Xiaofeng Ling
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Xiaofeng Ling has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Yu Zhu
School of Information Science and Engineering, East China University of Science and Technology, Shanghai, China
4 shared publications - Nan Wang
School of Information Science and Engineering, East China University of Science and Technology, Shanghai, China
3 shared publications - Rui Wang
Tongji University
3 shared publications - Hongqing Zhu
School of Information Science and Engineering, East China University of Science and Technology, Shanghai, 200237, China
2 shared publications Nan Wang
School of Information Science and Engineering, East China University of Science and Technology, Shanghai, China
2 shared publications- Pengyu Wang
Sports Artificial Intelligence Institute, Capital University of Physical Education and Sports, Beijing, China
2 shared publications
Latest Publications
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (3) , 4561-4574
DDCFusion: Dynamic Depth Compensation Fusion for Camera–Radar 3-D Object Detection
The effective representation and feature extraction from sparse point clouds of 4D millimeter-wave(4D-MMW) radar pose a significant challenge in 3D object detection. This paper proposes DDCFusion, a novel radar-camera fusion network that advances measurement precision by dynamically compensating for depth errors in sparse radar data. DDCFusion achieves this by exploiting the physical properties of 4D-MMW radar to improve measurement reliability and reduce depth uncertainty, which enhances depth measurement confidence in the view transform by integrating RCS-derived reflectivity metrics. The occupancy-weighted radar branch prioritizes image regions with high-confidence radar returns, minimizing measurement noise in view transform operation. Furthermore, DDCFusion optimizes spatial measurement consistency in Bird’s-Eye-View (BEV) space by modeling cross-sensor dependencies through the Global Feature Slice Coordinate Attention (GFSCA) fusion module. Experimental validation on the VoD and TJ4DRadSet datasets demonstrates superior measurement accuracy, achieving 51.08% mAP on VoD and 34.61% mAP on TJ4DRadSet—outperforming existing methods in depth error reduction and robustness to sparsity. Ablation studies verify the measurement-centric design: RCS-guided diffusion improves small-object detection (e.g., pedestrians), while DBSCAN-based clustering refines large-object localization (e.g., vehicles). The network demonstrates significant improvements in depth accuracy and robustness to sparse inputs while maintaining competitive inference latency with 138ms.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (3) , 4561-4574
DDCFusion: Dynamic Depth Compensation Fusion for Camera–Radar 3-D Object Detection
The effective representation and feature extraction from sparse point clouds of 4D millimeter-wave(4D-MMW) radar pose a significant challenge in 3D object detection. This paper proposes DDCFusion, a novel radar-camera fusion network that advances measurement precision by dynamically compensating for depth errors in sparse radar data. DDCFusion achieves this by exploiting the physical properties of 4D-MMW radar to improve measurement reliability and reduce depth uncertainty, which enhances depth measurement confidence in the view transform by integrating RCS-derived reflectivity metrics. The occupancy-weighted radar branch prioritizes image regions with high-confidence radar returns, minimizing measurement noise in view transform operation. Furthermore, DDCFusion optimizes spatial measurement consistency in Bird’s-Eye-View (BEV) space by modeling cross-sensor dependencies through the Global Feature Slice Coordinate Attention (GFSCA) fusion module. Experimental validation on the VoD and TJ4DRadSet datasets demonstrates superior measurement accuracy, achieving 51.08% mAP on VoD and 34.61% mAP on TJ4DRadSet—outperforming existing methods in depth error reduction and robustness to sparsity. Ablation studies verify the measurement-centric design: RCS-guided diffusion improves small-object detection (e.g., pedestrians), while DBSCAN-based clustering refines large-object localization (e.g., vehicles). The network demonstrates significant improvements in depth accuracy and robustness to sparse inputs while maintaining competitive inference latency with 138ms.
- Open Access
- 13 May 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (10) , 5478
High-Precision Positioning Method for Robot Acoustic Ranging Based on Self-Optimization of Base Stations
In response to the demand for high-precision positioning within confined or indoor environments, the application of acoustic ranging methods has been widely adopted by numerous engineers. Currently, time-of-flight (TOF)-based acoustic ranging positioning systems face challenges such as the susceptibility of sound velocity to environmental factors and the loss of acoustic signals at both short and long distances, which leads to a reduction in positioning accuracy. This paper addresses these issues by proposing a high-precision confidence interval weighting method for acoustic ranging and further introduces a method for base station deployment and self-optimization positioning within fixed indoor base station scenarios. The method is based on trilateration positioning, establishing criteria for the division of central and boundary areas. It categorizes mobile terminal nodes based on their coordinates from the previous moment, selects distance information from nearby base stations in different modes, and employs weights for decision-making and computation, ultimately yielding two-dimensional positioning coordinates. Experiments demonstrate that the proposed method can effectively enhance the positioning accuracy of acoustic positioning systems compared to traditional four-base station weighted average positioning algorithms.
- 13 May 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (10) , 5478
Open AccessHigh-Precision Positioning Method for Robot Acoustic Ranging Based on Self-Optimization of Base Stations
In response to the demand for high-precision positioning within confined or indoor environments, the application of acoustic ranging methods has been widely adopted by numerous engineers. Currently, time-of-flight (TOF)-based acoustic ranging positioning systems face challenges such as the susceptibility of sound velocity to environmental factors and the loss of acoustic signals at both short and long distances, which leads to a reduction in positioning accuracy. This paper addresses these issues by proposing a high-precision confidence interval weighting method for acoustic ranging and further introduces a method for base station deployment and self-optimization positioning within fixed indoor base station scenarios. The method is based on trilateration positioning, establishing criteria for the division of central and boundary areas. It categorizes mobile terminal nodes based on their coordinates from the previous moment, selects distance information from nearby base stations in different modes, and employs weights for decision-making and computation, ultimately yielding two-dimensional positioning coordinates. Experiments demonstrate that the proposed method can effectively enhance the positioning accuracy of acoustic positioning systems compared to traditional four-base station weighted average positioning algorithms.
- 23 February 2025
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 22 (2) , 1-12
FPGA-based 1D-CNN accelerator for real-time arrhythmia classification
- 23 February 2025
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 22 (2) , 1-12
FPGA-based 1D-CNN accelerator for real-time arrhythmia classification
- 4 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-13
TransMRE: Multiple Observation Planes Representation Encoding With Fully Sparse Voxel Transformers for 3-D Object Detection
The effective representation and feature extraction of 3D scenes from sparse and unstructured point clouds pose a significant challenge in 3D object detection. In this paper, we propose TransMRE, a network that enables fully sparse multiple observation plane feature fusion using LiDAR point clouds as single-modal input. TransMRE achieves this by sparsely factorizing a 3D voxel scene into three separate observation planes: X-Y plane, X-Z plane, and Y-Z plane. Additionally, we propose Observation Plane Sparse Fusion and Interaction to explore the internal relationship between different observation planes. The transformers mechanism is employed to realize feature attention within single observation plane and feature attention across multiple observation planes. This recursive application of attention is done during multiple observation plane projection feature aggregation to effectively model the entire 3D scene. This approach addresses the limitation of insufficient feature representation ability under a single bird’s-eye view constructed by extremely sparse point clouds. Furthermore, TransMRE maintains the full sparsity property of the entire network, eliminating the need to convert sparse feature maps into dense feature maps. As a result, it can be effectively applied to LiDAR point cloud data with large scanning ranges, such as Argoverse 2, while ensuring low computational complexity. Extensive experiments were conducted to evaluate the effectiveness of TransMRE, achieving 64.9 mAP and 70.4 NDS on the nuScenes detection benchmark, and 32.3 mAP on the Argoverse 2 detection benchmark. These results demonstrate that our method outperforms state-of-the-art methods.
- 4 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-13
TransMRE: Multiple Observation Planes Representation Encoding With Fully Sparse Voxel Transformers for 3-D Object Detection
The effective representation and feature extraction of 3D scenes from sparse and unstructured point clouds pose a significant challenge in 3D object detection. In this paper, we propose TransMRE, a network that enables fully sparse multiple observation plane feature fusion using LiDAR point clouds as single-modal input. TransMRE achieves this by sparsely factorizing a 3D voxel scene into three separate observation planes: X-Y plane, X-Z plane, and Y-Z plane. Additionally, we propose Observation Plane Sparse Fusion and Interaction to explore the internal relationship between different observation planes. The transformers mechanism is employed to realize feature attention within single observation plane and feature attention across multiple observation planes. This recursive application of attention is done during multiple observation plane projection feature aggregation to effectively model the entire 3D scene. This approach addresses the limitation of insufficient feature representation ability under a single bird’s-eye view constructed by extremely sparse point clouds. Furthermore, TransMRE maintains the full sparsity property of the entire network, eliminating the need to convert sparse feature maps into dense feature maps. As a result, it can be effectively applied to LiDAR point cloud data with large scanning ranges, such as Argoverse 2, while ensuring low computational complexity. Extensive experiments were conducted to evaluate the effectiveness of TransMRE, achieving 64.9 mAP and 70.4 NDS on the nuScenes detection benchmark, and 32.3 mAP on the Argoverse 2 detection benchmark. These results demonstrate that our method outperforms state-of-the-art methods.
- 2 April 2024
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (8) , 5265-5285
Occluded person re-identification based on parallel triplet augmentation and parameter-free token spatial attention
- 2 April 2024
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (8) , 5265-5285
Occluded person re-identification based on parallel triplet augmentation and parameter-free token spatial attention
- 22 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (1) , 750-756
Laser Direct Writing of Flexible Large-Area Graphene Sheet for Electrical Impedance Tomography Based Multimode Sensing Purposes
The high-performance electrical impedance tomography (EIT) based sensors rely on the formation of uniform and large-area conductive sensing layer in a highly-efficient and scalable method. In this study, a laser direct writing (LDW) technology is studied to prepare a large-area graphene layer for EIT-based sensors of pressure and temperature measurements. Through the optimization of the LDW parameters, the generated graphene sheet on a polyimide substrate with a minimized electrical anisotropy of 1.68, has a sensitivity of 1.76×10 -4 kPa -1 for applied pressure and a temperature coefficient of resistance of -2.26×10 -4 °C -1 . Besides, the EIT sensor demonstrates a great potential for temperature and pressure sensing purposes, which can be employed to detect the surface cracks and impact damage as well.
- 22 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (1) , 750-756
Laser Direct Writing of Flexible Large-Area Graphene Sheet for Electrical Impedance Tomography Based Multimode Sensing Purposes
The high-performance electrical impedance tomography (EIT) based sensors rely on the formation of uniform and large-area conductive sensing layer in a highly-efficient and scalable method. In this study, a laser direct writing (LDW) technology is studied to prepare a large-area graphene layer for EIT-based sensors of pressure and temperature measurements. Through the optimization of the LDW parameters, the generated graphene sheet on a polyimide substrate with a minimized electrical anisotropy of 1.68, has a sensitivity of 1.76×10 -4 kPa -1 for applied pressure and a temperature coefficient of resistance of -2.26×10 -4 °C -1 . Besides, the EIT sensor demonstrates a great potential for temperature and pressure sensing purposes, which can be employed to detect the surface cracks and impact damage as well.
- 5 October 2022
- Published by Elsevier in European Polymer Journal
- Vol. 180
Wearable, antibacterial, and self-healable modular sensors for monitoring joints movement ultra-sensitively
- 5 October 2022
- Published by Elsevier in European Polymer Journal
- Vol. 180
Wearable, antibacterial, and self-healable modular sensors for monitoring joints movement ultra-sensitively
- Open Access
- 11 January 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 10858-10870
ATT Squeeze U-Net: A Lightweight Network for Forest Fire Detection and Recognition
Forest fire is becoming one of the most significant natural disasters at the expense of ecology and economy. In this article, we develop an effective SqueezeNet based asymmetric encoder-decoder U-shape architecture, Attention U-Net and SqueezeNet (ATT Squeeze U-Net), mainly functions as an extractor and a discriminator of forest fire. This model takes attention mechanism to highlight useful features and suppress irrelevant contents by embedding Attention Gate (AG) units in the skip connection of U-shape structure. In this way, salient features are emphasized so that the proposed method could be competent at forest fire segmentation tasks with a small number of parameters. Specifically, we first replace classical convolution layer by a depthwise one and engage a Channel Shuffle operation as a feature communicator in the Fire module of classical SqueezeNet. Then, this modified SqueezeNet is employed as a substitution of the encoder of Attention U-Net and a corresponding DeFire module designed is combined into the decoder as well. Finally, to classify true fire, we take use of a fragment of the encoder in ATT Squeeze U-Net. The experimental results of modified SqueezeNet integrated Attention U-Net show that a competitive accuracy at 0.93 and an average prediction time at 0.89 second per image are achieved for reliable real-time forest fire detection.
- 11 January 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 10858-10870
Open AccessATT Squeeze U-Net: A Lightweight Network for Forest Fire Detection and Recognition
Forest fire is becoming one of the most significant natural disasters at the expense of ecology and economy. In this article, we develop an effective SqueezeNet based asymmetric encoder-decoder U-shape architecture, Attention U-Net and SqueezeNet (ATT Squeeze U-Net), mainly functions as an extractor and a discriminator of forest fire. This model takes attention mechanism to highlight useful features and suppress irrelevant contents by embedding Attention Gate (AG) units in the skip connection of U-shape structure. In this way, salient features are emphasized so that the proposed method could be competent at forest fire segmentation tasks with a small number of parameters. Specifically, we first replace classical convolution layer by a depthwise one and engage a Channel Shuffle operation as a feature communicator in the Fire module of classical SqueezeNet. Then, this modified SqueezeNet is employed as a substitution of the encoder of Attention U-Net and a corresponding DeFire module designed is combined into the decoder as well. Finally, to classify true fire, we take use of a fragment of the encoder in ATT Squeeze U-Net. The experimental results of modified SqueezeNet integrated Attention U-Net show that a competitive accuracy at 0.93 and an average prediction time at 0.89 second per image are achieved for reliable real-time forest fire detection.
- Open Access
- 11 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 31198-31213
Content-Based Superpixel Matching Using Spatially Constrained Student’s-t Mixture Model and Scale-Invariant Key-Superpixels
This paper addresses an image matching methodology designed for correspondence problem in computer vision. Firstly, a novel superpixel segmentation model driven by spatially constrained Student's-t mixture model (SMM) is proposed. The tails of Student's t-distribution are heavier than that of traditional Gaussian distribution, therefore, SMM is more insensitive to outliers and noise. In this model, a spatially constraint term based on Markov random field (MRF) is designed, so that good boundary adherence and intensity homogeneity would be achieved. Next, by constructing an adaptive superpixel Gaussian filter and a superpixel salient detector, this paper establishes an innovative key-superpixel detection method by building a superpixel scale-space pyramid. Different from conventional keypoint based detection, two images could then be matched directly in a superpixel-to-superpixel manner. During the matching process, a combinatorial feature descriptor that merges color, shape, gradient and texture features is set up to distinguish each considered key-superpixel. One main advantage of this approach is that implementation time would be largely reduced by less matching demand for key-superpixels and few corresponding local features. Some experiments on datasets at the end would demonstrate a relatively better performance of our model.
- 11 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 31198-31213
Open AccessContent-Based Superpixel Matching Using Spatially Constrained Student’s-t Mixture Model and Scale-Invariant Key-Superpixels
This paper addresses an image matching methodology designed for correspondence problem in computer vision. Firstly, a novel superpixel segmentation model driven by spatially constrained Student's-t mixture model (SMM) is proposed. The tails of Student's t-distribution are heavier than that of traditional Gaussian distribution, therefore, SMM is more insensitive to outliers and noise. In this model, a spatially constraint term based on Markov random field (MRF) is designed, so that good boundary adherence and intensity homogeneity would be achieved. Next, by constructing an adaptive superpixel Gaussian filter and a superpixel salient detector, this paper establishes an innovative key-superpixel detection method by building a superpixel scale-space pyramid. Different from conventional keypoint based detection, two images could then be matched directly in a superpixel-to-superpixel manner. During the matching process, a combinatorial feature descriptor that merges color, shape, gradient and texture features is set up to distinguish each considered key-superpixel. One main advantage of this approach is that implementation time would be largely reduced by less matching demand for key-superpixels and few corresponding local features. Some experiments on datasets at the end would demonstrate a relatively better performance of our model.
- 14 June 2019
- Published by Springer Nature
- p. 665-672
A Flexible Broadband Single RF Architecture Based on Time-Modulated Array
- 14 June 2019
- Published by Springer Nature
- p. 665-672
A Flexible Broadband Single RF Architecture Based on Time-Modulated Array