Publications

You can also find my articles on my Google Scholar profile.

A ‘Cluster-then-Estimate’ Natural Language Processing (NLP) Approach for Classifying Maritime Incident Severity Based on Textual Descriptions

Published in Accident Analysis & Prevention, 2026

Textual incident description is a vital source for understanding the severity of maritime incident. In the maritime industry, relevant authorities and companies typically rely on manual methods to estimate incident severity based on textual descriptions. However, manual estimation is less efficient for assessing vessels’ operational risk or managing historical incident archives, where a large volume of incidents is involved. Therefore, this study proposes a ‘cluster-then-estimate’ approach which uses Natural Language Processing (NLP) techniques to automatically estimate the severity level of incidents based upon their textual descriptions. In the proposed approach, Latent Dirichlet Allocation (LDA) is used to group the preprocessed textual descriptions into multiple clusters, with each cluster representing an incident type. Then, Bidirectional Encoder Representation from Transformers (BERT) model is fine-tuned for each …

Recommended citation: T Chen, M Liang, WS Lee, Y Cai, Q Meng. (2026). A 'Cluster-then-Estimate' Natural Language Processing (NLP) Approach for Classifying Maritime Incident Severity Based on Textual Descriptions. Accident Analysis & Prevention, 228, 108413. https://www.sciencedirect.com/science/article/pii/S0001457526000229

Realistic Curriculum Reinforcement Learning for Autonomous and Sustainable Marine Vessel Navigation

Published in Proceedings of the AAAI Conference on Artificial Intelligence, 2026

Sustainability is becoming increasingly critical in the maritime transport, encompassing both environmental and social impacts, such as Greenhouse Gas (GHG) emissions and navigational safety. Traditional vessel navigation heavily relies on human experience, often lacking autonomy and emission awareness, and is prone to human errors that may compromise safety. In this paper, we propose a Curriculum Reinforcement Learning (CRL) framework integrated with a realistic, data-driven marine simulation environment and a machine learning-based fuel consumption prediction module. The simulation environment is constructed using real-world vessel movement data and enhanced with a Diffusion Model to simulate dynamic maritime conditions. Vessel fuel consumption is estimated using historical operational data and learning-based regression. The surrounding environment is represented as image-based inputs to capture spatial complexity. We design a lightweight, policy-based CRL agent with a comprehensive reward mechanism that considers safety, emissions, timeliness, and goal completion. This framework effectively handles complex tasks progressively while ensuring stable and efficient learning in continuous action spaces. We validate the proposed approach in a sea area of the Indian Ocean, demonstrating its efficacy in enabling sustainable and safe vessel navigation.

Recommended citation: X Zhang, Z Xiao, M Liang, T Liu, H Li, W Zhang. (2026). Realistic Curriculum Reinforcement Learning for Autonomous and Sustainable Marine Vessel Navigation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(46). https://ojs.aaai.org/index.php/AAAI/article/view/41313

An attention-enhanced deep learning model for detecting vessel anomalous behavior

Published in Ocean Engineering, 2026

Intelligent surveillance is critical for ensuring the safety and efficiency of maritime traffic, especially in busy waterways where the volume of vessel traffic presents substantial monitoring challenges. To effectively detect vessel anomalies and minimize potential risks, we propose a novel deep learning model called Seq2Seq-Attention for Vessel Anomalous Behavior Detection (SA-VABD). Specifically, the speed-aware Douglas-Peucker algorithm is employed to compress trajectory data to reduce data complexity and improve processing efficiency. This compression allows us to extract high-level attributes of vessel trajectories more effectively. Building on these attributes, we present a seq2seq-driven model designed to capture both static and dynamic characteristics of vessels. To further improve the model’s detection capability, an attention mechanism is embedded into the Seq2Seq network. This allows the model to …

Recommended citation: C Xu, S Zhou, M Liang, Y Zhang, M Zhang, RW Liu. (2026). An attention-enhanced deep learning model for detecting vessel anomalous behavior. Ocean Engineering, 343, 123239. https://www.sciencedirect.com/science/article/pii/S0029801825029221

Big Data Fusion-Driven Geospatial Knowledge Graph Construction Method for Sustainable Smart Cities

Published in Sustainable Cities and Society, 2026

Urban planning faces increasing challenges in integrating and analyzing multi-source geospatial data due to inconsistencies in spatial resolution, data latency, and processing inefficiency. Traditional geographic information systems (GIS) and remote sensing models typically rely on a single data source, limiting their ability to deliver accurate and comprehensive insights for smart city development. This paper proposes a Big Data Fusion-Driven Geospatial Knowledge Graph framework (BDF-GeoKG) to address these limitations by integrating vector, raster, text, and image data. The proposed framework follows a structured process of entity extraction, relationship construction, attribute extraction, and entity alignment to establish a unified geospatial knowledge graph. Entity extraction identifies geographic objects and attributes from multi-source data. Relationship construction defines spatial and semantic connections …

Recommended citation: Y Duan, M Liang*, Y Li, R Gao, J Chen, ZS Chen, H Wang. (2026). Big Data Fusion-Driven Geospatial Knowledge Graph Construction Method for Sustainable Smart Cities. Sustainable Cities and Society. https://www.sciencedirect.com/science/article/pii/S2210670725008972

A fine-grained predictive optimization framework for dynamic eco-routing of electric vehicles

Published in Computers & Industrial Engineering, 2026

As electric vehicle (EV) adoption grows globally, optimizing energy consumption during travel is essential for extending range and improving urban sustainability. This paper presents a novel eco-routing framework, Predictive Optimization for Eco-Routing of Electric Vehicles (POE-EV), which is designed with a two-step approach: first, a deep learning model is developed to predict fine-grained energy consumption across road segments, accounting for complex spatiotemporal driving conditions. Next, Dijkstra’s algorithm generates high-quality initial routes based on these predictions. These routes are refined through a Genetic Algorithm (GA), optimizing the balance between minimizing travel time and reducing energy consumption. This sequential process of prediction followed by optimization provides key advantages, with Dijkstra’s algorithm offering efficient path generation and GA exploring multiple solutions for …

Recommended citation: Q Liu, Y Li, M Liang*, R Gao, J Zhang, Y Liu. (2026). A fine-grained predictive optimization framework for dynamic eco-routing of electric vehicles. Computers & Industrial Engineering, 111662. https://www.sciencedirect.com/science/article/pii/S0360835225008083

A big data-driven meta classifier for cross-sectional classification of vessel incident risk

Published in Ocean Engineering, 2025

Major vessel incidents, such as collisions, groundings, and oil spills, impose significant financial losses on shipping stakeholders. To assess the risk of future incidents, this study proposes a big data-driven classification approach to identify incident-prone vessels. Based on a comprehensive global vessel intelligence dataset (2017–2022), we develop four base machine learning classifiers and a meta classifier with hybrid feature selection. The meta classifier validated on out-of-sample data in year 2022, demonstrates superior classification precision and sorting ability, outperforming state-of-the-art models and an industry-adopted vehicle incident risk assessment method. The proposed solution provides an effective tool for stakeholders to evaluate vessel incident risk levels, aiding decisions on vessel purchases, charters, marine insurance premium design, and risk control. By offering a practical and scalable …

Recommended citation: Y Cai, T Chen, M Liang, Q Meng. (2025). A big data-driven meta classifier for cross-sectional classification of vessel incident risk. Ocean Engineering, 342, 123058. https://www.sciencedirect.com/science/article/pii/S0029801825027416

Identifying Acute Thoracolumbar Vertebral Compression Fractures From Low-Quality Small-Sample X-Ray Images: A Transfer Learning-Based Approach

Published in IEEE Journal of Biomedical and Health Informatics, 2025

Timely and accurate diagnosis of acute thoracolumbar vertebral compression fractures in X-ray images is critical for initiating prompt and effective treatment, preventing potential neurological damage and long-term disability. Recent advancements in artificial intelligence (AI) have significantly improved medical imaging analysis, providing sophisticated tools to assist clinicians in diagnosing acute thoracolumbar vertebral compression fractures. Nonetheless, detecting these fractures through imaging remains challenging due to the complex overlapping of bony structures in the thoracolumbar region, variability in fracture patterns, and often subtle nature of these injuries. Additionally, the limited availability and sometimes poor quality of medical images further complicate accurate AI-based detection. Addressing these challenges, this study introduces a transfer learning model optimized for recognizing acute …

Recommended citation: Y Wang, W Li, S Chen, Y Yang, A Fan, C Lei, Y Kou, N Han, F Xue, et al. (2025). Identifying Acute Thoracolumbar Vertebral Compression Fractures From Low-Quality Small-Sample X-Ray Images: A Transfer Learning-Based Approach. IEEE Journal of Biomedical and Health Informatics. https://ieeexplore.ieee.org/abstract/document/11265857/

Ship anomalous behavior detection based on interval prediction of multiple vessel trajectories

Published in Engineering Applications of Artificial Intelligence, 2025

In response to the increasing complexity of maritime traffic situations, effective maritime surveillance is essential to ensure the safety of maritime activities. This paper leverages Automatic Identification Systems (AIS) data in conjunction with deep learning techniques and statistical methods to achieve adaptive matching and prediction of multiple vessel trajectories, as well as the identification of anomalous vessel behaviors. Initially, an improved Hausdorff distance is employed to accommodate preprocessed trajectory data, and an enhanced Density-Based Spatial Clustering of Applications with Noise(DBSCAN) algorithm is utilized for cluster analysis, facilitating adaptive multi-vessel trajectory matching. Subsequently, considering the inter-vessel influence zones, a multi-vessel behavior feature prediction is realized based on the completed adaptive vessel trajectory matching. Furthermore, distribution characteristics of …

Recommended citation: W Liu, C Chen, Y Peng, M Liang. (2025). Ship anomalous behavior detection based on interval prediction of multiple vessel trajectories. Engineering Applications of Artificial Intelligence, 160, 111693. https://www.sciencedirect.com/science/article/pii/S0952197625016951

UPTM-LLM: Large Language Models-Powered Urban Pedestrian Travel Modes Recognition for Intelligent Transportation System

Published in Applied Soft Computing, 2025

With the progression of urbanization and information technology, Intelligent Transportation Systems (ITS) are increasingly crucial for improving urban efficiency and reducing traffic congestion. A key element within ITS is Urban Pedestrian Travel Mode (UPTM) recognition, which aids traffic management through data-driven optimization and congestion mitigation. Deep learning (DL) has advanced transportation research but struggles with the complex semantics of pedestrian movement, including temporal context (e.g., peak/off-peak hours, workdays/weekends), origin-destination POI categories, and the underlying characteristics of GPS trajectories (encompassing kinematics and GIS features). The emergence of large language models (LLMs), known for their large-scale parameters and deep architectures, has enhanced the capacity to interpret such complex semantics. Leveraging this capability, we propose UPTM …

Recommended citation: Y Li, Y Zhan, M Liang*, Y Zhang, J Liang. (2025). UPTM-LLM: Large Language Models-Powered Urban Pedestrian Travel Modes Recognition for Intelligent Transportation System. Applied Soft Computing, 113999. https://www.sciencedirect.com/science/article/pii/S1568494625013122

Reliable vessel trajectory clustering: A maritime shipping network-driven computational method

Published in Ocean Engineering, 2025

Vessel trajectory clustering has attracted growing attention in maritime applications, such as anomaly detection, route planning, behavior prediction and situational awareness, etc. However, the clustering results often suffer from low efficiency and robustness, especially for the massive vessel trajectories. To achieve satisfactory clustering results, this work proposes a maritime shipping network-driven vessel trajectory clustering method, highly dependent on the accurate and robust extraction of network nodes (i.e., waypoints). We first exploit the sub-trajectory-sensitive Douglas-Peucker (DP) and Density Peak Clustering (DPC) algorithms, respectively, to identify the turning and crossing feature points. The whole raw feature points are then obtained by combining these turning and crossing feature points with the starting and ending points of vessel trajectories. The final waypoints can be extracted accordingly by …

Recommended citation: C Xu, S Zhou, M Liang, Z Liu, RW Liu. (2025). Reliable vessel trajectory clustering: A maritime shipping network-driven computational method. Ocean Engineering, 336, 121691. https://www.sciencedirect.com/science/article/pii/S0029801825013976

A Dynamic Ensemble Deep Randomized Neural Network Using Deep Autoregressive Features for Wave Height Forecasting With Missing Values

Published in IEEE Journal of Oceanic Engineering, 2025

Wave energy is an essential part of sustainable energy. Precise forecasts of wave height assist in the reliable control of wave energy converters and the intelligent operation of electricity generation. However, the severe and extreme environment poses a significant challenge for accurate sensor recording, resulting in a huge number of missing values at random. The missing values exist in multiple explanatory variables, significantly deteriorating the performance of the classical machine learning models. This article aims to enhance the accuracy of significant wave height forecasting with data imperfections by proposing a flexible dynamic ensemble framework and an ensemble deep randomized neural network. First, the proposed dynamic ensemble framework disaggregates the whole forecasting task into multiple subtasks based on the number of missing values. For each subtask, any missing values imputation …

Recommended citation: R Gao, S Yang, M Yuan, Z Wang, M Liang*, PN Suganthan, KF Yuen. (2025). A Dynamic Ensemble Deep Randomized Neural Network Using Deep Autoregressive Features for Wave Height Forecasting With Missing Values. IEEE Journal of Oceanic Engineering. https://ieeexplore.ieee.org/abstract/document/11131328/

Natural language processing and text mining in transportation: Current status, challenges, and future roadmap

Published in Expert Systems with Applications, 2025

The transportation sector is generating and accumulating an increasing amount of unstructured data from a variety of sources. As a result, Natural Language Processing (NLP) and text mining are becoming critical for their capability to automatically process and interpret human language in transportation research. However, there is a lack of a thorough review of existing research in this field, as well as a detailed guide on how to use these techniques in transportation studies. This paper offers an updated review of NLP and text mining techniques, including the latest developments in Large Language Models (LLMs), tailored for comprehensive transportation modeling across land, maritime, and aviation sectors. It highlights the data sources and methodologies used in previous studies, provides an analysis of word representation, sentiment analysis, external NLP toolkits, language diversity, and performance …

Recommended citation: X Zhang, R Gao, Z Xiao, K Wang, T Liu, M Liang, J Zhang. (2025). Natural language processing and text mining in transportation: Current status, challenges, and future roadmap. Expert Systems with Applications, 129050. https://www.sciencedirect.com/science/article/pii/S0957417425026673

Spatial-Frequency Fusion Network With Learnable Fractional Fourier Transform for Remote Sensing Imaging Enhancement

Published in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2025

Atmospheric haze significantly degrades the quality of remote sensing images, reducing visibility, distorting spectral information, and impairing downstream tasks such as land cover classification and infrastructure layout analysis. To overcome these challenges, this article proposes a novel spatial–frequency fusion network (termed SFFNet) with a learnable fractional Fourier transform for efficient remote sensing imaging enhancement. In the spatial domain, the SFFNet uses a multiscale spatial pyramid pooling block to capture both fine-grained details and global contextual information, while residual connections ensure robust feature learning and spatial detail preservation. In the frequency domain, a self-learned fractional Fourier transform module adaptively extracts haze-relevant features, leveraging a learnable parameter to dynamically adjust the fractional order of the transform. Furthermore, an attentive …

Recommended citation: W Xu, M Liang, Y Lu, R Gao, D Yang. (2025). Spatial-Frequency Fusion Network With Learnable Fractional Fourier Transform for Remote Sensing Imaging Enhancement. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. https://ieeexplore.ieee.org/abstract/document/11071378/

Discontinuous Parsimony Embedding Empowered Transformer for Shipping Market Forecasting

Published in 2025 International Joint Conference on Neural Networks (IJCNN), 2025

The profitability and survival of ship-owning companies in the global shipping market are deeply intertwined with accurate forecasts of ship prices and charter rates. Effective detection of market shortfalls and capitalization on temporal arbitrage opportunities are essential for maintaining a competitive edge. Traditional forecasting models, while adept at handling various multivariate time series tasks, predominantly focus on embedding synchronous time lags, often neglecting asynchronous dependencies. This paper introduces the Shipping Transformer (SFormer), a novel forecasting model designed to address this gap by integrating a discontinuous and parsimonious embedding strategy. This approach effectively captures lead-lag relationships between explanatory and target series. To further enhance forecasting performance, we introduce a cross-dimension attention module that uncovers cross-series …

Recommended citation: R Gao, Z Wang, M Hu, M Liang, PN Suganthan. (2025). Discontinuous Parsimony Embedding Empowered Transformer for Shipping Market Forecasting. 2025 International Joint Conference on Neural Networks (IJCNN), 1-8. https://ieeexplore.ieee.org/abstract/document/11229012/

Data-driven impact analysis of chokepoint on multi-scale maritime networks: A case study of the Taiwan Strait

Published in Transportation Research Part E: Logistics and Transportation Review, 2025

Maritime chokepoints are crucial nodes in international supply chains, facilitating most of the global freight transport and the distribution of industrial raw materials and goods. However, these chokepoints are vulnerable to disruptions from severe weather, traffic accidents, and unpredictable factors, significantly impacting global maritime networks. Therefore, it is crucial to understand the role of chokepoints to improve the robustness of the global maritime network. This paper aims to assess the influence of a crucial maritime chokepoint, with a specific focus on the Taiwan Strait, and its effects on the global maritime network. Specifically, five maritime networks with different scales are constructed based on massive global Automatic Identification System (AIS) data. The Taiwan Strait is selected as an important maritime chokepoint in the multi-scale networks. Then, the dependence of the Taiwan Strait on various ports …

Recommended citation: M Liang, Y Cai, T Chen, H Wang, Q Meng. (2025). Data-driven impact analysis of chokepoint on multi-scale maritime networks: A case study of the Taiwan Strait. Transportation Research Part E: Logistics and Transportation Review. https://www.sciencedirect.com/science/article/pii/S1366554525002200

Big-Data-Driven Vessel Destination Prediction for Smart Port Management

Published in Engineering Applications of Artificial Intelligence, 2025

The accurate prediction of vessel destinations is crucial for enhancing maritime traffic efficiency, optimizing port management, and improving regional economic analysis. However, destination information in Automatic Identification System (AIS) data is often missing or inaccurate, which undermines the reliability of maritime analytic. Traditional vessel destination prediction methods primarily focus on measuring trajectory similarities, which results in high computational complexity. This study develops a deep learning approach to vessel destination prediction by transforming the problem into an image classification task. Rasterized images of historical destination ports and vessel trajectories are generated, incorporating AIS data within a fixed spatial context. A multi-scale residual convolutional network is constructed to extract relevant trajectory and port distribution features. To enhance the representation of trajectory …

Recommended citation: J Chen, Q Zhang, M Liang*, C Peng, C Chen. (2025). Big-Data-Driven Vessel Destination Prediction for Smart Port Management. Engineering Applications of Artificial Intelligence. https://www.sciencedirect.com/science/article/pii/S0952197625008292

A graph attention network-based learning framework for automatic detection of abnormal vessel behaviors

Published in Ocean Engineering, 2025

With the rapid expansion of maritime activities, the need to detect abnormal vessel behaviors using advanced data-driven methods has become increasingly critical for ensuring maritime safety and efficiency. Existing approaches often overlook the temporal dependencies and feature correlations in vessel behaviors. This limitation reduces their ability to capture the complexities of maritime operations. To address these challenges, we propose GAT-AD, a novel graph attention network-based framework for anomaly detection in vessel behavior. Our framework incorporates three key components: (1) a graph attention module that combines temporal and feature attention to capture sequential and feature dependencies, (2) an embedding layer to extract latent information from vessel data, enhancing representation learning, and (3) a joint detection module that calculates anomaly scores using both reconstruction-based …

Recommended citation: M Liang, Y Zhang, Q Jin, W Liu. (2025). A graph attention network-based learning framework for automatic detection of abnormal vessel behaviors. Ocean Engineering. https://www.sciencedirect.com/science/article/pii/S0029801825004159

Improving Maritime Data: A Machine Learning-based Model for Missing Vessel Trajectories Reconstruction

Published in IEEE Transactions on Vehicular Technology, 2025

Advancements in maritime satellite technology have significantly impacted the maritime industry, enhancing both communication and safety at sea. These technological improvements have enabled Automatic Identification Systems (AIS) to transmit data through robust maritime communication networks. However, despite these advancements, AIS data often contain significant missing data due to limitations in both devices and network coverage. To overcome these limitations, this research presents an innovative approach for reconstructing missing points. In this paper, we first extracted the geometric shape and motion characteristics of vessels and constructed a decision tree using an adaptive sparse constraint mechanism to classify four types of vessel trajectories. Then, the vessel classification findings serve as input and the vessel acceleration features are constructed using the heading and velocity features of the …

Recommended citation: J Chen, M Liang*, C Pang, J Zhang, S Huo. (2025). Improving Maritime Data: A Machine Learning-based Model for Missing Vessel Trajectories Reconstruction. IEEE Transactions on Vehicular Technology. https://ieeexplore.ieee.org/abstract/document/10878129/

Underwater acoustic signal denoising algorithms: A survey of the state-of-the-art

Published in IEEE Transactions on Instrumentation and Measurement, 2025

Underwater acoustic signal (UAS) denoising is crucial for enhancing the reliability of underwater communication and monitoring systems by mitigating the effects of noise and improving signal clarity. The complex and dynamic nature of underwater environments presents unique challenges that make effective denoising essential for accurate data interpretation and system performance. This article comprehensively reviews recent advances in UAS denoising, focusing on its critical role in improving these systems. The review begins by addressing the fundamental challenges in UAS processing, such as signal attenuation, noise variability, and environmental impacts. It then categorizes and analyzes various denoising algorithms, including conventional, decomposition-based, and learning-based approaches, discussing their applications, strengths, and limitations. Additionally, the article reviews evaluation metrics …

Recommended citation: R Gao, M Liang*, H Dong, X Luo, PN Suganthan. (2025). Underwater acoustic signal denoising algorithms: A survey of the state-of-the-art. IEEE Transactions on Instrumentation and Measurement. https://ieeexplore.ieee.org/abstract/document/10935817/

Multi-Frequency Spatial-Temporal Graph Neural Network for Short-Term Metro OD Demand Prediction during Public Health Emergencies

Published in Transportation, 2025

Short-term metro OD demand prediction during public health emergencies is a crucial task for the effective management and operation of metro systems. However, such emergencies tend to cause significant fluctuations in OD demand, making accurate prediction particularly challenging. To tackle this problem, this paper proposes a Multi-Frequency Spatial-Temporal Graph Neural Network (MFST-GNN) to accurately predict the metro OD demand during public health emergencies. Specifically, multiple OD demand patterns, including real-time, daily, and weekly OD demand are leveraged to extract the periodicity spatial-temporal features of OD demand. A novel multi-frequency temporal feature extraction module is developed to capture the periodic temporal features, while an adaptive spatial feature extraction module is introduced to learn the complex hidden spatial features. Moreover, event-related information is …

Recommended citation: J Zhang, S Zhang, H Zhao, Y Yang, M Liang*. (2025). Multi-Frequency Spatial-Temporal Graph Neural Network for Short-Term Metro OD Demand Prediction during Public Health Emergencies. Transportation. https://link.springer.com/article/10.1007/s11116-025-10582-0

Deep-TCP: Multi-source data fusion for deep learning-powered tropical cyclone intensity prediction to enhance urban sustainability

Published in Information Fusion, 2025

Tropical cyclones (TC) exert a profound impact on cities, causing extensive damage and losses. Thus, TC Intensity Prediction is crucial for creating sustainable cities as it enables proactive measures to be taken, including evacuation planning, infrastructure reinforcement, and emergency response coordination. In this study, we propose a Deep learning-powered TC Intensity Prediction (Deep-TCP) framework. In particular, Deep-TCP contains a data constraint module for fusing data features from multiple sources and establishing a unified global representation. To capture the spatiotemporal attributes, a Spatial-Temporal Attention (ST-Attention) module is built to distill insights from environmental variables. To improve the robustness and stability of the predictions, an encoder-decoder module that utilizes the ConvGPU unit is introduced to enhance feature maps. Then, a novel feature enhancement module is built to …

Recommended citation: S Jiang, M Liang*, C Wang, H Fan, Y Ma. (2025). Deep-TCP: Multi-source data fusion for deep learning-powered tropical cyclone intensity prediction to enhance urban sustainability. Information Fusion, 114, 102670. https://www.sciencedirect.com/science/article/pii/S1566253524004482

Wave energy forecasting: A state-of-the-art survey and a comprehensive evaluation

Published in Applied Soft Computing, 2025

Wave energy, a promising renewable energy source, has the potential to diversify the global energy mix significantly. Accurate forecasting of significant wave height (SWH) is crucial for enhancing the efficiency and reliability of wave energy conversion systems. As interest in this field grows, research into SWH forecasting has expanded dramatically. This comprehensive survey evaluates sixteen SWH forecasting methods, including Persistence, decision trees, deep neural networks, random neural networks, and random forests. The paper begins by establishing a detailed taxonomy that categorizes SWH forecasting algorithms, providing a framework to interpret the complexities of different methodological approaches. We then explore the interconnections between ensemble learning and decomposition-based frameworks and the integration of individual forecasting techniques within ensemble and hybrid models. In …

Recommended citation: R Gao, X Zhang, M Liang*, PN Suganthan, H Dong. (2025). Wave energy forecasting: A state-of-the-art survey and a comprehensive evaluation. Applied Soft Computing, 112652. https://www.sciencedirect.com/science/article/pii/S1568494624014261

Integrating GPU-Accelerated for Fast Large-Scale Vessel Trajectories Visualization in Maritime IoT Systems

Published in IEEE Transactions on Intelligent Transportation Systems, 2025

With the advancement of satellite communication technology, the maritime Internet of Things (IoT) has made significant progress. As a result, vast amounts of Automatic Identification System (AIS) data from global vessels are transmitted to various maritime stakeholders through Maritime IoT systems. AIS data contains a large amount of dynamic and static information that requires effective and intuitive visualization for comprehensive analysis. However, two major deficiencies challenge current visualization models: a lack of consideration for interactions between distant pixels and low efficiency. To address these issues, we developed a large-scale vessel trajectories visualization algorithm, called the Non-local Kernel Density Estimation (NLKDE) algorithm, which incorporates a non-local convolution process. It accurately calculates the density distribution of vessel trajectories by considering correlations between …

Recommended citation: M Liang, K Liu, R Gao, Y Li. (2025). Integrating GPU-Accelerated for Fast Large-Scale Vessel Trajectories Visualization in Maritime IoT Systems. IEEE Transactions on Intelligent Transportation Systems, 26(3), 4048-4065. https://ieeexplore.ieee.org/abstract/document/10824219/

Multi-Sensor Fusion-Driven Surface Vessel Identification and Tracking Using Unmanned Aerial Vehicles for Maritime Surveillance

Published in IEEE Transactions on Consumer Electronics, 2025

The rapidly-developing artificial intelligence and edge computing technologies have been actively promoting the evolution of autonomous vehicles. The flexible and powerful unmanned aerial vehicles (UAVs) have become suitable application platforms for maritime surveillance. In the maritime Internet of Things (IoT), both airborne cameras and automatic identification system (AIS), which, respectively, provide visual and positioning data, have become the frequently-used and cost-effective sensors. It becomes necessary to fuse the AIS and visual data to detect the visual appearances of surface vessels with obtaining the abundant information on position, movements, and identity, etc. In this work, we propose to develop a multi-sensor fusion-driven computational method (termed MSF-VIT) for highly-reliable surface vessel identification and tracking under different conditions. In particular, it mainly consists of two …

Recommended citation: C Zhao, M Bao, J Liang, M Liang, RW Liu, G Pang. (2025). Multi-Sensor Fusion-Driven Surface Vessel Identification and Tracking Using Unmanned Aerial Vehicles for Maritime Surveillance. IEEE Transactions on Consumer Electronics. https://ieeexplore.ieee.org/abstract/document/10833660/

Exploring key factors for long-term vessel incident risk prediction

Published in Reliability Engineering & System Safety, 2025

Factor analysis acts a pivotal role in enhancing maritime safety. Most previous studies conduct factor analysis within the framework of incident-related label prediction, where the developed models can be categorized into short-term and long-term prediction models. The long-term models offer a more strategic approach, enabling more proactive risk management, compared to the short-term ones. Nevertheless, few studies have devoted to rigorously identifying the key factors for the long-term prediction and undertaking comprehensive factor analysis. Hence, this study aims to delve into the key factors for predicting the incident risk levels in the subsequent year given a specific datestamp. The majority of candidate factors potentially contributing to the incident risk are collected from vessels’ historical safety performance data spanning up to five years. An improved embedded feature selection method, which integrates …

Recommended citation: T Chen, H Wang, Y Cai, M Liang, Q Meng. (2025). Exploring key factors for long-term vessel incident risk prediction. Reliability Engineering & System Safety, 253, 110565. https://www.sciencedirect.com/science/article/pii/S0951832024006379

Estimation of vessel link-level travel time distribution: A directed network-driven approach

Published in Ocean Engineering, 2024

Accurate vessel travel time estimation is essential for operational efficiency and route optimization. Despite the prevalent use of the automatic identification system (AIS) in garnering multifaceted real-time data, ambiguities and inaccuracies in travel time predictions persist. It leads to planning uncertainties, inefficient resource allocations, and heightened operational costs in maritime logistics. To addresses these issues, this paper proposed a nuanced method to enhance the precision and reliability of vessel travel time estimations. Firstly, a directed maritime network is constructed by extracting essential information from AIS-based historical vessel trajectories. This lays the foundation for the subsequent analytical processes. Secondly, the non-parametric kernel density estimation (KDE) is applied to this constructed network, enabling the estimation of vessel travel time distributions across various network links. The non …

Recommended citation: M Liang, J Su, R Gao, RW Liu, Y Zhan. (2025). Estimation of vessel link-level travel time distribution: A directed network-driven approach. Ocean Engineering, 313, 119371. https://www.sciencedirect.com/science/article/pii/S0029801824027094

From ports to routes: Extracting multi-scale shipping networks using massive AIS data

Published in Ocean Engineering, 2024

Maritime transportation is a critical component of global trade and commerce. To ensure maritime safety, fixed shipping routing has been established in many complex waters. However, there is currently a lack of comprehensive digital shipping networks in wide-range maritime areas. To better understand the navigational patterns, this paper proposes a data-driven extraction framework for multi-scale shipping networks, including port-, node-, and route-level shipping networks. It is essentially a hierarchical approach, which progresses from port to route. In particular, for the extraction of port-level shipping networks, the clustering in quest (CLIQUE) and alpha-shapes algorithms are employed to accurately extract the boundaries and spatial extents of individual ports. For the node-level shipping network extraction, an adaptive Douglas-Peucker algorithm is developed to identify crucial feature points, and CLIQUE clustering is further exploited to extract the network waypoints. A novel slice-based traffic flow fitting algorithm is finally introduced to extract the route-level shipping network. To verify the performance of shipping network extraction methods, comprehensive experiments are conducted using the massive Automatic Identification System (AIS) data in different water areas. The experimental results have demonstrated that our method was capable of extracting multi-scale shipping networks, revealing traffic characteristics and vessel behaviours. Overall, the method proposed herein is useful for shipping logistic analysis and provides a foundation for several potential maritime applications, including route planning, trajectory prediction, and others.

Recommended citation: Liu, R. W., Zhou, S., Liang, M., Gao, R., & Wang, H. (2024). From ports to routes: Extracting multi-scale shipping networks using massive AIS data. Ocean Engineering, 311, 118969. https://doi.org/10.1016/j.oceaneng.2024.118969

AISClean: AIS data-driven vessel trajectory reconstruction under uncertain conditions

Published in Ocean Engineering, 2024

In maritime transportation, intelligent vessel surveillance has become increasingly prevalent and widespread by collecting and analyzing high massive spatial data from automatic identification system (AIS). The state-of-the-art AIS devices contain various functionalities, such as position transmission, tracking navigation, etc. Widely equipped shipboard AIS devices provide a large amount of real-time and historical vessel trajectory data for maritime management. However, the original AIS data often suffers from unwanted noise (i.e., poorly tracked timestamped points for vessel trajectories) and missing (i.e., no data is received or transmitted for a long term) data during signal acquisition, transmission, and analog-to-digital conversion. This degradation in data quality poses significant risks, including potential miscalculations in vessel collision avoidance systems, inaccuracies in emission calculations, and challenges in

Recommended citation: Liang, M., Su, J., Liu, R. W., & Lam, J. S. L. (2024). AISClean: AIS data-driven vessel trajectory reconstruction under uncertain conditions. Ocean Engineering, 306, 117987. https://doi.org/10.1016/j.oceaneng.2024.117987

AIS-based vessel trajectory compression: a systematic review and software development

Published in IEEE Open Journal of Vehicular Technology, 2024

With the advancement of satellite and 5G communication technologies, vehicles can transmit and exchange data from anywhere in the world. It has resulted in the generation of massive spatial trajectories, particularly from the Automatic Identification System (AIS) for surface vehicles. The massive AIS data lead to high storage requirements and computing costs, as well as low data transmission efficiency. These challenges highlight the critical importance of vessel trajectory compression for surface vehicles. However, the complexity and diversity of vessel trajectories and behaviors make trajectory compression imperative and challenging in maritime applications. Therefore, trajectory compression has been one of the hot spots in research on trajectory data mining. The major purpose of this work is to provide a comprehensive reference source for beginners involved in vessel trajectory compression. The current trajectory compression methods could be broadly divided into two types, batch (offline) and online modes. The principles and pseudo-codes of these methods will be provided and discussed in detail. In addition, compressive experiments on several publicly available data sets have been implemented to evaluate the batch and online compression methods in terms of computation time, compression ratio, trajectory similarity, and trajectory length loss rate. Finally, we develop a flexible and open software, called AISCompress, for AIS-based batch and online vessel trajectory compression. The conclusions and associated future works are also given to inspire future applications in vessel trajectory compression.

Recommended citation: Liu, R. W., Zhou, S., Yin, S., Shu, Y., & Liang, M. (2024). AIS-based vessel trajectory compression: a systematic review and software development. IEEE Open Journal of Vehicular Technology, 5, 1193-1214. https://doi.org/10.1109/OJVT.2024.3443675

A survey of distance-based vessel trajectory clustering: Data pre-processing, methodologies, applications, and experimental evaluation

Published in arXiv, 2024

Vessel trajectory clustering, a crucial component of the maritime intelligent transportation systems, provides valuable insights for applications such as anomaly detection and trajectory prediction. This paper presents a comprehensive survey of the most prevalent distance-based vessel trajectory clustering methods, which encompass two main steps: trajectory similarity measurement and clustering. Initially, we conducted a thorough literature review using relevant keywords to gather and summarize pertinent research papers and datasets. Then, this paper discussed the principal methods of data pre-processing that prepare data for further analysis. The survey progresses to detail the leading algorithms for measuring vessel trajectory similarity and the main clustering techniques used in the field today. Furthermore, the various applications of trajectory clustering within the maritime context are explored. Finally, the paper evaluates the effectiveness of different algorithm combinations and pre-processing methods through experimental analysis, focusing on their impact on the performance of distance-based trajectory clustering algorithms. The experimental results demonstrate the effectiveness of various trajectory clustering algorithms and notably highlight the significant improvements that trajectory compression techniques contribute to the efficiency and accuracy of trajectory clustering. This comprehensive approach ensures a deep understanding of current capabilities and future directions in vessel trajectory clustering.

Recommended citation: Liang, M., Liu, R. W., Gao, R., Xiao, Z., Zhang, X., & Wang, H. (2024). A survey of distance-based vessel trajectory clustering: Data pre-processing, methodologies, applications, and experimental evaluation. arXiv preprint arXiv:2407.11084. https://arxiv.org/abs/2407.11084

Multi-ship encounter situation graph structure learning for ship collision avoidance based on AIS big data with spatio-temporal edge and node attention graph convolutional networks

Published in Ocean Engineering, 2024

With the increasing number of ships on the sea, the frequency multi-ship encounters situation was becoming more common than two-ship encounter. The complexity and risk of the navigation will exponentially increase with the more ships involved. Relying solely on the International Regulations for Preventing Collisions at Sea (COLREGS) (objective knowledge) was insufficient to handle the multi-ship intelligent collision avoidance problem, also needed the ship officer's good seamanship (subjective knowledge). In this study, we propose a methodology that combines subjective insights from AIS big data with objective analysis through multi-ship encounters recognition with graph convolutional networks (GCN). (1) The ship encounter 8-azimuths map was utilized to identify the two-ship encounter situation (25 types) from the AIS data. (2) Identify the multi-ship encounters trajectory data by cross-matching the two

Recommended citation: Gao, M., Liang, M., Zhang, A., Hu, Y., & Zhu, J. (2024). Multi-ship encounter situation graph structure learning for ship collision avoidance based on AIS big data with spatio-temporal edge and node attention graph convolutional networks. Ocean Engineering, 301, 117605. https://doi.org/10.1016/j.oceaneng.2024.117605

Spatio-Temporal Multi-Graph Transformer Network for Joint Prediction of Multiple Vessel Trajectories

Published in Engineering Applications of Artificial Intelligence, 2024

The vessel trajectory prediction plays a vital role in guaranteeing traffic safety for unmanned surface vehicles and autonomous surface vessels. By leveraging advanced satellite communication technology, AIS provides massive vessel trajectories, significantly enhancing maritime safety and decision-making. This research proposes a spatio-temporal multi-graph transformer network (ST-MGT), aiming to predict multiple vessel trajectories simultaneously. This innovative model amalgamates the capabilities of graph convolutional networks (GCNs) and transformer models to proficiently address the spatial and temporal interactions amongst vessels. The ST-MGT is comprised of three crucial layers. The temporal transformer layer employs sophisticated temporal transformer and memory mechanisms to discern the intricate temporal correlations between vessel movements. The spatial multi-graph transformer layer constructs a comprehensive multi-graph representation to illuminate spatial correlations between vessels. It incorporates a spatial graph convolutional network and transformer to meticulously understand and interpret the diverse and complex spatial interactions amongst varying vessels. Lastly, the Regularized LSTM (RegLSTM) layer is implemented for predicting vessel trajectories accurately, based on the unraveled spatio-temporal patterns. Extensive and meticulous experiments reveal that our proposed ST-MGT method transcends other state-of-the-art prediction models in robustness and accuracy. The model’s capability to facilitate multi-vessel and multi-step prediction showcases its immense potential and adaptability in intricate and multifaceted navigation environments, underscoring its practical applicability and significance in enhancing maritime navigational safety.

Recommended citation: Liu, R. W., Zheng, W., & Liang, M. (2024). Spatio-temporal multi-graph transformer network for joint prediction of multiple vessel trajectories. Engineering Applications of Artificial Intelligence, 129, 107625. http://doi.org/10.1016/j.engappai.2023.107625

PiracyAnalyzer: Spatial Temporal Patterns Analysis of Global Piracy Incidents

Published in Reliability Engineering & System Safety, 2024

Maritime piracy incidents present significant threats to maritime security, resulting in material damages and jeopardizing the safety of crews. Despite the scope of the issue, existing research has not adequately explored the diverse risks and theoretical implications involved. To fill that gap, this paper aims to develop a comprehensive framework for analyzing global piracy incidents. The framework assesses risk levels and identifies patterns from spatial, temporal, and spatio-temporal dimensions, which facilitates the development of informed anti-piracy policy decisions. Firstly, the paper introduces a novel risk assessment mechanism for piracy incidents and constructs a dataset encompassing 3,716 recorded incidents from 2010 to 2021. Secondly, this study has developed a visualization and analysis framework capable of examining piracy incidents through the identification of clusters, outliers, and hot spots. Thirdly, a number of experiments are conducted on the constructed dataset to scrutinize current spatial-temporal patterns of piracy accidents. In experiments, we analyze the current trends in piracy incidents on temporal, spatial, and spatio-temporal dimensions to provide a detailed examination of piracy incidents. The paper contributes new understandings of piracy distribution and patterns, thereby enhancing the effectiveness of anti-piracy measures.

Recommended citation: Liang, M., Li, H., Liu, R. W., Lam, J. S. L., & Yang, Z. (2024). PiracyAnalyzer: Spatial temporal patterns analysis of global piracy incidents. Reliability Engineering & System Safety, 243, 109877. https://doi.org/10.1016/j.ress.2023.109877

Unsupervised Maritime Anomaly Detection for Intelligent Situational Awareness Using AIS Data

Published in Knowledge-Based Systems, 2024

With the mandatory implementation of the automatic identification system and the rapid advancement of relevant satellite communication technologies, a vast amount of vessel trajectory data has been amassed. It has catalyzed advancements in the field of maritime anomaly detection, significantly contributing to the enhancement of intelligent maritime situational awareness. However, detecting abnormal vessel trajectories from massive data is a highly challenging task that requires extensive manual effort. To address the challenge, this paper develops an unsupervised deep learning method for detecting abnormal vessel trajectories. Specifically, we first generate the informative trajectory images by remapping the raw vessel trajectories into two-dimensional matrices while maintaining the spatio-temporal properties. Then, a Wasserstein generative adversarial network with gradient penalty (WGAN-GP) is trained on normal vessel trajectories. Next, an encoder is trained to map these trajectory images to a latent space. Finally, the trained encoder and WGAN-GP are used to design an anomaly score for detecting abnormal vessel trajectories and anomalies in vessel trajectories. Experimental evaluations using multiple evaluation metrics on diverse simulated and real-world maritime environments validate the efficacy of the proposed method in unsupervised maritime anomaly detection. The results indicate that the proposed model can provide accurate anomaly detection on vessel trajectories without manual labeling.

Recommended citation: Liang, M., Weng, L., Gao, R., Li, Y., & Du, L. (2024). Unsupervised maritime anomaly detection for intelligent situational awareness using AIS data. Knowledge-Based Systems, 284, 111313. https://doi.org/10.1016/j.knosys.2023.111313

QSD-LSTM: Vessel trajectory prediction using long short-term memory with quaternion ship domain

Published in Applied Ocean Research, 2023

Vessel trajectory prediction is a critical aspect of ensuring maritime traffic safety and avoiding collisions. The long short-term memory (LSTM) network and its extensions have represented powerful ability of vessel trajectory prediction. However, the previous studies often did not take dynamic interactions between neighboring vessels into account. Additionally, in complex traffic conditions, trajectory prediction will acquire uncertainty, and these potential negative factors can limit the prediction of future trajectory. To enhance the prediction performance, we propose an interactive vessel trajectory prediction framework (i.e., QSD-LSTM) based on LSTM, which is embedded with the quaternion ship domain (QSD). The QSD is beneficial for avoiding unwanted collision between neighboring vessels. In addition, the operation of trajectory clustering is further incorporated into our trajectory prediction framework, potentially leading to more robust prediction results. Numerous experiments have been implemented on realistic automatic identification system (AIS)-based vessel trajectories to compare our QSD-LSTM with several state-of-the-art prediction methods. The prediction results have demonstrated the superior performance of our method in terms of both quantitative and qualitative evaluations.

Recommended citation: Liu, R. W., Hu, K., Liang, M., Li, Y., Liu, X., & Yang, D. (2023). QSD-LSTM: Vessel trajectory prediction using long short-term memory with quaternion ship domain. Applied Ocean Research, 136, 103592. https://doi.org/10.1016/j.apor.2023.103592

Ship collision risk analysis: Modeling, visualization and prediction

Published in Ocean Engineering, 2022

Ship collisions are the primary threat to traffic safety in the sea, which can seriously threaten human lives, the environment and material assets. Therefore, the detection and analyze of ship collision risks have important theoretical significance and application value. To improve maritime safety and efficiency, we propose a modeling, visualization and prediction framework to analyze ship collision risk. In particular, to fully consider the maneuverability of the ship, we introduce the quaternion ship domain (QSD) into the vessel conflict ranking operator (VCRO). In addition, to further analyze and better understand collision risk, the kernel density estimation (KDE) model is employed to visualize the ship collision risk. The ship collision risk usually contains underlying patterns and laws. Thus, we proposed a convolutional long short-term memory network (ConvLSTM) model, which can extract spatial-temporal features and predict spatial-temporal risk. Finally, to verify the reliability and robustness of the framework, we conducted extensive experiments on the automatic identification system (AIS) data of Chengshantou water. The results show that the framework demonstrates superiority in risk calculation, visualization and prediction. Theoretically, the framework proposed in this paper can serve maritime intelligent transportation system well.

Recommended citation: Liu, R. W., Huo, X., Liang, M., & Wang, K. (2022). Ship collision risk analysis: modeling, visualization and prediction. Ocean Engineering, 266, 112895. https://doi.org/10.1016/j.oceaneng.2022.112895

Fine-Grained Vessel Traffic Flow Prediction with a Spatio-Temporal Multi-Graph Convolutional Network

Published in IEEE Transactions on Intelligent Transportation Systems, 2022

The accurate and robust prediction of vessel traffic flow is gaining importance in maritime intelligent transportation system (ITS), such as vessel traffic services, maritime spatial planning, and traffic safety management, etc. To achieve fine-grained vessel traffic flow prediction, we will first generate the maritime traffic network (which is essentially a graph), and then propose a graph-driven neural network. In particular, to represent various correlations among spatio-temporal vessel traffic flow, we tend to extract the feature points (i.e., starting, way and ending points) by utilizing the knowledge of vessel positioning data. These feature points are essentially related to the geometrical structures of massive vessel trajectories collected from massive automatic identification system (AIS) records, contributing to the generation of maritime traffic network. We then propose a spatio-temporal multi-graph convolutional network (STMGCN)-based vessel traffic flow prediction method by exploiting multiple types of inherent correlations in the generated maritime graph. The proposed STMGCN mainly contains one spatial multi-graph convolutional layer and two temporal gated convolutional layers, beneficial for extracting spatial and temporal traffic flow patterns. The main benefit of our graph-driven prediction method is that it takes full advantage of the maritime graph and multi-graph learning. Comprehensive experiments have been implemented on realistic AIS dataset to compare our method with several state-of-the-art prediction methods. The fine-grained prediction results have demonstrated our superior performance in terms of both accuracy and robustness.

Recommended citation: Liang, M., Liu, R. W., Zhan, Y., Li, H., Zhu, F., & Wang, F. Y. (2022). Fine-grained vessel traffic flow prediction with a spatio-temporal multigraph convolutional network. IEEE Transactions on Intelligent Transportation Systems, 23(12), 23694-23707. http://10.1109/TITS.2022.3199160

STMGCN: Mobile Edge Computing-Empowered Vessel Trajectory Prediction Using Spatio-Temporal Multi-Graph Convolutional Network

Published in IEEE Transactions on Industrial Informatics, 2022

The revolutionary advances in machine learning and data mining techniques have contributed greatly to the rapid developments of maritime Internet of Things (IoT). In maritime IoT, the spatio-temporal vessel trajectories, collected from the hybrid satellite-terrestrial automatic identification system (AIS) base stations, are of considerable importance for promoting traffic situation awareness and vessel traffic services, etc. To guarantee traffic safety and efficiency, it is essential to robustly and accurately predict the AIS-based vessel trajectories (i.e., the future positions of vessels) in maritime IoT. In this work, we propose a spatio-temporal multigraph convolutional network (STMGCN)-based trajectory prediction framework using the mobile edge computing (MEC) paradigm. Our STMGCN is mainly composed of three different graphs, which are, respectively, reconstructed according to the social force, the time to closest point of approach, and the size of surrounding vessels. These three graphs are then jointly embedded into the prediction framework by introducing the spatio-temporal multigraph convolutional layer. To further enhance the prediction performance, the self-attention temporal convolutional layer is proposed to further optimize STMGCN with fewer parameters. Owing to the high interpretability and powerful learning ability, STMGCN is able to achieve superior prediction performance in terms of both accuracy and robustness. The reliable prediction results are potentially beneficial for traffic safety management and intelligent vehicle navigation in MEC-enabled maritime IoT.

Recommended citation: Liu, R. W., Liang, M., Nie, J., Yuan, Y., Xiong, Z., Yu, H., & Guizani, N. (2022). STMGCN: Mobile edge computing-empowered vessel trajectory prediction using spatio-temporal multigraph convolutional network. IEEE Transactions on Industrial Informatics, 18(11), 7977-7987. http://10.1109/TII.2022.3165886

Deep learning-powered vessel trajectory prediction for improving smart traffic services in maritime Internet of Things

Published in IEEE Transactions on Network Science and Engineering, 2022

The maritime Internet of Things (IoT) has recently emerged as a revolutionary communication paradigm where a large number of moving vessels are closely interconnected in intelligent maritime networks. However, the tremendous growth of vessel trajectories, collected from the combined satellite-terrestrial AIS (automatic identification system) base stations, could lead to unsatisfactory maritime safety and efficacy. To promote smart traffic services in maritime IoT, it is necessary to accurately and robustly predict the spatiotemporal vessel trajectories. It is beneficial for collision avoidance, maritime surveillance, and abnormal behavior detection, etc. Motivated by the strong learning capacity of deep neural networks, this work proposes an AIS data-driven trajectory prediction framework, whose main component is a long short term memory (LSTM) network. In particular, the vessel traffic conflict situation modeling, generated using the dynamic AIS data and social force concept, is embedded into the LSTM network to guarantee high-accuracy vessel trajectory prediction. In addition, a mixed loss function is reconstructed to make our prediction results more reliable and robust in different navigation environments. Several quantitative and qualitative experiments have been implemented on realistic AIS-based vessel trajectories. Our results have demonstrated that the proposed method could achieve satisfactory prediction performance in terms of accuracy and robustness.

Recommended citation: Liu, R. W., Liang, M., Nie, J., Lim, W. Y. B., Zhang, Y., & Guizani, M. (2022). Deep learning-powered vessel trajectory prediction for improving smart traffic services in maritime Internet of Things. IEEE Transactions on Network Science and Engineering, 9(5), 3080-3094. http://10.1109/TNSE.2022.3140529

MVFFNet: Multi-view feature fusion network for imbalanced ship classification

Published in Pattern Recognition Letters, 2021

The accurate classification of moving ships is of fundamental importance to maritime authorities for ensuring the safety and security of shipping operations. With the wide use of automatic identification systems (AISs), which allow ships to receive identification/location information from nearby ships, it is feasible to classify the types of ships by analysing ship behaviour from AIS-based trajectories. However, the imbalanced features of AIS data make it difficult to achieve satisfactory classification results in the presence of several different types of ships. To overcome these potential limitations, we propose a multi-view feature fusion network (MVFFNet) to achieve accurate ship classification with imbalanced data. To guarantee the powerful representation and generalization abilities of MVFFNet, we first extract several multi-view features (i.e., motion features and morphological features) from AIS-based ship trajectories. Several kinematic variables related to ship behaviour are empirically adopted as motion features. The morphological features are automatically extracted via convolutional auto-encoder (CAE) networks. CAE networks are capable of optimally learning the features from informative trajectory images, which are strictly related to the original ship trajectories. The bidirectional gated recurrent unit (BiGRU) network is then proposed to combine multi-view features to generate the ship classification results. In addition, a hybrid loss function is presented to handle the imbalance problem of ship types, potentially leading to enhanced robustness and accuracy of ship classification. Comprehensive experiments on two realistic datasets have demonstrated that our proposed MVFFNet consistently outperforms other competing methods in terms of classification accuracy and robustness.

Recommended citation: Liang, M., Zhan, Y., & Liu, R. W. (2021). MVFFNet: Multi-view feature fusion network for imbalanced ship classification. Pattern Recognition Letters, 51, 26-32. https://doi.org/10.1016/j.patrec.2021.07.024

CNN-enabled visibility enhancement framework for vessel detection under haze environment

Published in Journal of advanced transportation, 2021

Maritime images captured under haze environment often have a terrible visual effect, making it easy to overlook important information. To avoid the failure of vessel detection caused by fog, it is necessary to preprocess the collected hazy images for recovering vital information. In this paper, a novel CNN-enabled visibility dehazing framework is proposed, consisting of two subnetworks, that is, Coarse Feature Extraction Module (C-FEM) and Fine Feature Fusion Module (F-FFM). Specifically, C-FEM is a multiscale haze feature extraction network, which can learn information from three scales. Correspondingly, F-FFM is an improved encoder-decoder network to fuse multiscale information obtained by C-FEM and enhance the visual effect of the final output. Meanwhile, a hybrid loss function is designed for monitoring the multiscale output of C-FEM and the final result of F-FFM simultaneously. It is worth mentioning that massive maritime images are considered the training dataset to further adapt the vessel detection task under haze environment. Comprehensive experiments on synthetic and realistic images have verified the superior effectiveness and robustness of our CNN-enabled visibility dehazing framework compared to several state-of-the-art methods. Our method preprocesses images before vessel detection to demonstrate our framework has the capacity of promoting maritime video surveillance.

Recommended citation: Lu, Y., Guo, Y., & Liang, M. (2021). CNN-enabled visibility enhancement framework for vessel detection under haze environment. Journal of advanced transportation, 2021, 1-14. https://doi.org/10.1155/2021/5598390

An unsupervised learning method with convolutional auto-encoder for vessel trajectory similarity computation

Published in Ocean Engineering, 2021

To achieve reliable mining results for massive vessel trajectories, one of the most important challenges is how to efficiently compute the similarities between different vessel trajectories. The computation of vessel trajectory similarity has recently attracted increasing attention in the maritime data mining research community. However, traditional shape- and warping-based methods often suffer from several drawbacks such as high computational cost and sensitivity to unwanted artifacts and non-uniform sampling rates, etc. To eliminate these drawbacks, we propose an unsupervised learning method which automatically extracts low-dimensional features through a convolutional auto-encoder (CAE). In particular, we first generate the informative trajectory images by remapping the raw vessel trajectories into two-dimensional matrices while maintaining the spatio-temporal properties. Based on the massive vessel trajectories collected, the CAE can learn the low-dimensional representations of informative trajectory images in an unsupervised manner. The trajectory similarity is finally equivalent to efficiently computing the similarities between the learned low-dimensional features, which strongly correlate with the raw vessel trajectories. Comprehensive experiments on realistic data sets have demonstrated that the proposed method largely outperforms traditional trajectory similarity computation methods in terms of efficiency and effectiveness. The high-quality trajectory clustering performance could also be guaranteed according to the CAE-based trajectory similarity computation results.

Recommended citation: Liang, M., Liu, R. W., Li, S., Xiao, Z., Liu, X., & Lu, F. (2021). An unsupervised learning method with convolutional auto-encoder for vessel trajectory similarity computation. Ocean Engineering, 225, 108803. http://doi.org/10.1016/j.oceaneng.2021.108803

Liang Maohan (梁茂晗)

Publications