Intelligent deep fusion network for anomaly identification in maritime transportation systems

Djenouri, Youcef; Belhadi, Asma; Djenouri, Djamel; Srivastava, Gautam; Lin, Jerry Chun-Wei

Abstract

This paper introduces a novel deep learning architecture for identifying outliers in the context of intelligent transportation systems. The use of a convolutional neural network with decomposition is explored to find abnormal behavior in maritime data. The set of maritime data is first decomposed into similar clusters containing homogeneous data, and then a convolutional neural network is used for each data cluster. Different models are trained (one per cluster), and each model is learned from highly correlated data. Finally, the results of the models are merged using a simple but efficient fusion strategy. To verify the performance of the proposed framework, intensive experiments were conducted on marine data. The results show the superiority of the proposed framework compared to the baseline solutions in terms of several accuracy metrics.

• Journal Article Type: Article
• Acceptance Date: Feb 24, 2022
• Online Publication Date: Feb 24, 2022
• Publication Date: Feb 28, 2023
• Deposit Date: Mar 3, 2022
• Publicly Available Date: Mar 9, 2022
• Journal: IEEE Transactions on Intelligent Transportation Systems
• Print ISSN: 1524-9050
• Electronic ISSN: 1558-0016
• Publisher: Institute of Electrical and Electronics Engineers
• Peer Reviewed: Peer Reviewed
• Volume: 24
• Issue: 2
• Pages: 2392-2400
• DOI: https://doi.org/10.1109/TITS.2022.3151490
• Keywords: Computer Science Applications; Mechanical Engineering; Automotive Engineering
• Public URL: https://uwe-repository.worktribe.com/output/9096794
• Publisher URL: https://ieeexplore.ieee.org/document/9721146