Chang Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Chang Xu has a total of 39 co-authors. Here are the top co-authors by number of shared publications:
- Qingwu Li
College of Information Science and Engineering, Hohai University, Changzhou, China
12 shared publications - Yaqin Zhou
College of Internet of Things Engineering, Hohai University, Changzhou 213022, China
5 shared publications - Dabing Yu
The Key Laboratory of Jinan Digital Twins and Intelligent Water Conservancy, Shandong Water Conservancy Survey and Design Institute Co., Jinan 250013, China
4 shared publications - Qingkai Zhou
College of Internet of Things Engineering, Hohai University, Jiangsu, China
3 shared publications Yaqin Zhou
3 shared publications- Yunpeng Ma
Jiangsu Key Laboratory of Power Transmission and Distribution Equipment Technology, Hohai University, Changzhou, China
3 shared publications
Latest Publications
- Open Access
- 23 October 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (21) , 3870
Mathematical Analysis of Backdoor-Induced Melanoma Risk in Modality-Heterogeneous Federated Learning
Aggregate accuracy and conventional attack success rates cannot distinguish pre-existing diagnostic errors from melanoma false negatives induced by a trigger. We evaluate melanoma-to-nevus risk downgrading in source-heterogeneous federated learning using paired clean and triggered predictions. Four source-defined clients represent clinical photography, smartphone imaging, and dermoscopy. The primary endpoint, trigger-induced paired false negatives (trigFN), counts clean-correct melanoma cases that change to a nevus prediction after triggering. In centralized screening, sticker and blend triggers produce 35 and 31 trigFN among 77 melanoma cases, whereas a square trigger produces none. Across four seeds, Full-Phase poisoning produces trigFN and a -percentage-point melanoma-sensitivity loss; Late-Phase poisoning produces trigFN and a -point loss. Under Full-Phase poisoning, NormClip with reduces trigFN to and the sensitivity loss to points, while mean balanced accuracy (BACC) changes from to . External evaluation shows smaller within-model trigger-associated losses with clipping, although absolute melanoma sensitivity on untriggered images is lower. Paired clinical endpoints reveal harm that aggregate utility can obscure. Within the evaluated four-client, single-attacker, non-adaptive setting, norm-constrained aggregation reduces but does not eliminate harm.- 23 October 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (21) , 3870
Open AccessMathematical Analysis of Backdoor-Induced Melanoma Risk in Modality-Heterogeneous Federated Learning
Aggregate accuracy and conventional attack success rates cannot distinguish pre-existing diagnostic errors from melanoma false negatives induced by a trigger. We evaluate melanoma-to-nevus risk downgrading in source-heterogeneous federated learning using paired clean and triggered predictions. Four source-defined clients represent clinical photography, smartphone imaging, and dermoscopy. The primary endpoint, trigger-induced paired false negatives (trigFN), counts clean-correct melanoma cases that change to a nevus prediction after triggering. In centralized screening, sticker and blend triggers produce 35 and 31 trigFN among 77 melanoma cases, whereas a square trigger produces none. Across four seeds, Full-Phase poisoning produces trigFN and a -percentage-point melanoma-sensitivity loss; Late-Phase poisoning produces trigFN and a -point loss. Under Full-Phase poisoning, NormClip with reduces trigFN to and the sensitivity loss to points, while mean balanced accuracy (BACC) changes from to . External evaluation shows smaller within-model trigger-associated losses with clipping, although absolute melanoma sensitivity on untriggered images is lower. Paired clinical endpoints reveal harm that aggregate utility can obscure. Within the evaluated four-client, single-attacker, non-adaptive setting, norm-constrained aggregation reduces but does not eliminate harm. - Open Access
- 10 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 92-98
Anti-Aliasing Gaussian Splatting 3D Reconstruction Method for Power Equipment
- 10 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 92-98
Open AccessAnti-Aliasing Gaussian Splatting 3D Reconstruction Method for Power Equipment
- 11 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35 (10577149) , 5048-5063
Asymmetric Feature Consistency Reinforcement Network for Visual-Depth-Thermal Salient Object Detection and a New Benchmark
Fusing Visual-Depth-Thermal (VDT) data holds immense potential for robust Salient Object Detection (SOD) in complex environments. However, current research is constrained by dataset scarcity and the limitations of symmetric direct fusion strategies. To address these gaps, we first construct a comprehensive benchmark named LiTR-2654, comprising 2,654 spatially aligned VDT image triplets captured via LiDAR and dual-modality cameras. This dataset features high diversity and reduced center bias, designed to advance practical applications. With this benchmark, we propose the Asymmetric Feature Consistency Reinforcement Network (AFCRNet), effectively utilizing triple-modality cues to achieve accurate SOD. AFCRNet comprises mainly two core technical innovations: “Unify-then-Integrate” fusion strategy investigates modality-complementary information and context-guided decoder module enables the common focus of multi-level features. Specifically, cross-level thermal and visual features are densely interacted to obtain consistent feature representations. Meanwhile, taking depth features as supplements, same-level triple-modality features are integrated with the attention mechanism, significantly suppressing complex background interference and highlighting salient objects. To further improve the segmentation accuracy, high-level contextual information is introduced into multi-level features to accurately distinguish salient objects, and edge supervision is also utilized to optimize the object contour. Comprehensive analysis of different methods is conducted on published and self-built datasets, demonstrating the superiority of the proposed method. The constructed novel benchmark will be made publicly available at: github.com/215HH/LiTR-2654.
- 11 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35 (10577149) , 5048-5063
Asymmetric Feature Consistency Reinforcement Network for Visual-Depth-Thermal Salient Object Detection and a New Benchmark
Fusing Visual-Depth-Thermal (VDT) data holds immense potential for robust Salient Object Detection (SOD) in complex environments. However, current research is constrained by dataset scarcity and the limitations of symmetric direct fusion strategies. To address these gaps, we first construct a comprehensive benchmark named LiTR-2654, comprising 2,654 spatially aligned VDT image triplets captured via LiDAR and dual-modality cameras. This dataset features high diversity and reduced center bias, designed to advance practical applications. With this benchmark, we propose the Asymmetric Feature Consistency Reinforcement Network (AFCRNet), effectively utilizing triple-modality cues to achieve accurate SOD. AFCRNet comprises mainly two core technical innovations: “Unify-then-Integrate” fusion strategy investigates modality-complementary information and context-guided decoder module enables the common focus of multi-level features. Specifically, cross-level thermal and visual features are densely interacted to obtain consistent feature representations. Meanwhile, taking depth features as supplements, same-level triple-modality features are integrated with the attention mechanism, significantly suppressing complex background interference and highlighting salient objects. To further improve the segmentation accuracy, high-level contextual information is introduced into multi-level features to accurately distinguish salient objects, and edge supervision is also utilized to optimize the object contour. Comprehensive analysis of different methods is conducted on published and self-built datasets, demonstrating the superiority of the proposed method. The constructed novel benchmark will be made publicly available at: github.com/215HH/LiTR-2654.
- 10 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-17
Frequency Introduced Cascade Feature Fusion Network via Knowledge Distillation for Infrared Small Target Detection
Infrared small target detection (IRSTD) plays a crucial role in numerous applications and has benefited significantly from recent advances in deep learning. However, existing methods often face two major challenges: (1) difficulty in detecting non-salient small targets embedded in complex backgrounds, and (2) substantial computational costs during inference, which limit practical deployment. To address these issues, we propose the Frequency Introduced Cascade Feature Fusion Network (FICFFNet), which leverages high-frequency information from infrared images to enhance target saliency, and incorporates a knowledge distillation (KD) framework to balance detection accuracy and efficiency. Specifically, infrared and frequency features are jointly extracted and adaptively fused, enabling effective integration of spatial and frequency-domain cues and improving localization of tiny objects with local temperature variations. Adjacent-level features are further aggregated to combine low-level spatial details with high-level contextual semantics, ensuring robust multi-scale prediction. A lightweight student network is constructed by simplifying the architecture, while multiple distillation paths transfer crucial knowledge from the dual-stream features and prediction maps to the single-stream representation. We also construct two infrared tiny bird detection datasets by collecting and annotating infrared images from transformer substations. Extensive experiments on both self-built and public datasets demonstrate that FICFFNet achieves superior detection performance while maintaining high computational efficiency. Our code will be made public at: github.com/HHUyxt/IRTBD.
- 10 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-17
Frequency Introduced Cascade Feature Fusion Network via Knowledge Distillation for Infrared Small Target Detection
Infrared small target detection (IRSTD) plays a crucial role in numerous applications and has benefited significantly from recent advances in deep learning. However, existing methods often face two major challenges: (1) difficulty in detecting non-salient small targets embedded in complex backgrounds, and (2) substantial computational costs during inference, which limit practical deployment. To address these issues, we propose the Frequency Introduced Cascade Feature Fusion Network (FICFFNet), which leverages high-frequency information from infrared images to enhance target saliency, and incorporates a knowledge distillation (KD) framework to balance detection accuracy and efficiency. Specifically, infrared and frequency features are jointly extracted and adaptively fused, enabling effective integration of spatial and frequency-domain cues and improving localization of tiny objects with local temperature variations. Adjacent-level features are further aggregated to combine low-level spatial details with high-level contextual semantics, ensuring robust multi-scale prediction. A lightweight student network is constructed by simplifying the architecture, while multiple distillation paths transfer crucial knowledge from the dual-stream features and prediction maps to the single-stream representation. We also construct two infrared tiny bird detection datasets by collecting and annotating infrared images from transformer substations. Extensive experiments on both self-built and public datasets demonstrate that FICFFNet achieves superior detection performance while maintaining high computational efficiency. Our code will be made public at: github.com/HHUyxt/IRTBD.
- Open Access
Thermal Imaging-Based Abnormal Heating Detection for High-Voltage Power Equipment
Thermal infrared imaging could detect hidden faults in various types of high-voltage power equipment, which is of great significance for power inspections. However, there are still certain issues with thermal-imaging-based abnormal heating detection methods due to varying appearances of abnormal regions and complex temperature interference from backgrounds. To solve these problems, a contour-based instance segmentation network is first proposed to utilize thermal (T) and visual (RGB) images, realizing high-accuracy segmentation against complex and changing environments. Specifically, modality-specific features are encoded via two-stream backbones and fused in spatial, channel, and frequency domains. In this way, modality differences are well handled, and effective complementary information is extracted for object detection and contour initialization. The transformer decoder is further utilized to explore the long-range relationships between contour points with background points, and to achieve the deformation of contour points. Then, the auto-encoder-based reconstruction network is developed to learn the distribution of power equipment using the proposed random argument strategy. Meanwhile, the UNet-like discriminative network directly explores the differences between the reconstructed and original image, capturing the deviation of poor reconstruction regions for abnormal heating detection. Many images are acquired in transformer substations with different weathers and day times to build the datasets with pixel-level annotation. Several extensive experiments are conducted for qualitative and quantitative evaluation, while the comparison results fully prove the effectiveness and robustness of the proposed instance segmentation method. The practicality and performance of the proposed abnormal heating detection method are evaluated on image patches with different kinds of insulators.
Open AccessThermal Imaging-Based Abnormal Heating Detection for High-Voltage Power Equipment
Thermal infrared imaging could detect hidden faults in various types of high-voltage power equipment, which is of great significance for power inspections. However, there are still certain issues with thermal-imaging-based abnormal heating detection methods due to varying appearances of abnormal regions and complex temperature interference from backgrounds. To solve these problems, a contour-based instance segmentation network is first proposed to utilize thermal (T) and visual (RGB) images, realizing high-accuracy segmentation against complex and changing environments. Specifically, modality-specific features are encoded via two-stream backbones and fused in spatial, channel, and frequency domains. In this way, modality differences are well handled, and effective complementary information is extracted for object detection and contour initialization. The transformer decoder is further utilized to explore the long-range relationships between contour points with background points, and to achieve the deformation of contour points. Then, the auto-encoder-based reconstruction network is developed to learn the distribution of power equipment using the proposed random argument strategy. Meanwhile, the UNet-like discriminative network directly explores the differences between the reconstructed and original image, capturing the deviation of poor reconstruction regions for abnormal heating detection. Many images are acquired in transformer substations with different weathers and day times to build the datasets with pixel-level annotation. Several extensive experiments are conducted for qualitative and quantitative evaluation, while the comparison results fully prove the effectiveness and robustness of the proposed instance segmentation method. The practicality and performance of the proposed abnormal heating detection method are evaluated on image patches with different kinds of insulators.
- 24 February 2023
- Published by Elsevier in Optics and Lasers in Engineering
- Vol. 165
Color Polarization Image Super-Resolution Reconstruction via a Cross-Branch Supervised Learning Strategy
- 24 February 2023
- Published by Elsevier in Optics and Lasers in Engineering
- Vol. 165
Color Polarization Image Super-Resolution Reconstruction via a Cross-Branch Supervised Learning Strategy
- Open Access
- 1 January 2023
- Published by Wiley in Journal of Sensors
- Vol. 2023 (1)
Direction Consistency‐Guided Lightweight Power Line Detection Network for Aerial Images
Accurate detection of power lines in aerial images is of great significance in ensuring grid security. However, complex power line scenarios and the thin and light structure of power lines both make it difficult to detect power lines accurately. Most of the existing approaches use traditional deep learning methods, using networks with a large number of parameters, computation, and memory occupation, thus making them not lightweight enough to perform on mobile devices. Based on this, a lightweight power line detection network based on direction consistency and location attention is proposed. The network is designed with a coordinate‐aware feature extraction layer, which performs feature extraction by four‐layer stacking to achieve faster inference speed while ensuring the network has fewer parameters. This layer is also able to sense the coordinates of the center pixel of the convolution in the image during the convolution process, thus preserving the location information of the power lines. In order to enhance the power line representation, a two‐stage context‐guided module is later utilized to simultaneously learn local features, surrounding context, and global context. Then, the features are input into a Gaussian kernel estimation module and features are aggregated in the corresponding directions through Gaussian kernels of eight different directions. The main directions of the power lines in the image and the corresponding Gaussian convolution kernels are obtained by filtering the feature responses. In addition, a kernel‐guided decoder module is proposed to take advantage of the estimated power line features in the main direction of Gaussian kernel aggregation. This module can effectively enhance the power line representation and maintain the continuity of power lines. Meanwhile, low‐level features are introduced to recover the edge details to realize high performance in distinguishing dense power lines. Both ablation experiments and comparison experiments on the transmission towers and power lines aerial‐image and Power Line Aerial Image Dataset show that the proposed power line detection network has a good segmentation performance in complex scenarios. The proposed method performs the best in the comparison experiments, improving over the suboptimal method by 3.51% on average for the max F‐measure metric.
- 1 January 2023
- Published by Wiley in Journal of Sensors
- Vol. 2023 (1)
Open AccessDirection Consistency‐Guided Lightweight Power Line Detection Network for Aerial Images
Accurate detection of power lines in aerial images is of great significance in ensuring grid security. However, complex power line scenarios and the thin and light structure of power lines both make it difficult to detect power lines accurately. Most of the existing approaches use traditional deep learning methods, using networks with a large number of parameters, computation, and memory occupation, thus making them not lightweight enough to perform on mobile devices. Based on this, a lightweight power line detection network based on direction consistency and location attention is proposed. The network is designed with a coordinate‐aware feature extraction layer, which performs feature extraction by four‐layer stacking to achieve faster inference speed while ensuring the network has fewer parameters. This layer is also able to sense the coordinates of the center pixel of the convolution in the image during the convolution process, thus preserving the location information of the power lines. In order to enhance the power line representation, a two‐stage context‐guided module is later utilized to simultaneously learn local features, surrounding context, and global context. Then, the features are input into a Gaussian kernel estimation module and features are aggregated in the corresponding directions through Gaussian kernels of eight different directions. The main directions of the power lines in the image and the corresponding Gaussian convolution kernels are obtained by filtering the feature responses. In addition, a kernel‐guided decoder module is proposed to take advantage of the estimated power line features in the main direction of Gaussian kernel aggregation. This module can effectively enhance the power line representation and maintain the continuity of power lines. Meanwhile, low‐level features are introduced to recover the edge details to realize high performance in distinguishing dense power lines. Both ablation experiments and comparison experiments on the transmission towers and power lines aerial‐image and Power Line Aerial Image Dataset show that the proposed power line detection network has a good segmentation performance in complex scenarios. The proposed method performs the best in the comparison experiments, improving over the suboptimal method by 3.51% on average for the max F‐measure metric.
- 23 October 2022
- Published by Elsevier in Knowledge-Based Systems
- Vol. 258
Asymmetric cross-modal activation network for RGB-T salient object detection
- 23 October 2022
- Published by Elsevier in Knowledge-Based Systems
- Vol. 258
Asymmetric cross-modal activation network for RGB-T salient object detection
- 21 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 33 (4) , 1577-1592
Dual-Space Graph-Based Interaction Network for RGB-Thermal Semantic Segmentation in Electric Power Scene
Real-time scene comprehension is the basis for automatic electric power inspection. However, existing RGB-based scene comprehension methods may achieve unsatisfied performance when dealing with complex scenarios, insufficient illumination or occluded appearances. To solve this problem, by cooperating visual and thermal images, the Dual-Space Graph-based Interaction Network (DSGBINet) is proposed to achieve all-day time semantic segmentation of power equipment in high-voltage power transmission line and electric transformer substation scenes. Specifically, modality-specific features are first extracted via two separate backbone networks with the same architecture. Multi-modality high-level features are first fused via long-range relationship in coordinate space. Then, multi-modality features from regular grids are further clustered and assigned to vertices in feature space. Cross-graph and inner-graph regional relations are utilized for reasoning and enhancement, which could exploit the mutual benefits and extract rich contextual information in a semantic view. Furthermore, to overcome the huge scale difference and the inherent characteristic of thermal images, the idea of multi-task learning is integrated into the decoding process. The edge detection and semantic segmentation are achieved collaboratively, which could segment the different power equipment more accurately and completely. The comparative and ablation experiments on the proposed two RGB-T semantic segmentation datasets evaluate the effectiveness and robustness of the proposed network compared with existing state-of-the-art methods. The extended experiments on the public datasets further demonstrate the superiority of the proposed method. Our dataset and code will be released at: https://github.com/hhujiang/DSGBINet.
- 21 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 33 (4) , 1577-1592
Dual-Space Graph-Based Interaction Network for RGB-Thermal Semantic Segmentation in Electric Power Scene
Real-time scene comprehension is the basis for automatic electric power inspection. However, existing RGB-based scene comprehension methods may achieve unsatisfied performance when dealing with complex scenarios, insufficient illumination or occluded appearances. To solve this problem, by cooperating visual and thermal images, the Dual-Space Graph-based Interaction Network (DSGBINet) is proposed to achieve all-day time semantic segmentation of power equipment in high-voltage power transmission line and electric transformer substation scenes. Specifically, modality-specific features are first extracted via two separate backbone networks with the same architecture. Multi-modality high-level features are first fused via long-range relationship in coordinate space. Then, multi-modality features from regular grids are further clustered and assigned to vertices in feature space. Cross-graph and inner-graph regional relations are utilized for reasoning and enhancement, which could exploit the mutual benefits and extract rich contextual information in a semantic view. Furthermore, to overcome the huge scale difference and the inherent characteristic of thermal images, the idea of multi-task learning is integrated into the decoding process. The edge detection and semantic segmentation are achieved collaboratively, which could segment the different power equipment more accurately and completely. The comparative and ablation experiments on the proposed two RGB-T semantic segmentation datasets evaluate the effectiveness and robustness of the proposed network compared with existing state-of-the-art methods. The extended experiments on the public datasets further demonstrate the superiority of the proposed method. Our dataset and code will be released at: https://github.com/hhujiang/DSGBINet.
- 5 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 60 (01962892) , 1-17
A Cross-Level Spectral–Spatial Joint Encode Learning Framework for Imbalanced Hyperspectral Image Classification
Convolutional neural networks (CNNs) have dominated the research of hyperspectral image (HSI) classification, attributing to the superior feature representation capacity. Fast patch-free global learning (FPGA) as a fast learning framework for HSI classification has received wide interest. Despite their promising results from the perspective of fast inference, recent works have difficulty modeling spectral–spatial relationships with imbalanced samples. In this article, we revisit the encoder–decoder-based fully convolutional network (FCN) and propose a cross-level spectral–spatial joint encoding (CLSJE) framework for imbalanced HSI classification. First, a multiscale input encoder and multiple-to-one multiscale features connection are introduced to obtain abundant features and facilitate multiscale contextual information flow between the encoder and the decoder. Second, in the encoder layer, we propose the spectral–spatial joint attention (SSJA) mechanism consisting of high-frequency spatial attention (HFSA) and spectral-transform channel attention (STCA). HFSA and STCA encode spectral–spatial features jointly to improve the learning of the discriminative spectral–spatial features. Powered by these two components, CLSJE enjoys a high capability to capture both spatial and spectral dependencies for HSI classification. Besides, a class-proportion sampling strategy is developed to increase the attention to insufficiency samples. Extensive experiments demonstrate the superiority of our proposed CLSJE both at classification accuracy and inference speed, and show the state-of-the-art results on four benchmark datasets. Code can be obtained at https://github.com/yudadabing/CLSJE .
- 5 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 60 (01962892) , 1-17
A Cross-Level Spectral–Spatial Joint Encode Learning Framework for Imbalanced Hyperspectral Image Classification
Convolutional neural networks (CNNs) have dominated the research of hyperspectral image (HSI) classification, attributing to the superior feature representation capacity. Fast patch-free global learning (FPGA) as a fast learning framework for HSI classification has received wide interest. Despite their promising results from the perspective of fast inference, recent works have difficulty modeling spectral–spatial relationships with imbalanced samples. In this article, we revisit the encoder–decoder-based fully convolutional network (FCN) and propose a cross-level spectral–spatial joint encoding (CLSJE) framework for imbalanced HSI classification. First, a multiscale input encoder and multiple-to-one multiscale features connection are introduced to obtain abundant features and facilitate multiscale contextual information flow between the encoder and the decoder. Second, in the encoder layer, we propose the spectral–spatial joint attention (SSJA) mechanism consisting of high-frequency spatial attention (HFSA) and spectral-transform channel attention (STCA). HFSA and STCA encode spectral–spatial features jointly to improve the learning of the discriminative spectral–spatial features. Powered by these two components, CLSJE enjoys a high capability to capture both spatial and spectral dependencies for HSI classification. Besides, a class-proportion sampling strategy is developed to increase the attention to insufficiency samples. Extensive experiments demonstrate the superiority of our proposed CLSJE both at classification accuracy and inference speed, and show the state-of-the-art results on four benchmark datasets. Code can be obtained at https://github.com/yudadabing/CLSJE .