Changzhong Zou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Changzhong Zou has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
Changwu Zou
4 shared publicationsRui Hu
2 shared publications- Youshen Xia
Anhui University
2 shared publications - Can Zhang
College of Computer and Data Science, Fuzhou University, Fuzhou, China
1 shared publication Fanjing Fu
Fuzhou University
1 shared publication- Liqin Huang
Fuzhou University
1 shared publication
Latest Publications
- Open Access
- 9 July 2026
- Published by Elsevier in SSRN Electronic Journal
Feature Disentanglement and Region-Aware TaskInteraction for Remote Sensing Semantic ChangeDetection
Semantic Change Detection (SCD) entails the concurrent execution of change localization and semantic recognition. However, existing methods primarily face two criticallimitations. Firstly, existing multi-task semantic change detectionmethods commonly employ shared deep task-specific featurespaces to simultaneously handle both semantic segmentation andchange detection tasks. However, this processing strategy maypotentially impact multi-task performance. Secondly, it is diffi-cult for pixel-level interactions to model long-range contextualrelationships and remains susceptible to noise. In this study,we introduce a Feature Disentanglement and Region-AwareTask Interaction (FD-RATI) method. A Feature Disentanglementmodule is designed, consisting of a Change Detail EnhancementBranch (CDEB) and a Semantic-Spatial Interaction Branch (SSIB). CDEB processes change-sensitive boundary details throughcontextual modeling and selective fusion mechanisms. SSIBis responsible for handling high-level semantic features andadaptively aggregating multi-scale semantic context, therebyimproving semantic consistency while preserving fine-grainedstructures. Furthermore, to overcome the limitations of pixellevel interactions, the Region-Aware Task Interaction (RATI)module is proposed. RATI aggregates pixel-level features intolearnable region prototypes and enables global context reasoningwithin a compact region space. After region-level interaction,the features are projected back into the pixel space. FD-RATIachieves state-of-the-art results, as evidenced by comprehensiveevaluations conducted on multiple benchmark datasets.
- 9 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessFeature Disentanglement and Region-Aware TaskInteraction for Remote Sensing Semantic ChangeDetection
Semantic Change Detection (SCD) entails the concurrent execution of change localization and semantic recognition. However, existing methods primarily face two criticallimitations. Firstly, existing multi-task semantic change detectionmethods commonly employ shared deep task-specific featurespaces to simultaneously handle both semantic segmentation andchange detection tasks. However, this processing strategy maypotentially impact multi-task performance. Secondly, it is diffi-cult for pixel-level interactions to model long-range contextualrelationships and remains susceptible to noise. In this study,we introduce a Feature Disentanglement and Region-AwareTask Interaction (FD-RATI) method. A Feature Disentanglementmodule is designed, consisting of a Change Detail EnhancementBranch (CDEB) and a Semantic-Spatial Interaction Branch (SSIB). CDEB processes change-sensitive boundary details throughcontextual modeling and selective fusion mechanisms. SSIBis responsible for handling high-level semantic features andadaptively aggregating multi-scale semantic context, therebyimproving semantic consistency while preserving fine-grainedstructures. Furthermore, to overcome the limitations of pixellevel interactions, the Region-Aware Task Interaction (RATI)module is proposed. RATI aggregates pixel-level features intolearnable region prototypes and enables global context reasoningwithin a compact region space. After region-level interaction,the features are projected back into the pixel space. FD-RATIachieves state-of-the-art results, as evidenced by comprehensiveevaluations conducted on multiple benchmark datasets.
- Open Access
- 20 May 2026
- Published by Elsevier in SSRN Electronic Journal
Remote Sensing Semantic Change Detection via Feature Disentanglement and Region-Aware Task Interaction
Semantic Change Detection (SCD) entails the concurrent execution of change localization and semantic recognition. However, existing methods primarily face two criticallimitations. Firstly, existing multi-task semantic change detectionmethods commonly employ shared deep task-specific featurespaces to simultaneously handle both semantic segmentation andchange detection tasks. However, this processing strategy maypotentially impact multi-task performance. Secondly, reliance onpixel-level interactions struggles to model extended contextualrelationships and remains susceptible to noise. In this work, weintroduce a Feature Disentanglement and Region-Aware TaskInteraction (FD-RATI) method. We introduce a feature disentanglement module composed of two branches to address taskconflicts. The first branch is the CD-Branch, which can preservelow-level details and boundary information to ensure accuratechange localization. The second branch is the SS-Branch, whichcan capture high-level semantic context to enhance recognitionperformance. Furthermore, to overcome the limitations of pixellevel interactions, the Region-Aware Task Interaction (RATI)module is proposed. RATI aggregates pixel-level features intolearnable region prototypes to enable global context reasoningwithin a compact region space. The features refined by regioninteraction are then projected back into the pixel space toimprove feature discriminability. FD-RATI achieves state-of-theart results, as evidenced by comprehensive evaluations conductedon multiple benchmark datasets.
- 20 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessRemote Sensing Semantic Change Detection via Feature Disentanglement and Region-Aware Task Interaction
Semantic Change Detection (SCD) entails the concurrent execution of change localization and semantic recognition. However, existing methods primarily face two criticallimitations. Firstly, existing multi-task semantic change detectionmethods commonly employ shared deep task-specific featurespaces to simultaneously handle both semantic segmentation andchange detection tasks. However, this processing strategy maypotentially impact multi-task performance. Secondly, reliance onpixel-level interactions struggles to model extended contextualrelationships and remains susceptible to noise. In this work, weintroduce a Feature Disentanglement and Region-Aware TaskInteraction (FD-RATI) method. We introduce a feature disentanglement module composed of two branches to address taskconflicts. The first branch is the CD-Branch, which can preservelow-level details and boundary information to ensure accuratechange localization. The second branch is the SS-Branch, whichcan capture high-level semantic context to enhance recognitionperformance. Furthermore, to overcome the limitations of pixellevel interactions, the Region-Aware Task Interaction (RATI)module is proposed. RATI aggregates pixel-level features intolearnable region prototypes to enable global context reasoningwithin a compact region space. The features refined by regioninteraction are then projected back into the pixel space toimprove feature discriminability. FD-RATI achieves state-of-theart results, as evidenced by comprehensive evaluations conductedon multiple benchmark datasets.
- 15 April 2026
- Published by Taylor & Francis in International Journal of Remote Sensing
- Vol. 47 (11) , 4619-4640
Multi-scale dual-domain fusion net based on improved transformer for unsupervised spectral reconstruction
1. Hyperspectral imaging possesses information across the entire electromagnetic spectrum and contains richer information than normal RGB images that only have visible light information. Thus, HSI ...
- 15 April 2026
- Published by Taylor & Francis in International Journal of Remote Sensing
- Vol. 47 (11) , 4619-4640
Multi-scale dual-domain fusion net based on improved transformer for unsupervised spectral reconstruction
1. Hyperspectral imaging possesses information across the entire electromagnetic spectrum and contains richer information than normal RGB images that only have visible light information. Thus, HSI ...
- 1 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 20 (1545598X) , 1-5
A New Semisupervised Method for Detecting Semantic Changes in Remote Sensing Images
The growing availability of high-quality remote sensing imagery has led to increased interest in semantic change detection. Supervised methods for this task have shown significant performance improvements, but acquiring labeled data is often challenging and expensive. To confront this challenge, we propose a semi-supervised approach for semantic change detection in remote sensing images using an innovative teacher-student model. We use a convolutional neural network (CNN) in the teacher model and a fusion design combining CNN and vision transformer in the student model, with the rationale that CNN require fewer training samples compared to vision transformer and fusion network allows us to leverage the advantages of both. To further enhance the model’s performance, we propose a novel data augmentation approach by interchanging bitemporal images as well as their labels. The principle for that is the change from one moment to another and vice versa are two different changes and can therefore be used to augment the training dataset. More importantly, this method does not reduce its reliability because no noise is brought to the remote sensing images. By adopting this approach, we are able to better utilize the small labeled dataset to increase the precision of the model while maintaining the robustness. According to the experimental results, the proposed method outperforms several state-of-the-art methods and achieves an improvement compared to Bi-SRNet in mIoU/SeK/OA of 3.25/4.42/1.78, 6.37/11.72/2.91 on the SECOND and Landsat-SCD dataset, respectively.
- 1 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 20 (1545598X) , 1-5
A New Semisupervised Method for Detecting Semantic Changes in Remote Sensing Images
The growing availability of high-quality remote sensing imagery has led to increased interest in semantic change detection. Supervised methods for this task have shown significant performance improvements, but acquiring labeled data is often challenging and expensive. To confront this challenge, we propose a semi-supervised approach for semantic change detection in remote sensing images using an innovative teacher-student model. We use a convolutional neural network (CNN) in the teacher model and a fusion design combining CNN and vision transformer in the student model, with the rationale that CNN require fewer training samples compared to vision transformer and fusion network allows us to leverage the advantages of both. To further enhance the model’s performance, we propose a novel data augmentation approach by interchanging bitemporal images as well as their labels. The principle for that is the change from one moment to another and vice versa are two different changes and can therefore be used to augment the training dataset. More importantly, this method does not reduce its reliability because no noise is brought to the remote sensing images. By adopting this approach, we are able to better utilize the small labeled dataset to increase the precision of the model while maintaining the robustness. According to the experimental results, the proposed method outperforms several state-of-the-art methods and achieves an improvement compared to Bi-SRNet in mIoU/SeK/OA of 3.25/4.42/1.78, 6.37/11.72/2.91 on the SECOND and Landsat-SCD dataset, respectively.
- Open Access
- 2 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 15 (19391404) , 9497-9508
Enhanced Channel Attention Network With Cross-Layer Feature Fusion for Spectral Reconstruction in the Presence of Gaussian Noise
Spectral reconstruction from RGB images has made significant progress. Previous works usually utilized the noise-free RGB images as input to reconstruct the corresponding hyperspectral images (HSIs). However, due to instrumental limitation or atmospheric interference, it is inevitable to suffer from noise (e.g., Gaussian noise) in the actual image acquisition process, which further increases the difficulty of spectral reconstruction. In this paper, we propose an enhanced channel attention network (ECANet) to learn a nonlinear mapping from noisy RGB images to clean HSIs. The backbone of our proposed ECANet is stacked with multiple enhanced channel attention (ECA) blocks. The ECA block is the dual residual version of channel attention block, which makes the network focus on key auxiliary information and features that are more conducive to spectral reconstruction. For the case that the input RGB images are disturbed by Gaussian noise, cross-layer feature fusion (CLFF) unit is used to concatenate the multiple feature maps at different depths for more powerful feature representations. In addition, we design a novel combined loss function as the constraint of the ECANet to achieve more accurate reconstruction result. Experimental results on two HSI benchmarks, CAVE and NTIRE 2020, demonstrate that the effectiveness of our method in terms of both visual and quantitative over other state-of-the-art methods.
- 2 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 15 (19391404) , 9497-9508
Open AccessEnhanced Channel Attention Network With Cross-Layer Feature Fusion for Spectral Reconstruction in the Presence of Gaussian Noise
Spectral reconstruction from RGB images has made significant progress. Previous works usually utilized the noise-free RGB images as input to reconstruct the corresponding hyperspectral images (HSIs). However, due to instrumental limitation or atmospheric interference, it is inevitable to suffer from noise (e.g., Gaussian noise) in the actual image acquisition process, which further increases the difficulty of spectral reconstruction. In this paper, we propose an enhanced channel attention network (ECANet) to learn a nonlinear mapping from noisy RGB images to clean HSIs. The backbone of our proposed ECANet is stacked with multiple enhanced channel attention (ECA) blocks. The ECA block is the dual residual version of channel attention block, which makes the network focus on key auxiliary information and features that are more conducive to spectral reconstruction. For the case that the input RGB images are disturbed by Gaussian noise, cross-layer feature fusion (CLFF) unit is used to concatenate the multiple feature maps at different depths for more powerful feature representations. In addition, we design a novel combined loss function as the constraint of the ECANet to achieve more accurate reconstruction result. Experimental results on two HSI benchmarks, CAVE and NTIRE 2020, demonstrate that the effectiveness of our method in terms of both visual and quantitative over other state-of-the-art methods.
- 25 April 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 19 (1545598X) , 1-5
An Effective Road Centerline Extraction Method From VHR
Road digitizing is a labor-intensive task. For reducing labor cost, this letter presents an effective semiautomatic road delineation method based on geodesic distance field (GDF) and piecewise polygon fitting. The main components of the approach are optimal circle calculation, soft road center kernel density estimation (KDE), fast GDF generation, and piecewise polygon fitting model. First, the adaptive circular template is proposed to automatically measure the road width. Next, the soft road center kernel density is estimated for fast GDF generation, which supports the extraction of road centerline between two adjacent seeds. Finally, piecewise polygon fitting is used to refine the road centerline. Extensive experiments demonstrate that the proposed algorithm is efficient and robust in road delineation. The proposed approach takes almost the same time to extract any length of road segment given fixed image size, and no hyperparameters needs to be set.
- 25 April 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 19 (1545598X) , 1-5
An Effective Road Centerline Extraction Method From VHR
Road digitizing is a labor-intensive task. For reducing labor cost, this letter presents an effective semiautomatic road delineation method based on geodesic distance field (GDF) and piecewise polygon fitting. The main components of the approach are optimal circle calculation, soft road center kernel density estimation (KDE), fast GDF generation, and piecewise polygon fitting model. First, the adaptive circular template is proposed to automatically measure the road width. Next, the soft road center kernel density is estimated for fast GDF generation, which supports the extraction of road centerline between two adjacent seeds. Finally, piecewise polygon fitting is used to refine the road centerline. Extensive experiments demonstrate that the proposed algorithm is efficient and robust in road delineation. The proposed approach takes almost the same time to extract any length of road segment given fixed image size, and no hyperparameters needs to be set.
- 1 February 2018
- Published by Elsevier in Signal Processing: Image Communication
- Vol. 60, 29-41
Bayesian dictionary learning for hyperspectral image super resolution in mixed Poisson–Gaussian noise
- 1 February 2018
- Published by Elsevier in Signal Processing: Image Communication
- Vol. 60, 29-41
Bayesian dictionary learning for hyperspectral image super resolution in mixed Poisson–Gaussian noise
- 1 January 2018
- Published by Elsevier in Neurocomputing
- Vol. 275, 430-437
Restoration of hyperspectral image contaminated by Poisson noise using spectral unmixing
- 1 January 2018
- Published by Elsevier in Neurocomputing
- Vol. 275, 430-437
Restoration of hyperspectral image contaminated by Poisson noise using spectral unmixing
- 20 January 2010
- Published by China Science Publishing & Media Ltd. in Journal of Computer Applications
- Vol. 30 (1) , 127-129
Node data prediction based on SVR in wireless sensor network
- 20 January 2010
- Published by China Science Publishing & Media Ltd. in Journal of Computer Applications
- Vol. 30 (1) , 127-129
Node data prediction based on SVR in wireless sensor network