Lehao Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Lehao Huang has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Bo-Hui Tang
Yunnan Key Laboratory of Quantitative Remote Sensing/Yunnan International Joint Laboratory for Integrated Sky-Ground Intelligent Monitoring of Mountain Hazards, Faculty of Land Resources Engineering, Kunming University of Science and Technology, Kunming 650093, China
5 shared publications - Zhen Zhang
Kunming University of Science and Technology
5 shared publications - Qingwang Wang
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China
4 shared publications - Chenxi Liu
Faculty of Land Resource Engineering, Kunming University of Science and Technology, Kunming, China
2 shared publications - Linhuan Jiang
Faculty of Land Resource Engineering, Kunming University of Science and Technology, Kunming, China
2 shared publications - Tao Shen
Kunming University of Science and Technology
2 shared publications
Latest Publications
- Top Cited PapersOpen Access
- 20 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19 (19391404) , 6840-6874
Artificial Intelligence for Remote Sensing: Progress, Challenges, and Perspectives
Recent breakthroughs in artificial intelligence (AI), particularly in deep learning, have transformed the processing and analysis of large-scale remote sensing (RS) data. These advancements deliver remarkable performance in tasks such as imagery interpretation and quantitative inversion, significantly boosting the efficiency and reach of RS interpretation. AI extends the utility of RS data across diverse fields, including disaster monitoring, environmental protection, and urban planning, cementing its role as a cornerstone technology in RS. This review systematically explores the paradigm transfer in RS driven by AI, proposing a dual-dimensional “data-task” evolution framework that traces the transition from single-modal, single-task models to multi-modal, multi-task architectures in imagery interpretation and quantitative RS research. Specifically, architectural innovations overcome the limitations of single-sensor systems by fusing optical, SAR, and textual data, enabling feature reuse and enhancing cross-task synergy through collaborative mechanisms. The emergence of RS agents introduces a closed-loop intelligence paradigm—integrating multi-modal perception, automated task planning, and interactive decision feedback—transforming RS interpretation from passive perception to active cognition. Despite these advances, significant challenges persist, such as data heterogeneity, edge computing limitations, and model security vulnerabilities, all of which are thoroughly assessed in this review. Looking ahead, we advocate next-generation deep learning architectures that integrate expert knowledge and human feedback to address the evolving demands of Earth observation. This review charts a clear path toward more reliable, adaptable, and intelligent RS systems, heralding a new era of AI-powered Earth observation.
- 20 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19 (19391404) , 6840-6874
Top Cited PapersOpen AccessArtificial Intelligence for Remote Sensing: Progress, Challenges, and Perspectives
Recent breakthroughs in artificial intelligence (AI), particularly in deep learning, have transformed the processing and analysis of large-scale remote sensing (RS) data. These advancements deliver remarkable performance in tasks such as imagery interpretation and quantitative inversion, significantly boosting the efficiency and reach of RS interpretation. AI extends the utility of RS data across diverse fields, including disaster monitoring, environmental protection, and urban planning, cementing its role as a cornerstone technology in RS. This review systematically explores the paradigm transfer in RS driven by AI, proposing a dual-dimensional “data-task” evolution framework that traces the transition from single-modal, single-task models to multi-modal, multi-task architectures in imagery interpretation and quantitative RS research. Specifically, architectural innovations overcome the limitations of single-sensor systems by fusing optical, SAR, and textual data, enabling feature reuse and enhancing cross-task synergy through collaborative mechanisms. The emergence of RS agents introduces a closed-loop intelligence paradigm—integrating multi-modal perception, automated task planning, and interactive decision feedback—transforming RS interpretation from passive perception to active cognition. Despite these advances, significant challenges persist, such as data heterogeneity, edge computing limitations, and model security vulnerabilities, all of which are thoroughly assessed in this review. Looking ahead, we advocate next-generation deep learning architectures that integrate expert knowledge and human feedback to address the evolving demands of Earth observation. This review charts a clear path toward more reliable, adaptable, and intelligent RS systems, heralding a new era of AI-powered Earth observation.
- 20 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-16
Frequency-Enhanced Spatial–Spectral Network for Hyperspectral Imagery Reconstruction From Multispectral Imagery
Hyperspectral imagery (HSI) delivers detailed spectral information critical for remote sensing applications such as high-accuracy land cover classification, quantitative parameter retrieval, and environmental monitoring. However, satellite-borne HSI often suffers from limited spatial resolution owing to inherent sensor constraints, whereas airborne HSI is constrained by restricted spatial coverage. Reconstructing high-resolution HSI from multispectral imagery emerges as a promising strategy to address these challenges. In this study, we propose the Frequency-Enhanced Spatial-Spectral Network (FESSN), a computationally efficient architecture that innovatively integrates multi-domain fusion across spatial, spectral, and frequency domains to achieve superior reconstruction performance. A key innovation is the neural network-driven frequency enhanced modulation (FEM), which adaptively refines spectral amplitudes and phases via fast Fourier transform, providing interpretable, parameter-efficient enhancements to bridge spatial-spectral modeling gaps. Meanwhile, a Mamba-based Multi-Scale Spatial Fusion module (MMSAF) that seamlessly integrates local features with long-range dependencies, and a U-shaped Spectral Module (USEM) that integrates Mamba and attention mechanisms to model inter-group and intra-group relationships, while adhering to spectral sparsity priors. The experimental results demonstrate FESSN outperforming six state-of-the-art methods in metrics like RMSE (up to 5.31% improvement), PSNR, SAM, ERGAS, and SSIM. Downstream tasks in land cover classification further validate its utility, positioning FESSN as a breakthrough in efficient, high-fidelity HSI reconstruction. To facilitate reproducibility and further research, the code will be publicly available at https://github.com/KustAIRS/TGRS-FESSN.
- 20 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-16
Frequency-Enhanced Spatial–Spectral Network for Hyperspectral Imagery Reconstruction From Multispectral Imagery
Hyperspectral imagery (HSI) delivers detailed spectral information critical for remote sensing applications such as high-accuracy land cover classification, quantitative parameter retrieval, and environmental monitoring. However, satellite-borne HSI often suffers from limited spatial resolution owing to inherent sensor constraints, whereas airborne HSI is constrained by restricted spatial coverage. Reconstructing high-resolution HSI from multispectral imagery emerges as a promising strategy to address these challenges. In this study, we propose the Frequency-Enhanced Spatial-Spectral Network (FESSN), a computationally efficient architecture that innovatively integrates multi-domain fusion across spatial, spectral, and frequency domains to achieve superior reconstruction performance. A key innovation is the neural network-driven frequency enhanced modulation (FEM), which adaptively refines spectral amplitudes and phases via fast Fourier transform, providing interpretable, parameter-efficient enhancements to bridge spatial-spectral modeling gaps. Meanwhile, a Mamba-based Multi-Scale Spatial Fusion module (MMSAF) that seamlessly integrates local features with long-range dependencies, and a U-shaped Spectral Module (USEM) that integrates Mamba and attention mechanisms to model inter-group and intra-group relationships, while adhering to spectral sparsity priors. The experimental results demonstrate FESSN outperforming six state-of-the-art methods in metrics like RMSE (up to 5.31% improvement), PSNR, SAM, ERGAS, and SSIM. Downstream tasks in land cover classification further validate its utility, positioning FESSN as a breakthrough in efficient, high-fidelity HSI reconstruction. To facilitate reproducibility and further research, the code will be publicly available at https://github.com/KustAIRS/TGRS-FESSN.
- Open Access
- 5 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 24617-24631
MHS-Mamba: A Multihierarchical Semantic Model for UAV Hyperspectral Image Classification
Hyperspectral image (HSI) captured by unmanned aerial vehicles (UAVs) are distinguished by superior spatial resolution and intricate spectral detail, with widespread applications in precise environmental monitoring, agricultural management, and biotic stress detection. The classification of UAV-acquired HSIs constitutes the cornerstone of these sophisticated monitoring endeavors. However, the ultra-high spatial resolution inherent to these images engenders considerable spatial heterogeneity and spectral variability. Existing deep learning classification methods overlook multi-level semantic features and global contextual information of images, leading to significant issues of salt-and-pepper noise and recurrent misclassifications. To address these challenges, this paper proposes a novel classification framework, Multi-Hierarchical Semantics Mamba (MHS-Mamba), tailored specifically for UAV-based hyperspectral image classification. The framework integrates two innovative components: the Multi-Hierarchical Semantics Spectral-Spatial Network (MHS-SSNet), which adeptly extracts spatial, spectral, and semantic features from UAV-derived HSIs, and the Linear Spectral Mamba (LS Mamba) module, which effectively models and amalgamates short- and long-range spectral dependencies. Experimental results on agricultural and coastal urban scene datasets demonstrate that the proposed method achieves overall accuracies of 96.58%, 96.98%, and 95.47%, outperforming state-of-the-art approaches by 1.99%, 3.22%, and 3.82%, respectively. These results highlight the robustness and generalizability of the proposed method, which holds strong potential for practical applications in precision agriculture and fine-scale land-cover monitoring. The source code is released at https://github.com/KustAIRS/JSTARS-MHS-Mamba.
- 5 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 24617-24631
Open AccessMHS-Mamba: A Multihierarchical Semantic Model for UAV Hyperspectral Image Classification
Hyperspectral image (HSI) captured by unmanned aerial vehicles (UAVs) are distinguished by superior spatial resolution and intricate spectral detail, with widespread applications in precise environmental monitoring, agricultural management, and biotic stress detection. The classification of UAV-acquired HSIs constitutes the cornerstone of these sophisticated monitoring endeavors. However, the ultra-high spatial resolution inherent to these images engenders considerable spatial heterogeneity and spectral variability. Existing deep learning classification methods overlook multi-level semantic features and global contextual information of images, leading to significant issues of salt-and-pepper noise and recurrent misclassifications. To address these challenges, this paper proposes a novel classification framework, Multi-Hierarchical Semantics Mamba (MHS-Mamba), tailored specifically for UAV-based hyperspectral image classification. The framework integrates two innovative components: the Multi-Hierarchical Semantics Spectral-Spatial Network (MHS-SSNet), which adeptly extracts spatial, spectral, and semantic features from UAV-derived HSIs, and the Linear Spectral Mamba (LS Mamba) module, which effectively models and amalgamates short- and long-range spectral dependencies. Experimental results on agricultural and coastal urban scene datasets demonstrate that the proposed method achieves overall accuracies of 96.58%, 96.98%, and 95.47%, outperforming state-of-the-art approaches by 1.99%, 3.22%, and 3.82%, respectively. These results highlight the robustness and generalizability of the proposed method, which holds strong potential for practical applications in precision agriculture and fine-scale land-cover monitoring. The source code is released at https://github.com/KustAIRS/JSTARS-MHS-Mamba.
- Top Cited PapersOpen Access
- 25 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 3099-3124
UAV Hyperspectral Remote Sensing Image Classification: A Systematic Review
In recent years, significant advances in Unmanned Aerial Vehicle (UAV) technology and hyperspectral remote sensing have spurred rapid and innovative developments in UAV-based HyperSpectral Image (HSI) classification across a range of fields, including environmental monitoring, precision agriculture, forest health assessment, and disaster management. Compared to spaceborne platforms, the spectra of ground objects observed by UAV platforms exhibit notable variations, presenting more pronounced challenges for accurate classification. This paper provides an in-depth and systematic review of UAV HSI classification techniques, systematically examining the evolution from traditional machine learning approaches, such as sparse coding, compressed sensing, and kernel methods, to cutting-edge deep learning frameworks, including Convolutional Neural Networks (CNNs), Transformer models, Recurrent Neural Networks (RNNs), Graph Convolutional Networks (GCNs), Generative Adversarial Networks (GANs), and hybrid models. Although traditional methods demonstrate effectiveness in certain scenarios, their limitations become increasingly apparent when dealing with high-dimensional, nonlinear spectral data. In contrast, deep learning-based models excel at capturing intricate relationships between spectral and spatial features, significantly boosting classification accuracy and emerging as the dominant paradigm in the field. And the WHU-Hi hyperspectral remote sensing dataset are utilized as a case study to elucidate the advantages and limitations of various deep learning methods through rigorous qualitative and quantitative comparisons. The potential of UAV hyperspectral image classification techniques in addressing high-dimensional data and complex scenarios is also thoroughly described. Furthermore, this paper delves into cutting-edge research trends, such as lightweight model development, hyperspectral large models, multi-source data fusion, and model interpretability, while also highlighting future trends for UAV hyperspectral remote sensing classification technology, particularly in real-time monitoring and intelligent applications.
- 25 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 3099-3124
Top Cited PapersOpen AccessUAV Hyperspectral Remote Sensing Image Classification: A Systematic Review
In recent years, significant advances in Unmanned Aerial Vehicle (UAV) technology and hyperspectral remote sensing have spurred rapid and innovative developments in UAV-based HyperSpectral Image (HSI) classification across a range of fields, including environmental monitoring, precision agriculture, forest health assessment, and disaster management. Compared to spaceborne platforms, the spectra of ground objects observed by UAV platforms exhibit notable variations, presenting more pronounced challenges for accurate classification. This paper provides an in-depth and systematic review of UAV HSI classification techniques, systematically examining the evolution from traditional machine learning approaches, such as sparse coding, compressed sensing, and kernel methods, to cutting-edge deep learning frameworks, including Convolutional Neural Networks (CNNs), Transformer models, Recurrent Neural Networks (RNNs), Graph Convolutional Networks (GCNs), Generative Adversarial Networks (GANs), and hybrid models. Although traditional methods demonstrate effectiveness in certain scenarios, their limitations become increasingly apparent when dealing with high-dimensional, nonlinear spectral data. In contrast, deep learning-based models excel at capturing intricate relationships between spectral and spatial features, significantly boosting classification accuracy and emerging as the dominant paradigm in the field. And the WHU-Hi hyperspectral remote sensing dataset are utilized as a case study to elucidate the advantages and limitations of various deep learning methods through rigorous qualitative and quantitative comparisons. The potential of UAV hyperspectral image classification techniques in addressing high-dimensional data and complex scenarios is also thoroughly described. Furthermore, this paper delves into cutting-edge research trends, such as lightweight model development, hyperspectral large models, multi-source data fusion, and model interpretability, while also highlighting future trends for UAV hyperspectral remote sensing classification technology, particularly in real-time monitoring and intelligent applications.
- Open Access
- 27 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 17 (19391404) , 6529-6543
PCCN-MSS: Parallel Convolutional Classification Network Combined Multi-Spatial Scale and Spectral Features for UAV-Borne Hyperspectral With High Spatial Resolution Imagery
Hyperspectral remote sensing images with high spatial resolution (H 2 imagery) have an abundant spatial-spectral information, holding tremendous potential for remote sensing fine-grained monitoring and classification. However, challenges such as high spatial heterogeneity, severe intra-class spectral variability and poor signal-to-noise ratio especially in unmanned aerial vehicle (UAV) hyperspectral imagery constrain and hinder the performance of fine-grained classification. Convolutional neural network (CNN) emerges as a formidable and excellent tool for image mining and feature extraction, offering effective utility for land cover classification. In this paper, a parallel convolutional classification network model based on multi-modal filters (including ICA-2D-FPN and SA-3D-CNN branching structures) PCCN-MSS is proposed for precise H 2 imagery classification. The ICA-2D-FPN branch integrates independent component analysis (ICA) into 2D-CNN to extract the multi-spatial scale and spectral information of H 2 imagery by Feature Pyramid Networks (FPN), meanwhile the SA-3D-CNN branch is designed to extract the spatial and spectral information by combining spectral attention (SA) mechanism and 3D-CNN. Taking hyperspectral imagery of unmanned aerial vehicles containing vegetation and artifactual material ground as an example, the proposed PCCN-MSS model achieves an overall accuracy (OA) of 78.18%, which outperforms by 9.58% than the compared methods. The proposed PCCN-MSS method can mitigate the classification issues of severe salt-and-pepper noise and inaccurate boundary, delivering more satisfactory classification results with robust classification performance and remarkable advantages for H 2 imagery.
- 27 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 17 (19391404) , 6529-6543
Open AccessPCCN-MSS: Parallel Convolutional Classification Network Combined Multi-Spatial Scale and Spectral Features for UAV-Borne Hyperspectral With High Spatial Resolution Imagery
Hyperspectral remote sensing images with high spatial resolution (H 2 imagery) have an abundant spatial-spectral information, holding tremendous potential for remote sensing fine-grained monitoring and classification. However, challenges such as high spatial heterogeneity, severe intra-class spectral variability and poor signal-to-noise ratio especially in unmanned aerial vehicle (UAV) hyperspectral imagery constrain and hinder the performance of fine-grained classification. Convolutional neural network (CNN) emerges as a formidable and excellent tool for image mining and feature extraction, offering effective utility for land cover classification. In this paper, a parallel convolutional classification network model based on multi-modal filters (including ICA-2D-FPN and SA-3D-CNN branching structures) PCCN-MSS is proposed for precise H 2 imagery classification. The ICA-2D-FPN branch integrates independent component analysis (ICA) into 2D-CNN to extract the multi-spatial scale and spectral information of H 2 imagery by Feature Pyramid Networks (FPN), meanwhile the SA-3D-CNN branch is designed to extract the spatial and spectral information by combining spectral attention (SA) mechanism and 3D-CNN. Taking hyperspectral imagery of unmanned aerial vehicles containing vegetation and artifactual material ground as an example, the proposed PCCN-MSS model achieves an overall accuracy (OA) of 78.18%, which outperforms by 9.58% than the compared methods. The proposed PCCN-MSS method can mitigate the classification issues of severe salt-and-pepper noise and inaccurate boundary, delivering more satisfactory classification results with robust classification performance and remarkable advantages for H 2 imagery.