Jiansheng Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jiansheng Wang has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Qingli Li
East China Normal University, Shanghai, China
6 shared publications - Yan Wang
East China Normal University, Shanghai, China
3 shared publications Fanli Lin
Shanghai Key Laboratory of Multidimensional Information Processing East China Normal University Shanghai China
2 shared publications- Haofei Song
East China Normal University
2 shared publications Huaiyuan Wang
Shanghai Key Laboratory of Multidimensional Information Processing East China Normal University Shanghai China
2 shared publications- Siyuan Xu
East China Normal University
2 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 12669-12678
Towards Reliable Stain Transfer: An Iterative Data-Model Co-Optimization Framework Based on Multimodal Expert-Guided Assessment
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 12669-12678
Open AccessTowards Reliable Stain Transfer: An Iterative Data-Model Co-Optimization Framework Based on Multimodal Expert-Guided Assessment
- Open Access
- 5 October 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Planets
- Vol. 131 (10)
Lunar Mineralogical Evidence Provided by Chang'e‐5 Lunar Soil Microscopic Images and Raman Spectroscopy
The Chang'e‐5 lunar regolith samples are returned samples of young mare basalt, possessing significant scientific research value. This paper performs nondestructive analysis on one of the Chang'e‐5 basalt samples using scanning electron microscopy and micro‐Raman spectroscopy. Through the analysis of the mineralogical characteristics, modal abundance, and impact effects of the sample, it is found that the sample is relatively consistent in composition with most Chang'e‐5 lunar regolith samples, being low‐titanium mare basalt that has undergone a certain degree of impact events. The analysis of quartz and plagioclase indicates that the impact pressure experienced by the young mare basalt is limited to a maximum of 20–25 GPa, with a minimum impact pressure of at least 5.2 GPa. Additionally, both this sample and the Chang'e‐5 lunar regolith samples should likely originate from a low‐titanium magma source, which may possibly have formed through the partial melting of a mixed magma composed of early refractory cumulates and late ilmenite‐bearing cumulates, followed by fractional crystallization in a shallow magma reservoir before being erupted onto the lunar surface, and is compositionally distinct from Apollo and CE‐6 basalts as an evolved, low‐Ti late‐stage basalts. This research can provide mineralogical evidence for understanding the thermal evolution of the Moon, particularly the late‐stage volcanic activity and internal thermodynamic changes.
- 5 October 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Planets
- Vol. 131 (10)
Open AccessLunar Mineralogical Evidence Provided by Chang'e‐5 Lunar Soil Microscopic Images and Raman Spectroscopy
The Chang'e‐5 lunar regolith samples are returned samples of young mare basalt, possessing significant scientific research value. This paper performs nondestructive analysis on one of the Chang'e‐5 basalt samples using scanning electron microscopy and micro‐Raman spectroscopy. Through the analysis of the mineralogical characteristics, modal abundance, and impact effects of the sample, it is found that the sample is relatively consistent in composition with most Chang'e‐5 lunar regolith samples, being low‐titanium mare basalt that has undergone a certain degree of impact events. The analysis of quartz and plagioclase indicates that the impact pressure experienced by the young mare basalt is limited to a maximum of 20–25 GPa, with a minimum impact pressure of at least 5.2 GPa. Additionally, both this sample and the Chang'e‐5 lunar regolith samples should likely originate from a low‐titanium magma source, which may possibly have formed through the partial melting of a mixed magma composed of early refractory cumulates and late ilmenite‐bearing cumulates, followed by fractional crystallization in a shallow magma reservoir before being erupted onto the lunar surface, and is compositionally distinct from Apollo and CE‐6 basalts as an evolved, low‐Ti late‐stage basalts. This research can provide mineralogical evidence for understanding the thermal evolution of the Moon, particularly the late‐stage volcanic activity and internal thermodynamic changes.
- Open Access
- 17 September 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (18) , 3202
ESA-MDN: An Ensemble Self-Attention Enhanced Mixture Density Framework for UAV Multispectral Water Quality Parameter Retrieval
Urban rivers, as crucial components of ecosystems, serve multiple functions, including flood control, drainage, and landscape services. However, with the acceleration of urbanization, factors such as industrial wastewater discharge, domestic sewage leakage, and surface runoff pollution have led to increasingly severe degradation of water quality in urban rivers. Unmanned aerial vehicle (UAV) remote sensing technology, with its sub-meter spatial resolution and operational flexibility, demonstrates significant advantages in the detailed monitoring of complex urban water systems. This study proposes an Ensemble Self-Attention Enhanced Mixture Density Network (ESA-MDN), which integrate an ensemble learning framework with a mixture density network and incorporates a self-attention mechanism for feature enhancement. This approach better captures the nonlinear relationships between water quality parameters and remote sensing features, achieving high-precision modeling of water quality parameter distributions. The resulting spatiotemporal distribution maps provide valuable support for pollution source identification and management decision making. The model successfully retrieved five water quality parameters, Chl-a, TSS, COD, TP, and DO, and validation metrics such as R2, RMSE, MAE, MSE, MAPE, bias, and slope were utilized. Key metrics for the ESA-MDN test set were as follows: Chl-a (R2 = 0.98, RMSE = 0.31), TSS (R2 = 0.93, RMSE = 0.27), COD (R2 = 0.93, RMSE = 0.39), TP (R2 = 0.99, RMSE = 0.02), and DO (R2 = 0.88, RMSE = 0.1). The results indicated that ESA-MDN can effectively extract water quality parameters from multispectral remote sensing data, with the generated spatiotemporal water quality distribution maps providing crucial support for pollution source identification and emergency response decision making.
- 17 September 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (18) , 3202
Open AccessESA-MDN: An Ensemble Self-Attention Enhanced Mixture Density Framework for UAV Multispectral Water Quality Parameter Retrieval
Urban rivers, as crucial components of ecosystems, serve multiple functions, including flood control, drainage, and landscape services. However, with the acceleration of urbanization, factors such as industrial wastewater discharge, domestic sewage leakage, and surface runoff pollution have led to increasingly severe degradation of water quality in urban rivers. Unmanned aerial vehicle (UAV) remote sensing technology, with its sub-meter spatial resolution and operational flexibility, demonstrates significant advantages in the detailed monitoring of complex urban water systems. This study proposes an Ensemble Self-Attention Enhanced Mixture Density Network (ESA-MDN), which integrate an ensemble learning framework with a mixture density network and incorporates a self-attention mechanism for feature enhancement. This approach better captures the nonlinear relationships between water quality parameters and remote sensing features, achieving high-precision modeling of water quality parameter distributions. The resulting spatiotemporal distribution maps provide valuable support for pollution source identification and management decision making. The model successfully retrieved five water quality parameters, Chl-a, TSS, COD, TP, and DO, and validation metrics such as R2, RMSE, MAE, MSE, MAPE, bias, and slope were utilized. Key metrics for the ESA-MDN test set were as follows: Chl-a (R2 = 0.98, RMSE = 0.31), TSS (R2 = 0.93, RMSE = 0.27), COD (R2 = 0.93, RMSE = 0.39), TP (R2 = 0.99, RMSE = 0.02), and DO (R2 = 0.88, RMSE = 0.1). The results indicated that ESA-MDN can effectively extract water quality parameters from multispectral remote sensing data, with the generated spatiotemporal water quality distribution maps providing crucial support for pollution source identification and emergency response decision making.
- Open Access
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 4475-4484
GeNSeg-Net: A General Segmentation Framework for Any Nucleus in Immunohistochemistry Images
Immunohistochemistry (IHC) plays a crucial role in understanding disease mechanisms, diagnosing pathology and guiding treatment decisions. The precise analysis heavily depends on accurate nucleus segmentation. However, segmentation is challenging due to significant inter- and intra-nucleus variability in morphology and distribution, stemming from inherent characteristics, imaging techniques, tissue differences and other factors. While current deep learning-based methods have shown promising results, their generalization performance is limited, inevitably requiring specific training data. To address the problem, we propose a novel Ge neral framework for Nucleus Seg mentation in IHC images (GeNSeg-Net). GeNSeg-Net effectively segments nuclei across diverse tissue types and imaging techniques with high variability using a small subset for training. It comprises an enhancement model and a segmentation model. Initially, all nuclei are enhanced to a uniform morphology with distinct features by the enhancement model through generation. The subsequent segmentation task is thereby simplified, leading to higher accuracy. We design a lightweight generator and discriminator to improve both enhancement quality and computational efficiency. Extensive experiments demonstrate the effectiveness of each component within GeNSeg-Net. Compared to existing methods, GeNSeg-Net achieves state-of-the-art (SOTA) segmentation accuracy and generalization performance on both private and public datasets, while maintaining highly competitive processing speed. Code is available at https://github.com/SikangSHU/GeNSeg-Net.
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 4475-4484
Open AccessGeNSeg-Net: A General Segmentation Framework for Any Nucleus in Immunohistochemistry Images
Immunohistochemistry (IHC) plays a crucial role in understanding disease mechanisms, diagnosing pathology and guiding treatment decisions. The precise analysis heavily depends on accurate nucleus segmentation. However, segmentation is challenging due to significant inter- and intra-nucleus variability in morphology and distribution, stemming from inherent characteristics, imaging techniques, tissue differences and other factors. While current deep learning-based methods have shown promising results, their generalization performance is limited, inevitably requiring specific training data. To address the problem, we propose a novel Ge neral framework for Nucleus Seg mentation in IHC images (GeNSeg-Net). GeNSeg-Net effectively segments nuclei across diverse tissue types and imaging techniques with high variability using a small subset for training. It comprises an enhancement model and a segmentation model. Initially, all nuclei are enhanced to a uniform morphology with distinct features by the enhancement model through generation. The subsequent segmentation task is thereby simplified, leading to higher accuracy. We design a lightweight generator and discriminator to improve both enhancement quality and computational efficiency. Extensive experiments demonstrate the effectiveness of each component within GeNSeg-Net. Compared to existing methods, GeNSeg-Net achieves state-of-the-art (SOTA) segmentation accuracy and generalization performance on both private and public datasets, while maintaining highly competitive processing speed. Code is available at https://github.com/SikangSHU/GeNSeg-Net.
- Open Access
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 10382-10391
LoFormer: Local Frequency Transformer for Image Deblurring
Due to the computational complexity of self-attention (SA), prevalent techniques for image deblurring often resort to either adopting localized SA or employing coarse-grained global SA methods, both of which exhibit drawbacks such as compromising global modeling or lacking fine-grained correlation. In order to address this issue by effectively modeling long-range dependencies without sacrificing fine-grained details, we introduce a novel approach termed Local Frequency Transformer (LoFormer). Within each unit of LoFormer, we incorporate a Local Channel-wise SA in the frequency domain (Freq-LC) to simultaneously capture cross-covariance within low- and high-frequency local windows. These operations offer the advantage of (1) ensuring equitable learning opportunities for both coarse-grained structures and fine-grained details, and (2) exploring a broader range of representational properties compared to coarse-grained global SA methods. Additionally, we introduce an MLP Gating mechanism complementary to Freq-LC, which serves to filter out irrelevant features while enhancing global learning capabilities. Our experiments demonstrate that LoFormer significantly improves performance in the image deblurring task, achieving a PSNR of 34.09 dB on the GoPro dataset with 126G FLOPs. https://github.com/DeepMed-Lab-ECNU/Single-Image-Deblur
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 10382-10391
Open AccessLoFormer: Local Frequency Transformer for Image Deblurring
Due to the computational complexity of self-attention (SA), prevalent techniques for image deblurring often resort to either adopting localized SA or employing coarse-grained global SA methods, both of which exhibit drawbacks such as compromising global modeling or lacking fine-grained correlation. In order to address this issue by effectively modeling long-range dependencies without sacrificing fine-grained details, we introduce a novel approach termed Local Frequency Transformer (LoFormer). Within each unit of LoFormer, we incorporate a Local Channel-wise SA in the frequency domain (Freq-LC) to simultaneously capture cross-covariance within low- and high-frequency local windows. These operations offer the advantage of (1) ensuring equitable learning opportunities for both coarse-grained structures and fine-grained details, and (2) exploring a broader range of representational properties compared to coarse-grained global SA methods. Additionally, we introduce an MLP Gating mechanism complementary to Freq-LC, which serves to filter out irrelevant features while enhancing global learning capabilities. Our experiments demonstrate that LoFormer significantly improves performance in the image deblurring task, achieving a PSNR of 34.09 dB on the GoPro dataset with 126G FLOPs. https://github.com/DeepMed-Lab-ECNU/Single-Image-Deblur