Baojing Pan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Baojing Pan has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Ze Yu
School of Electronic and Information Engineering, Beihang University, Beijing 100191, China
3 shared publications - Jindong Yu
School of Electronics and Information Engineering, Beihang University, Beijing, China
2 shared publications - Chao Li
Research Center for Space Optical Engineering, Harbin Institute of Technology, Harbin 150001, China
1 shared publication Chunsheng Li
School of Electronics and Information Engineering, Beihang University, Beijing, China
1 shared publication- Hanfu Li
Research Center for Space Optical Engineering, Harbin Institute of Technology, Harbin 150001, China
1 shared publication - Hezhi Sun
China Academy of Space Technology
1 shared publication
Latest Publications
- Open Access
- 1 May 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (9) , 1402
From Single-Look to Multi-Temporal SAR Despeckling: A Latent-Space Guided Transfer Learning Approach
Synthetic Aperture Radar (SAR) images are affected by speckle noise, which limits their application in fine object interpretation and quantitative analysis. Recent deep learning-based single-image SAR despeckling methods have made significant progress in spatial structure modeling but struggle to exploit temporal redundancy in multi-temporal data. Existing multi-temporal despeckling methods usually rely on complex spatiotemporal network structures, which are prone to overfitting or excessive smoothing of details when training samples are limited. To address these challenges, this paper proposes a latent-space-guided multi-temporal SAR despeckling method from the perspective of transfer learning and representation alignment, achieving effective knowledge transfer from single-image SAR despeckling to multi-temporal despeckling tasks. The method treats the single-image SAR despeckling task as a knowledge source domain, using stable latent space representations learned from the pre-trained single-image despeckling model as prior constraints. A latent space regularization mechanism is introduced during the training of the multi-temporal despeckling model, thereby establishing an explicit representation bridge between the 2D spatial model and the 3D spatiotemporal model. With this strategy, the multi-temporal model inherits the structural perception capability of the single-image model under limited training samples, improving speckle suppression while effectively maintaining image detail and structural consistency. Additionally, a pure convolutional network architecture is employed to support variable-length multi-temporal sequence input, enhancing the method’s adaptability under different temporal sampling conditions.
- 1 May 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (9) , 1402
Open AccessFrom Single-Look to Multi-Temporal SAR Despeckling: A Latent-Space Guided Transfer Learning Approach
Synthetic Aperture Radar (SAR) images are affected by speckle noise, which limits their application in fine object interpretation and quantitative analysis. Recent deep learning-based single-image SAR despeckling methods have made significant progress in spatial structure modeling but struggle to exploit temporal redundancy in multi-temporal data. Existing multi-temporal despeckling methods usually rely on complex spatiotemporal network structures, which are prone to overfitting or excessive smoothing of details when training samples are limited. To address these challenges, this paper proposes a latent-space-guided multi-temporal SAR despeckling method from the perspective of transfer learning and representation alignment, achieving effective knowledge transfer from single-image SAR despeckling to multi-temporal despeckling tasks. The method treats the single-image SAR despeckling task as a knowledge source domain, using stable latent space representations learned from the pre-trained single-image despeckling model as prior constraints. A latent space regularization mechanism is introduced during the training of the multi-temporal despeckling model, thereby establishing an explicit representation bridge between the 2D spatial model and the 3D spatiotemporal model. With this strategy, the multi-temporal model inherits the structural perception capability of the single-image model under limited training samples, improving speckle suppression while effectively maintaining image detail and structural consistency. Additionally, a pure convolutional network architecture is employed to support variable-length multi-temporal sequence input, enhancing the method’s adaptability under different temporal sampling conditions.
- 30 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-12
Collaborative Optimization for SAR Image Despeckling With Structure Preservation
Speckle is a significant problem in synthetic aperture radar (SAR) image processing, as it deteriorates the texture details and thus limits the precision of its high-accuracy applications. The impact of speckle on SAR images is related to their structural characteristics, which usually classify the images into homogeneous regions, heterogeneous regions, and extremely heterogeneous regions. In the latter two regions, structures such as edges and strong point targets generate a more complex speckle distribution which differs from that in homogeneous regions. Despeckling filters based on multiplicative models perform well in homogeneous regions but exhibit poor performance in the other two types of regions. To solve this problem, a collaborative optimization for SAR image despeckling with structure preservation is proposed in this paper, which leverages the coupling between the despeckling module and the structure extraction module. Guided by structure information, the proposed method achieves better preservation of edges and texture details while removing speckle. Experimental results based on real SAR images demonstrate that the proposed method effectively enhances the despeckling performance in all three types of regions.
- 30 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-12
Collaborative Optimization for SAR Image Despeckling With Structure Preservation
Speckle is a significant problem in synthetic aperture radar (SAR) image processing, as it deteriorates the texture details and thus limits the precision of its high-accuracy applications. The impact of speckle on SAR images is related to their structural characteristics, which usually classify the images into homogeneous regions, heterogeneous regions, and extremely heterogeneous regions. In the latter two regions, structures such as edges and strong point targets generate a more complex speckle distribution which differs from that in homogeneous regions. Despeckling filters based on multiplicative models perform well in homogeneous regions but exhibit poor performance in the other two types of regions. To solve this problem, a collaborative optimization for SAR image despeckling with structure preservation is proposed in this paper, which leverages the coupling between the despeckling module and the structure extraction module. Guided by structure information, the proposed method achieves better preservation of edges and texture details while removing speckle. Experimental results based on real SAR images demonstrate that the proposed method effectively enhances the despeckling performance in all three types of regions.
- Open Access
- 2 April 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (7) , 1260
Sea Clutter Suppression Based on Chaotic Prediction Model by Combining the Generator and Long Short-Term Memory Networks
Sea clutter usually greatly affects the target detection and identification performance of marine surveillance radars. In order to reduce the impact of sea clutter, a novel sea clutter suppression method based on chaos prediction is proposed in this paper. The method combines a generator trained by Generative Adversarial Networks (GAN) with a Long Short-Term Memory (LSTM) network to accomplish sea clutter prediction. By exploiting the generator’s ability to learn the distribution of unlabeled data, the accuracy of sea clutter prediction is improved compared with the classical LSTM-based model. Furthermore, effective suppression of sea clutter and improvements in the signal-to-clutter ratio of echo were achieved through clutter cancellation. Experimental results on real data demonstrated the effectiveness of the proposed method.
- 2 April 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (7) , 1260
Open AccessSea Clutter Suppression Based on Chaotic Prediction Model by Combining the Generator and Long Short-Term Memory Networks
Sea clutter usually greatly affects the target detection and identification performance of marine surveillance radars. In order to reduce the impact of sea clutter, a novel sea clutter suppression method based on chaos prediction is proposed in this paper. The method combines a generator trained by Generative Adversarial Networks (GAN) with a Long Short-Term Memory (LSTM) network to accomplish sea clutter prediction. By exploiting the generator’s ability to learn the distribution of unlabeled data, the accuracy of sea clutter prediction is improved compared with the classical LSTM-based model. Furthermore, effective suppression of sea clutter and improvements in the signal-to-clutter ratio of echo were achieved through clutter cancellation. Experimental results on real data demonstrated the effectiveness of the proposed method.