Xiaofang Kong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiaofang Kong has a total of 73 co-authors. Here are the top co-authors by number of shared publications:
- Minjie Wan
School of Electronic and Optical Engineering and Jiangsu Key Laboratory of Spectral Imaging and Intelligent Sense, Nanjing University of Science and Technology, Nanjing, China
13 shared publications - Guohua Gu
School of Electronic and Optical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China
12 shared publications - Qian Chen
School of Electronic and Optical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China
11 shared publications - Yunkai Xu
School of Electronic and Optical Engineering, the Jiangsu Key Laboratory of Spectral Imaging Intelligent Sense, and the Advanced Interdisciplinary Research Center for Optics, Nanjing University of Science and Technology, Nanjing, China
6 shared publications Weixian Qian
School of Electronic and Optical Engineering, Jiangsu Key Laboratory of Visual Sensing and Intelligent Perception, and the Advanced Interdisciplinary Research Center for Optics, Nanjing University of Science and Technology, Nanjing, China
5 shared publications- Ajun Shao
School of Electronic and Optical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China
4 shared publications
Latest Publications
- 6 August 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 333
DSTC-SRNet: A multi-frame super-resolution method based on deep spatio-temporal collaboration
- 6 August 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 333
DSTC-SRNet: A multi-frame super-resolution method based on deep spatio-temporal collaboration
- Open Access
- 1 July 2026
- Published by MDPI AG in Sustainability
- Vol. 18  (13) , 6659
Underwater Image Enhancement and Small Object Detection Method Based on RBE-CycleGAN and MSFDC-Net
Underwater object detection plays a vital role in marine exploration and resource exploitation. However, complex underwater environment leads to severe color deviation, blurring, and information loss of small targets, which greatly restrict detection performance. To address these problems, this paper integrates the Channel Attention and Spatial Attention Block (CASAB) attention mechanism into residual blocks based on generative adversarial networks to correct color distortion and improve the clarity of degraded underwater images. For underwater small object detection, MobileNetV2 is selected as the backbone network within the Faster R-CNN framework, and a multi-scale feature fusion strategy is adopted to reduce feature loss caused by repeated downsampling. In the detection head, coordinate attention and parallel dilated convolution are further integrated to suppress background noise and expand the receptive field of feature extraction. Experimental results on the Underwater Robot Professional Contest (URPC) dataset demonstrate that the proposed method yields gains of 10.06%, 9.43%, and 12.29% in three evaluation metrics: Underwater Image Quality Measure (UIQM), Underwater Colour Image Quality Evaluation (UCIQE) and Natural Image Quality Evaluator (NIQE), together with 7.81% in Mean Average Precision (mAP) and an 8.57% increase in Mean Recall (mRecall). These results demonstrate the effectiveness of all improvements.
- 1 July 2026
- Published by MDPI AG in Sustainability
- Vol. 18  (13) , 6659
Open AccessUnderwater Image Enhancement and Small Object Detection Method Based on RBE-CycleGAN and MSFDC-Net
Underwater object detection plays a vital role in marine exploration and resource exploitation. However, complex underwater environment leads to severe color deviation, blurring, and information loss of small targets, which greatly restrict detection performance. To address these problems, this paper integrates the Channel Attention and Spatial Attention Block (CASAB) attention mechanism into residual blocks based on generative adversarial networks to correct color distortion and improve the clarity of degraded underwater images. For underwater small object detection, MobileNetV2 is selected as the backbone network within the Faster R-CNN framework, and a multi-scale feature fusion strategy is adopted to reduce feature loss caused by repeated downsampling. In the detection head, coordinate attention and parallel dilated convolution are further integrated to suppress background noise and expand the receptive field of feature extraction. Experimental results on the Underwater Robot Professional Contest (URPC) dataset demonstrate that the proposed method yields gains of 10.06%, 9.43%, and 12.29% in three evaluation metrics: Underwater Image Quality Measure (UIQM), Underwater Colour Image Quality Evaluation (UCIQE) and Natural Image Quality Evaluator (NIQE), together with 7.81% in Mean Average Precision (mAP) and an 8.57% increase in Mean Recall (mRecall). These results demonstrate the effectiveness of all improvements.
- 26 February 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 316
An optimized additive bias field correction model for infrared image segmentation with intensity non-uniformity
- 26 February 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 316
An optimized additive bias field correction model for infrared image segmentation with intensity non-uniformity
- Open Access
- 20 January 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16  (2) , 1052
Adaptive Dual-Frequency Denoising Network-Based Strip Non-Uniformity Correction Method for Uncooled Long Wave Infrared Camera
The imaging quality of uncooled long wave infrared (IR) cameras is always limited by the stripe non-uniformity mainly caused by fixed pattern noise (FPN). In this paper, we propose an adaptive dual-frequency denoising network-based stripe non-uniformity correction (NUC) method, namely ADFDNet, to realize the balance between FPN removal and image detail preservation. Our ADFDNet takes the dual-frequency feature deconstruction module as its core, which decomposes the IR image into high-frequency and low-frequency features, and performs targeted processing through detail enhancement branches and sparse denoising branches. The former enhances the performance of detail preservation through multi-scale convolution and pixel attention mechanism, while the latter combines sparse attention mechanism and dilated convolution design to suppress high-frequency FPN. Furthermore, the dynamic weight fusion of features is realized using the adaptive dual-frequency fusion module, which better integrates detail information. In our study, a 420-pair image dataset covering different noise levels is constructed for better model training and evaluation. Experiments verify that the presented ADFDNet method significantly improves image clarity in both real and simulated noise scenes, and achieves a better balance between FPN suppression and detail preservation than other existing methods.
- 20 January 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16  (2) , 1052
Open AccessAdaptive Dual-Frequency Denoising Network-Based Strip Non-Uniformity Correction Method for Uncooled Long Wave Infrared Camera
The imaging quality of uncooled long wave infrared (IR) cameras is always limited by the stripe non-uniformity mainly caused by fixed pattern noise (FPN). In this paper, we propose an adaptive dual-frequency denoising network-based stripe non-uniformity correction (NUC) method, namely ADFDNet, to realize the balance between FPN removal and image detail preservation. Our ADFDNet takes the dual-frequency feature deconstruction module as its core, which decomposes the IR image into high-frequency and low-frequency features, and performs targeted processing through detail enhancement branches and sparse denoising branches. The former enhances the performance of detail preservation through multi-scale convolution and pixel attention mechanism, while the latter combines sparse attention mechanism and dilated convolution design to suppress high-frequency FPN. Furthermore, the dynamic weight fusion of features is realized using the adaptive dual-frequency fusion module, which better integrates detail information. In our study, a 420-pair image dataset covering different noise levels is constructed for better model training and evaluation. Experiments verify that the presented ADFDNet method significantly improves image clarity in both real and simulated noise scenes, and achieves a better balance between FPN suppression and detail preservation than other existing methods.
- 31 December 2025
- Published by Elsevier in Digital Signal Processing
- Vol. 172
Multiplicative bias field correction-based hybrid active contour model for infrared image segmentation
- 31 December 2025
- Published by Elsevier in Digital Signal Processing
- Vol. 172
Multiplicative bias field correction-based hybrid active contour model for infrared image segmentation
- 28 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Magnetics
- Vol. 61  (12) , 1-14
Discussion of the Influence of Multiple Factors on the Shielding of Magnetic Field in the Railgun’s Bore
Based on the finite element/boundary element coupling method, a three-dimensional (3D) numerical simulation model of transient electromagnetic field of railgun is established to simulate the diffusion of current and magnetic field on the shields with different material combinations in the bore, as well as the evolution process of magnetic field in shield’s cavity. The sliding electrical contact is defined between the rails and armature in this model. Combining with the theoretical analysis methods, the time of magnetic field diffusion in different shielding materials is evaluated. The influence of shields’ materials and the form of pulse current on the shielding effectiveness of the shield is analyzed through the simulation model. In order to investigate the effect of magnetic-structural coupling on the shielding effectiveness of shield, a magnetic-structural coupling simulation model of railgun launching process is also set up based on the multi-field coupling principle. The results reveal that in the strong magnetic field of bore, the shield made of copper material with higher conductivity can relatively significantly prevent the diffusion of magnetic field into the shield’s cavity. So its shielding effect is relatively good. With different form of pulse current, the dominant shielding mechanism will vary, resulting in differences of shielding effectiveness of shields in the bore. The distribution of magnetic field on the shield with and without considering the magnetic-structural coupling is different, especially with the passage of time, the difference is more obvious.
- 28 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Magnetics
- Vol. 61  (12) , 1-14
Discussion of the Influence of Multiple Factors on the Shielding of Magnetic Field in the Railgun’s Bore
Based on the finite element/boundary element coupling method, a three-dimensional (3D) numerical simulation model of transient electromagnetic field of railgun is established to simulate the diffusion of current and magnetic field on the shields with different material combinations in the bore, as well as the evolution process of magnetic field in shield’s cavity. The sliding electrical contact is defined between the rails and armature in this model. Combining with the theoretical analysis methods, the time of magnetic field diffusion in different shielding materials is evaluated. The influence of shields’ materials and the form of pulse current on the shielding effectiveness of the shield is analyzed through the simulation model. In order to investigate the effect of magnetic-structural coupling on the shielding effectiveness of shield, a magnetic-structural coupling simulation model of railgun launching process is also set up based on the multi-field coupling principle. The results reveal that in the strong magnetic field of bore, the shield made of copper material with higher conductivity can relatively significantly prevent the diffusion of magnetic field into the shield’s cavity. So its shielding effect is relatively good. With different form of pulse current, the dominant shielding mechanism will vary, resulting in differences of shielding effectiveness of shields in the bore. The distribution of magnetic field on the shield with and without considering the magnetic-structural coupling is different, especially with the passage of time, the difference is more obvious.
- 9 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 26  (12) , 22432-22446
SGA-YOLO: A Lightweight Real-Time Object Detection Network for UAV Infrared Images
The performance of existing object detection algorithms significantly degrades when applied to low-resolution infrared (IR) images captured by unmanned aerial vehicles (UAVs), which suffers from slow inference speed, low detection precision, and redundant network parameters. To tackle these issues, this paper proposes a lightweight real-time object detection network for UAV IR images, termed SGA-YOLO, which is designed based on the you only look at once version 8n (YOLOv8n) framework. First of all, the efficient SENetV2-neck enhances the correlation between different channels, which realizes efficient multi-scale feature fusion and improves detection precision. Subsequently, the lightweight S2GM backbone combines ShuffleNetV2-stride2 and C2f_Ghost modules, which significantly reduces the network parameters and increases inference speed. Finally, the adaptive fine-grained channel (AFGC) attention mechanism is coupled to further enhance detection precision and effectively mitigate background interference. Compared with the YOLOv8n, SGA-YOLO achieves a 27% reduction in network parameters, a 4.7% increment in precision, a 2.5% increment in recall rate, a 30.86% reduction in GFLOPs, a 16.3% increment in FPS, a 3.50% increment in mAP@0.5, and a 1.3% increment in mAP@0.5:0.95. In addition, it supports deployment on resource-constrained embedded system, offering a new perspective on designing lightweight UAV IR object detection networks for real-world applications in intelligent transportation systems. Our codes are available at https://github.com/gepengqiang2025/SGA-YOLO.
- 9 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 26  (12) , 22432-22446
SGA-YOLO: A Lightweight Real-Time Object Detection Network for UAV Infrared Images
The performance of existing object detection algorithms significantly degrades when applied to low-resolution infrared (IR) images captured by unmanned aerial vehicles (UAVs), which suffers from slow inference speed, low detection precision, and redundant network parameters. To tackle these issues, this paper proposes a lightweight real-time object detection network for UAV IR images, termed SGA-YOLO, which is designed based on the you only look at once version 8n (YOLOv8n) framework. First of all, the efficient SENetV2-neck enhances the correlation between different channels, which realizes efficient multi-scale feature fusion and improves detection precision. Subsequently, the lightweight S2GM backbone combines ShuffleNetV2-stride2 and C2f_Ghost modules, which significantly reduces the network parameters and increases inference speed. Finally, the adaptive fine-grained channel (AFGC) attention mechanism is coupled to further enhance detection precision and effectively mitigate background interference. Compared with the YOLOv8n, SGA-YOLO achieves a 27% reduction in network parameters, a 4.7% increment in precision, a 2.5% increment in recall rate, a 30.86% reduction in GFLOPs, a 16.3% increment in FPS, a 3.50% increment in mAP@0.5, and a 1.3% increment in mAP@0.5:0.95. In addition, it supports deployment on resource-constrained embedded system, offering a new perspective on designing lightweight UAV IR object detection networks for real-world applications in intelligent transportation systems. Our codes are available at https://github.com/gepengqiang2025/SGA-YOLO.
- 26 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25  (8) , 13357-13373
Small and Dim Target Detection Under Strong Clutter Based on Similarly of Gaussian and Motion Outlier Significance Using Moving Infrared Camera
The detection of small and dim target in cluttered environment remains a significant challenge. The similarity in spatial and temporal characteristics between strong clutter and small target presents a considerable difficulty for existing methods attempting to distinguish them. This paper proposes a method for detecting small and dim target in strong clutter environment when an infrared (IR) camera is in motion, based on the Similarity of Gaussian and Motion Outlier Significance (SOG-MOS). Initially, the Weighted Average Multidimensional Derivative (WAMD) and Similarity of Gaussian (SOG) are employed to select suspected dots. Subsequently, adaptive observable filtering based on residual analysis facilitates the correlation of suspected dots having the Markov property. Then, the Camera Global Motion (CGM) equation and the Target Local Motion (TLM) equation are derived from the projection relationship and available trajectory data. Ultimately, the true target is segmented based on Motion Outlier Significance (MOS) of clutter and target. Experimental results demonstrate that the method outperforms existing state-of-the-art techniques in terms of detection robustness.
- 26 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25  (8) , 13357-13373
Small and Dim Target Detection Under Strong Clutter Based on Similarly of Gaussian and Motion Outlier Significance Using Moving Infrared Camera
The detection of small and dim target in cluttered environment remains a significant challenge. The similarity in spatial and temporal characteristics between strong clutter and small target presents a considerable difficulty for existing methods attempting to distinguish them. This paper proposes a method for detecting small and dim target in strong clutter environment when an infrared (IR) camera is in motion, based on the Similarity of Gaussian and Motion Outlier Significance (SOG-MOS). Initially, the Weighted Average Multidimensional Derivative (WAMD) and Similarity of Gaussian (SOG) are employed to select suspected dots. Subsequently, adaptive observable filtering based on residual analysis facilitates the correlation of suspected dots having the Markov property. Then, the Camera Global Motion (CGM) equation and the Target Local Motion (TLM) equation are derived from the projection relationship and available trajectory data. Ultimately, the true target is segmented based on Motion Outlier Significance (MOS) of clutter and target. Experimental results demonstrate that the method outperforms existing state-of-the-art techniques in terms of detection robustness.
- 10 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62  (01962892) , 1-15
Underwater Image Restoration via Constrained Color Compensation and Background Light Color Space-Based Haze-Line Model
The quality of underwater imaging is significantly degraded by light scattering and absorption due to water body and suspended particles. To address the issues of color distortion and contrast degradation, we propose a novel underwater image restoration method based on constrained color compensation and background light color space-based haze-line model. The method begins by applying the constrained color compensation approach to provide targeted intensity adjustments for attenuated color channels. This process corrects color distortion while simultaneously mitigating the risks of over-compensation and insufficient color saturation. Subsequently, the haze-line model is employed in the transformed background light color space to restore the underwater image. Specifically, the color corrected underwater image is transformed to a novel color space, where the background light intensity serves as the origin. In this color space, all haze-lines are identified by grouping pixels with similar color characteristics through the superpixel clustering method. Then, the transmission distribution can be estimated based on the haze-line model. Finally, the scattered light components are removed by applying the underwater de-scattering model with the estimated transmission distribution to the luminance channel of the color corrected underwater image. Comparative experiments conducted on the UIEB and UCCS underwater image datasets demonstrate the superiority of the proposed method in terms of color correction and contrast enhancement when compared with state-of-the-art underwater image restoration techniques. Our codes are available at https://github.com/MinjieWan/C3HLM.
- 10 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62  (01962892) , 1-15
Underwater Image Restoration via Constrained Color Compensation and Background Light Color Space-Based Haze-Line Model
The quality of underwater imaging is significantly degraded by light scattering and absorption due to water body and suspended particles. To address the issues of color distortion and contrast degradation, we propose a novel underwater image restoration method based on constrained color compensation and background light color space-based haze-line model. The method begins by applying the constrained color compensation approach to provide targeted intensity adjustments for attenuated color channels. This process corrects color distortion while simultaneously mitigating the risks of over-compensation and insufficient color saturation. Subsequently, the haze-line model is employed in the transformed background light color space to restore the underwater image. Specifically, the color corrected underwater image is transformed to a novel color space, where the background light intensity serves as the origin. In this color space, all haze-lines are identified by grouping pixels with similar color characteristics through the superpixel clustering method. Then, the transmission distribution can be estimated based on the haze-line model. Finally, the scattered light components are removed by applying the underwater de-scattering model with the estimated transmission distribution to the luminance channel of the color corrected underwater image. Comparative experiments conducted on the UIEB and UCCS underwater image datasets demonstrate the superiority of the proposed method in terms of color correction and contrast enhancement when compared with state-of-the-art underwater image restoration techniques. Our codes are available at https://github.com/MinjieWan/C3HLM.
- 9 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73  (00189456) , 1-17
WTAPNet: Wavelet Transform-Based Augmented Perception Network for Infrared Small-Target Detection
Infrared (IR) small target detection plays a vital role in various applications of both civil and military areas, such as marine rescue, forest-fire prevention and precise guidance. However, challenges stemming from the small size of infrared targets and noise interference in complex backgrounds often limit the accuracy of IR small target detection algorithms. Owing to the rapid development of deep learning, numerous convolutional neural network (CNN)-based methods have emerged in recent years, but they are inevitable to encounter the risk of target loss in deep layers due to the use of pooling layers. To address this problem, we present a wavelet transform-based augmented perception network, namely WTAPNet, in this paper. First, an enhancement and enlarge (EE) module is designed to improve the network’s perceptual capability for IR small targets by magnifying image resolution and augmenting target features at the same time. Then, a discrete wavelet transform-based downsampling (DWTD) module and an inverse wavelet transform-based fusion (IWTF) module are proposed. These two modules collaboratively work to extract and fuse multi-scale features, which simultaneously reduces information loss. Finally, a bottom-up path fusion strategy is exploited to highlight and preserve small target features, which associates the lowest-level with the highest-level features and interconnects predictions from different hierarchical levels. Experimental results on NUDT-SIRST dataset and SIRST dataset demonstrate the superiority of our WTAPNet over other state-of-the-art IR small target detection methods in terms of F1-measure, recall and other indicators. The code is publicly available at https://github.com/hhy-NJUST/WTAPNet.
- 9 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73  (00189456) , 1-17
WTAPNet: Wavelet Transform-Based Augmented Perception Network for Infrared Small-Target Detection
Infrared (IR) small target detection plays a vital role in various applications of both civil and military areas, such as marine rescue, forest-fire prevention and precise guidance. However, challenges stemming from the small size of infrared targets and noise interference in complex backgrounds often limit the accuracy of IR small target detection algorithms. Owing to the rapid development of deep learning, numerous convolutional neural network (CNN)-based methods have emerged in recent years, but they are inevitable to encounter the risk of target loss in deep layers due to the use of pooling layers. To address this problem, we present a wavelet transform-based augmented perception network, namely WTAPNet, in this paper. First, an enhancement and enlarge (EE) module is designed to improve the network’s perceptual capability for IR small targets by magnifying image resolution and augmenting target features at the same time. Then, a discrete wavelet transform-based downsampling (DWTD) module and an inverse wavelet transform-based fusion (IWTF) module are proposed. These two modules collaboratively work to extract and fuse multi-scale features, which simultaneously reduces information loss. Finally, a bottom-up path fusion strategy is exploited to highlight and preserve small target features, which associates the lowest-level with the highest-level features and interconnects predictions from different hierarchical levels. Experimental results on NUDT-SIRST dataset and SIRST dataset demonstrate the superiority of our WTAPNet over other state-of-the-art IR small target detection methods in terms of F1-measure, recall and other indicators. The code is publicly available at https://github.com/hhy-NJUST/WTAPNet.