Yuxu Lu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yuxu Lu has a total of 42 co-authors. Here are the top co-authors by number of shared publications:
- Ryan Wen Liu
Wuhan University of Technology
12 shared publications - Yu Guo
School of Navigation, Wuhan University of Technology, Wuhan, China
10 shared publications - Dong Yang
Department of Logistics and Maritime Studies, The Hong Kong Polytechnic University, Hong Kong
7 shared publications - Fenghua Zhu
Institute of Automation, Chinese Academy of Sciences, Beijing, China
4 shared publications - Yuan Gao
Tsinghua University, Beijing, China
4 shared publications - Wenqi Ren
Shenzhen Campus of Sun Yat-sen University, Shenzhen, China
3 shared publications
Latest Publications
- 25 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. PP (15513203) , 1-11
Seeking Safety From Uncertainty: Probabilistic Vessel Trajectory Prediction With a Flow-Based Generative Model
Vessel trajectory prediction remains a critical focus in maritime safety, serving as a pivotal component in vessel traffic service and maritime autonomous surface ship. However, unmeasurable factors, such as human intentions and vessel failure, pose challenges to the reliability of predictions, particularly in multivessel encounters. In response to this, we propose the sequence generative flow (SGlow), a flow-based deep generative model that describes future motions of the vessels with a multimodal probability distribution. It mitigates navigation uncertainty by multiple probabilistic inferences, mapping a single observation to multiple possible predictions. By combining sequential modeling with flow-based distribution learning, SGlow provides more reliable and comprehensive trajectory forecasts for maritime risk warning and navigation decision-making. Compared to existing deep generative models, SGlow excels in multimodal prediction of vessel trajectory with two distinct advantages: first, precise modeling of trajectory distributions through exact log-likelihood optimization, and second, a mathematically traceable learning process that aligns with trajectory evolution dynamics. Experimental results show that SGlow achieves favorable performance compared with 11 benchmark deep learning models in both prediction accuracy and plausibility, enhancing maritime operations safety.
- 25 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. PP (15513203) , 1-11
Seeking Safety From Uncertainty: Probabilistic Vessel Trajectory Prediction With a Flow-Based Generative Model
Vessel trajectory prediction remains a critical focus in maritime safety, serving as a pivotal component in vessel traffic service and maritime autonomous surface ship. However, unmeasurable factors, such as human intentions and vessel failure, pose challenges to the reliability of predictions, particularly in multivessel encounters. In response to this, we propose the sequence generative flow (SGlow), a flow-based deep generative model that describes future motions of the vessels with a multimodal probability distribution. It mitigates navigation uncertainty by multiple probabilistic inferences, mapping a single observation to multiple possible predictions. By combining sequential modeling with flow-based distribution learning, SGlow provides more reliable and comprehensive trajectory forecasts for maritime risk warning and navigation decision-making. Compared to existing deep generative models, SGlow excels in multimodal prediction of vessel trajectory with two distinct advantages: first, precise modeling of trajectory distributions through exact log-likelihood optimization, and second, a mathematically traceable learning process that aligns with trajectory evolution dynamics. Experimental results show that SGlow achieves favorable performance compared with 11 benchmark deep learning models in both prediction accuracy and plausibility, enhancing maritime operations safety.
- 11 May 2026
- Published by Elsevier in Ocean Engineering
- Vol. 359
Resilient inland vessel trajectory prediction via visual-kinematic data imputation and interaction-aware graph learning
- 11 May 2026
- Published by Elsevier in Ocean Engineering
- Vol. 359
Resilient inland vessel trajectory prediction via visual-kinematic data imputation and interaction-aware graph learning
- Top Cited Papers
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 27 (5) , 5739-5754
Graph Learning-Driven Multi-Vessel Association: Fusing Multimodal Data for Maritime Intelligence
Ensuring maritime safety and optimizing traffic management in increasingly crowded and complex waterways require effective waterway monitoring. However, current methods struggle with challenges arising from multimodal data, such as dimensional disparities, mismatched target counts, vessel scale variations, occlusions, and asynchronous data streams from systems like the automatic identification system (AIS) and closed-circuit television (CCTV). Traditional multi-vessel association methods often struggle with these complexities, particularly in densely trafficked waterways. To overcome these issues, we propose a graph learning-driven multi-vessel association (named GMvA) method tailored for maritime multimodal data fusion. By integrating AIS and CCTV data, GMvA leverages time series learning and graph neural networks to capture the spatiotemporal features of vessel trajectories effectively. To enhance feature representation, the proposed method incorporates temporal graph attention and spatiotemporal attention, effectively capturing both local and global vessel interactions. Furthermore, a multi-layer perceptron-based uncertainty fusion module computes robust similarity scores, and the Hungarian algorithm is adopted to ensure globally consistent and accurate target matching. To validate the efficacy of our method, we have also constructed a new maritime multimodal dataset (termed MaritimeMmD) for data fusion. Extensive experiments demonstrate that GMvA delivers superior accuracy and robustness in multi-vessel association, outperforming existing methods even in challenging scenarios with high vessel density and incomplete or unevenly distributed AIS and CCTV data. The source dataset and code are available at https://github.com/LouisYxLu/GMvA
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 27 (5) , 5739-5754
Top Cited PapersGraph Learning-Driven Multi-Vessel Association: Fusing Multimodal Data for Maritime Intelligence
Ensuring maritime safety and optimizing traffic management in increasingly crowded and complex waterways require effective waterway monitoring. However, current methods struggle with challenges arising from multimodal data, such as dimensional disparities, mismatched target counts, vessel scale variations, occlusions, and asynchronous data streams from systems like the automatic identification system (AIS) and closed-circuit television (CCTV). Traditional multi-vessel association methods often struggle with these complexities, particularly in densely trafficked waterways. To overcome these issues, we propose a graph learning-driven multi-vessel association (named GMvA) method tailored for maritime multimodal data fusion. By integrating AIS and CCTV data, GMvA leverages time series learning and graph neural networks to capture the spatiotemporal features of vessel trajectories effectively. To enhance feature representation, the proposed method incorporates temporal graph attention and spatiotemporal attention, effectively capturing both local and global vessel interactions. Furthermore, a multi-layer perceptron-based uncertainty fusion module computes robust similarity scores, and the Hungarian algorithm is adopted to ensure globally consistent and accurate target matching. To validate the efficacy of our method, we have also constructed a new maritime multimodal dataset (termed MaritimeMmD) for data fusion. Extensive experiments demonstrate that GMvA delivers superior accuracy and robustness in multi-vessel association, outperforming existing methods even in challenging scenarios with high vessel density and incomplete or unevenly distributed AIS and CCTV data. The source dataset and code are available at https://github.com/LouisYxLu/GMvA
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 27 (10) , 1-16
CmIVTP: Cross-Modal Interaction-Based Vessel Trajectory Prediction for Maritime Intelligence
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 27 (10) , 1-16
CmIVTP: Cross-Modal Interaction-Based Vessel Trajectory Prediction for Maritime Intelligence
- Open Access
- 3 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 17610-17621
Spatial–Frequency Fusion Network With Learnable Fractional Fourier Transform for Remote Sensing Imaging Enhancement
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 paper proposes a novel spatial-frequency fusion network (termed SFFNet) with learnable fractional Fourier transform for efficient remote sensing imaging enhancement. In the spatial domain, SFFNet uses a multi-scale 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 transforms module adaptively extracts haze-relevant features, leveraging a learnable parameter to dynamically adjust the fractional order of the transform. Furthermore, an attentive frequency gate selectively emphasizes critical frequency features based on the local features of the input image. To effectively address the challenges of non-uniform haze distribution, a self-attention-guided fusion mechanism is introduced, synergistically integrating spatial and frequency information. Additionally, a hierarchical feature fusion strategy progressively refines multi-scale features throughout the dehazing process, ensuring comprehensive and accurate haze removal. Experimental results on both synthetic and real-world remote sensing datasets show that SFFNet achieves significant improvements in quantitative metrics and visual quality. Moreover, SFFNet demonstrates strong practical potential in remote sensing object detection by improving accuracy and robustness.
- 3 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 17610-17621
Open AccessSpatial–Frequency Fusion Network With Learnable Fractional Fourier Transform for Remote Sensing Imaging Enhancement
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 paper proposes a novel spatial-frequency fusion network (termed SFFNet) with learnable fractional Fourier transform for efficient remote sensing imaging enhancement. In the spatial domain, SFFNet uses a multi-scale 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 transforms module adaptively extracts haze-relevant features, leveraging a learnable parameter to dynamically adjust the fractional order of the transform. Furthermore, an attentive frequency gate selectively emphasizes critical frequency features based on the local features of the input image. To effectively address the challenges of non-uniform haze distribution, a self-attention-guided fusion mechanism is introduced, synergistically integrating spatial and frequency information. Additionally, a hierarchical feature fusion strategy progressively refines multi-scale features throughout the dehazing process, ensuring comprehensive and accurate haze removal. Experimental results on both synthetic and real-world remote sensing datasets show that SFFNet achieves significant improvements in quantitative metrics and visual quality. Moreover, SFFNet demonstrates strong practical potential in remote sensing object detection by improving accuracy and robustness.
- 1 July 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 294
Multi-view adaptive image enhancement with hierarchical attention for complex underground mining scenes
- 1 July 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 294
Multi-view adaptive image enhancement with hierarchical attention for complex underground mining scenes
- 6 December 2024
- Published by Springer Nature
- p. 255-272
OneRestore: A Universal Restoration Framework for Composite Degradation
- 6 December 2024
- Published by Springer Nature
- p. 255-272
OneRestore: A Universal Restoration Framework for Composite Degradation
- 17 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (12) , 19979-19994
Real-Time Multi-Scene Visibility Enhancement for Promoting Navigational Safety of Vessels Under Complex Weather Conditions
The visible-light camera, which is capable of environment perception and navigation assistance, has emerged as an essential imaging sensor for marine surface vessels in intelligent waterborne transportation systems (IWTS). However, the visual imaging quality inevitably suffers from several kinds of degradations (e.g., limited visibility, low contrast, color distortion, etc.) under complex weather conditions (e.g., haze, rain, and low-lightness). The degraded visual information will accordingly result in inaccurate environment perception and delayed operations for navigational risk. To promote the navigational safety of vessels, many computational methods have been presented to perform visual quality enhancement under poor weather conditions. However, most of these methods are essentially specific-purpose implementation strategies, only available for one specific weather type. To overcome this limitation, we propose to develop a general-purpose multi-scene visibility enhancement method, i.e., edge reparameterization-and attention-guided neural network (ERANet), to adaptively restore the degraded images captured under different weather conditions. In particular, our ERANet simultaneously exploits the channel attention, spatial attention, and reparameterization technology to enhance the visual quality while maintaining low computational cost. Extensive experiments conducted on standard and IWTS-related datasets have demonstrated that our ERANet could outperform several representative visibility enhancement methods in terms of both imaging quality and computational efficiency. The superior performance of IWTS-related object detection and scene segmentation could also be steadily obtained after ERANet-based visibility enhancement under complex weather conditions.
- 17 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (12) , 19979-19994
Real-Time Multi-Scene Visibility Enhancement for Promoting Navigational Safety of Vessels Under Complex Weather Conditions
The visible-light camera, which is capable of environment perception and navigation assistance, has emerged as an essential imaging sensor for marine surface vessels in intelligent waterborne transportation systems (IWTS). However, the visual imaging quality inevitably suffers from several kinds of degradations (e.g., limited visibility, low contrast, color distortion, etc.) under complex weather conditions (e.g., haze, rain, and low-lightness). The degraded visual information will accordingly result in inaccurate environment perception and delayed operations for navigational risk. To promote the navigational safety of vessels, many computational methods have been presented to perform visual quality enhancement under poor weather conditions. However, most of these methods are essentially specific-purpose implementation strategies, only available for one specific weather type. To overcome this limitation, we propose to develop a general-purpose multi-scene visibility enhancement method, i.e., edge reparameterization-and attention-guided neural network (ERANet), to adaptively restore the degraded images captured under different weather conditions. In particular, our ERANet simultaneously exploits the channel attention, spatial attention, and reparameterization technology to enhance the visual quality while maintaining low computational cost. Extensive experiments conducted on standard and IWTS-related datasets have demonstrated that our ERANet could outperform several representative visibility enhancement methods in terms of both imaging quality and computational efficiency. The superior performance of IWTS-related object detection and scene segmentation could also be steadily obtained after ERANet-based visibility enhancement under complex weather conditions.