTM) neural network, which are applied to infer pedestrian behaviour from motion tracks. We show that, though the neural network model achieves an accuracy of 80%, it requires long sequences to achieve this (100 samples or more). The SLDS, has a lower accuracy of 74%, but it achieves this result with short sequences (10 samples). To our knowledge, such a comparison on sequence length has not been considered in the literature before. The results provide a key intuition of the suitability of the models in time-critical problems.

show abstract

“…A BN has also been used to assist a surveillance system operator for maritime situational assessment [5], [6]. Expert knowledge has been used to inform the contextual and behavioral information within a scenario in which a dynamic Bayesian network was used for behavioral based classification of maritime vessels [7]. Additionally, a BN was used for the detection of unusual maritime activity where experts were used to set the network's conditional probability tables [8].…”

Section: Introductionmentioning

confidence: 99%

Imitating Radar Operator Decisions for Maritime Surveillance Missions Using Bayesian Networks

Brown

Anderson

2019

2019 International Radar Conference (RADAR)

View full text Add to dashboard Cite

This paper presents the use of Bayesian networks for learning decisions made by a human radar operator carrying out a maritime surveillance mission. By imitating the operator's decisions, a significant increase in the autonomy of a radar surveillance system can be achieved as well as potentially streamlining the qualification process. For maritime scenarios, current literature has only focused on using a Bayesian network (BN) for identification and assessment, and often assumes inputs from a generic surveillance sensor. Furthermore, in both the maritime surveillance and radar operations domain, there has been no investigation into using the operator's data in order to learn the decisions made throughout the mission. This paper uses a realtime radar simulation in order to obtain the scenario and radar information that would be observed by a human operator. In conjunction with a user interface, the simulation is further used to obtain operator decisions for a given mission with the maximum likelihood approach used to obtain the BN probabilities. The BN is then used in place of the operator for interfacing with the simulation in order to test the suitability of this method. Several typical scenarios are used to demonstrate the BN's operational ability relative to that of the operator. Additionally, the required data size for sufficient performance is investigated.

show abstract

Context-based behaviour modelling and classification of marine vessels in an abalone poaching situation

Cited by 7 publications

References 25 publications

Real-Time Maritime Traffic Anomaly Detection Based on Sensors and History Data Embedding

Real-Time Maritime Traffic Anomaly Detection Based on Sensors and History Data Embedding

Deep Learning and Statistical Models for Time-Critical Pedestrian Behaviour Prediction

Imitating Radar Operator Decisions for Maritime Surveillance Missions Using Bayesian Networks

Contact Info

Product

Resources

About