Hongming Chen
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
Hongming Chen has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Yufeng Li
The College of Electronic Information Engineering, Shenyang Aerospace University, Shenyang, China
6 shared publications - Xianping Fu
Dalian Maritime University
5 shared publications - Guoliang Yuan
The School of Information Science and Technology, Dalian Maritime University, Dalian, China
4 shared publications - Xiang Chen
School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China
4 shared publications - Yafei Wang
College of Information Science and Technology, Dalian Maritime University, Dalian, China
4 shared publications - Qi Ling
College of Information Science and Technology, Dalian Maritime University, Dalian, China
3 shared publications
Latest Publications
- 3 September 2026
- Published by Elsevier in Ocean Engineering
- Vol. 367
U-AsymSeg: Asymmetric dual-stream RGB-Polarization network for underwater semantic segmentation
- 3 September 2026
- Published by Elsevier in Ocean Engineering
- Vol. 367
U-AsymSeg: Asymmetric dual-stream RGB-Polarization network for underwater semantic segmentation
- 16 May 2026
- Published by Elsevier in Optics and Lasers in Engineering
- Vol. 204
Beyond RGB: A YCbCr-based structure-color decoupling network for underwater image enhancement
- 16 May 2026
- Published by Elsevier in Optics and Lasers in Engineering
- Vol. 204
Beyond RGB: A YCbCr-based structure-color decoupling network for underwater image enhancement
- 29 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 33 (10709908) , 1736-1740
Boosting Underwater Object Detection via Differential Attention
Underwater object detection inherently struggles with severe challenges, primarily characterized by low image contrast and target scale imbalance. To address these challenges, we propose Differential Feature Refinement Network (DFR-Net), a specialized architecture for underwater scenarios. We introduce the Differential Attention Feature Interaction (DAFI) module in the encoder, which utilizes a differential mechanism to implicitly cancel common-mode noise, significantly enhancing the feature Signal-to-Noise Ratio (SNR). Subsequently, we design the Aggregate-Diffuse FPN (AD-FPN), which integrates the Dilated Selective Aggregation (DSA) unit to achieve efficient detection of underwater multi-scale targets, and the Semantic-Guided Gating Fusion (SGF) unit to suppress noise accumulation during feature fusion. Experiments on DUO, RUOD, UDD, and UODD datasets demonstrate that the proposed DFR-Net outperforms the DEIM baseline by 2.8% on average. With only 4.5 M parameters and 10.5 G FLOPs, it achieves an inference speed of 22.5 FPS on an edge device, offering an optimal balance between performance and deployment efficiency.
- 29 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 33 (10709908) , 1736-1740
Boosting Underwater Object Detection via Differential Attention
Underwater object detection inherently struggles with severe challenges, primarily characterized by low image contrast and target scale imbalance. To address these challenges, we propose Differential Feature Refinement Network (DFR-Net), a specialized architecture for underwater scenarios. We introduce the Differential Attention Feature Interaction (DAFI) module in the encoder, which utilizes a differential mechanism to implicitly cancel common-mode noise, significantly enhancing the feature Signal-to-Noise Ratio (SNR). Subsequently, we design the Aggregate-Diffuse FPN (AD-FPN), which integrates the Dilated Selective Aggregation (DSA) unit to achieve efficient detection of underwater multi-scale targets, and the Semantic-Guided Gating Fusion (SGF) unit to suppress noise accumulation during feature fusion. Experiments on DUO, RUOD, UDD, and UODD datasets demonstrate that the proposed DFR-Net outperforms the DEIM baseline by 2.8% on average. With only 4.5 M parameters and 10.5 G FLOPs, it achieves an inference speed of 22.5 FPS on an edge device, offering an optimal balance between performance and deployment efficiency.
- Top Cited Papers
- 23 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 4690-4702
Towards Ultra-High-Definition Image Deraining: A Benchmark and an Efficient Method
Despite significant advancements in image deraining, most existing methods are carried out on low-resolution images, leaving their effectiveness on high-resolution images uncertain. This limitation becomes even more pronounced with the rise of ultra-high-definition (UHD) imaging. In this paper, we tackle the challenge of UHD image deraining and introduce 4K-Rain13 k, the first large-scale UHD image deraining dataset, featuring 13,000 paired images at 4 K resolution. Leveraging this dataset, we conduct a benchmark study on existing methods for processing UHD images. To better address this task, we propose UDR-Mixer, an efficient and effective architecture tailored for UHD image deraining. Our model comprises two key components: a spatial feature rearrangement layer, which captures long-range dependencies in UHD images, and a frequency feature modulation layer, which enhances high-fidelity image reconstruction. Extensive experiments demonstrate that our approach outperforms state-of-the-art methods while maintaining lower model complexity. The source code and proposed dataset are available at https://github.com/cschenxiang/UDR-Mixer.
- 23 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 4690-4702
Top Cited PapersTowards Ultra-High-Definition Image Deraining: A Benchmark and an Efficient Method
Despite significant advancements in image deraining, most existing methods are carried out on low-resolution images, leaving their effectiveness on high-resolution images uncertain. This limitation becomes even more pronounced with the rise of ultra-high-definition (UHD) imaging. In this paper, we tackle the challenge of UHD image deraining and introduce 4K-Rain13 k, the first large-scale UHD image deraining dataset, featuring 13,000 paired images at 4 K resolution. Leveraging this dataset, we conduct a benchmark study on existing methods for processing UHD images. To better address this task, we propose UDR-Mixer, an efficient and effective architecture tailored for UHD image deraining. Our model comprises two key components: a spatial feature rearrangement layer, which captures long-range dependencies in UHD images, and a frequency feature modulation layer, which enhances high-fidelity image reconstruction. Extensive experiments demonstrate that our approach outperforms state-of-the-art methods while maintaining lower model complexity. The source code and proposed dataset are available at https://github.com/cschenxiang/UDR-Mixer.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 33 (10709908) , 1641-1645
Toward All-in-One UAV Imagery Restoration via Prompt Learning With Multi-Scale Mamba
Despite significant progress in image restoration, most existing methods exhibit limited performance and generalization ability when handling multiple degradation types simultaneously. This challenge is particularly pronounced in UAV aerial imaging, where complex flight environments often introduce multiple degradation factors, severely reducing image quality and hindering downstream vision tasks. To address this issue, we construct the first integrated image restoration dataset for UAV aerial scenarios, named UAV-ARID, which includes five types of degradation: blur, noise, lowlight, raindrops, and haze. In addition, we propose a prompt-guided multi-scale mamba architecture for all-in-one UAV aerial image restoration. This framework integrates prompt learning with a multi-scale mamba architecture to dynamically capture image features and semantic information across scales. By generating context-rich conditional prompts, our model effectively guides the restoration process, recovering images from various types of degradation while maintaining low computational complexity for efficient restoration. Experimental results demonstrate that our approach achieves superior performance compared to state-of-the-art methods, validating its effectiveness.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 33 (10709908) , 1641-1645
Toward All-in-One UAV Imagery Restoration via Prompt Learning With Multi-Scale Mamba
Despite significant progress in image restoration, most existing methods exhibit limited performance and generalization ability when handling multiple degradation types simultaneously. This challenge is particularly pronounced in UAV aerial imaging, where complex flight environments often introduce multiple degradation factors, severely reducing image quality and hindering downstream vision tasks. To address this issue, we construct the first integrated image restoration dataset for UAV aerial scenarios, named UAV-ARID, which includes five types of degradation: blur, noise, lowlight, raindrops, and haze. In addition, we propose a prompt-guided multi-scale mamba architecture for all-in-one UAV aerial image restoration. This framework integrates prompt learning with a multi-scale mamba architecture to dynamically capture image features and semantic information across scales. By generating context-rich conditional prompts, our model effectively guides the restoration process, recovering images from various types of degradation while maintaining low computational complexity for efficient restoration. Experimental results demonstrate that our approach achieves superior performance compared to state-of-the-art methods, validating its effectiveness.
- 7 November 2025
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 23 (1) , 1-14
Dual supervision guided learning for underwater object detection
- 7 November 2025
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 23 (1) , 1-14
Dual supervision guided learning for underwater object detection
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12722-12728
SmokeBench: A Real-World Dataset for Surveillance Image Desmoking in Early-Stage Fire Scenes
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12722-12728
Open AccessSmokeBench: A Real-World Dataset for Surveillance Image Desmoking in Early-Stage Fire Scenes
- 3 January 2025
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 19 (2) , 1-8
Multi-scale representation for image deraining with state space model
- 3 January 2025
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 19 (2) , 1-8
Multi-scale representation for image deraining with state space model
- 3 January 2025
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 19 (2) , 1-9
Lightweight vision Mamba for weather-degraded remote sensing image restoration
- 3 January 2025
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 19 (2) , 1-9
Lightweight vision Mamba for weather-degraded remote sensing image restoration