Convolutional transformer attention network with few-shot learning for grassland degradation monitoring using UAV hyperspectral imagery

ABSTRACT In recent years, the desertification of grasslands has increased due to various factors, including both global warming and human activities. It is an essential basis for grassland degradation monitoring to monitor the dynamic change of desert grassland vegetation communities and distribution statistics. Although unmanned aerial vehicle (UAV) remote sensing images have allowed us to achieve dynamic real-time grassland monitoring, the distribution of desert grassland ground objects can be random and narrow, thus increasing the difficulty of sample labelling of remote sensing imagery. Therefore, to reduce the number of samples required for the model, this research proposes a convolutional transformer attention network (CTAN) to identify desert grassland ground objects and validate it on a self-collected desert grassland dataset. The network utilizes the transformer model to enhance its learning of global pixels so that it suppresses the transmission of background pixels within the network. Furthermore, the edge convolution module is designed to strengthen the network’s learning for edge pixels, improving its identification effect. The results show that the network provides 97.22% of overall accuracy (OA), 94.35% of average accuracy (AA), and 0.9398 of Kappa for ground object recognition in desert grassland. The model improves OA by 2.36–9.85% points compared to methods in the same field and 0.8–6.35% points compared to methods in hyperspectral imagery classification. The experimental results show the superior performance of the CTAN model for recognizing desert grassland objects, which helps the management and restoration of desert grasslands.

• Publication year

  2024

• Venue

  International Journal of Remote Sensing

• Publication date

  2024-03-11

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1080/01431161.2024.2326042
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• A Hyperspectral Image Classification Method Based on the Nonlocal Attention Mechanism of a Multiscale Convolutional Neural Network

  2023cited by this paper
• Identification and Classification of Small Sample Desert Grassland Vegetation Communities Based on Dynamic Graph Convolution and UAV Hyperspectral Imagery

  2023cited by this paper
• Evaluation of Stem Rust Disease in Wheat Fields by Drone Hyperspectral Imaging

  2023cited by this paper
• Research on Grassland Rodent Infestation Monitoring Methods Based on Dense Residual Networks and Unmanned Aerial Vehicle Remote Sensing

  2023cited by this paper
• Lightweight Spatial-Spectral Network Based on 3D-2D Multi-Group Feature Extraction Module for Hyperspectral Image Classification

  2023influential reference
• Seasonal Variations and Influencing Factors of Gross Nitrification Rate in Desert Steppe Soil

  2022cited by this paper
• Causes of overgrazing in Inner Mongolian grasslands: Searching for deep leverage points of intervention

  2022influential reference
• Graph-Feature-Enhanced Selective Assignment Network for Hyperspectral and Multispectral Data Classification

  2022cited by this paper
• Classification of desert grassland species based on a local-global feature enhancement network and UAV hyperspectral remote sensing

  2022influential reference
• Triple-branch ternary-attention mechanism network with deformable 3D convolution for hyperspectral image classification

  2022cited by this paper
• Research on deep learning method recognition and a classification model of grassland grass species based on unmanned aerial vehicle hyperspectral remote sensing

  2022cited by this paper
• Spectral-spatial hyperspectral image classification based on capsule network with limited training samples

  2022cited by this paper
• 3D-CNN based UAV hyperspectral imagery for grassland degradation indicator ground object classification research

  2021influential reference
• Residual Spectral-Spatial Attention Network for Hyperspectral Image Classification

  2021influential reference
• Aboveground mangrove biomass estimation in Beibu Gulf using machine learning and UAV remote sensing

  2021cited by this paper
• Assessing human-environment system sustainability based on Regional Safe and Just Operating Space: The case of the Inner Mongolia Grassland

  2021cited by this paper
• Scale effects on spatial heterogeneity of herbaceous vegetation in desert steppe depend on plant community type

  2021cited by this paper
• Mapping the occurrence and spatial distribution of noxious weed species with multisource data in degraded grasslands in the Three-River Headwaters Region, China.

  2021cited by this paper
• Hyperspectral image classification with multi-scale graph convolution network

  2021cited by this paper
• The response of soil nematode fauna to climate drying and warming in Stipa breviflora desert steppe in Inner Mongolia, China

  2020cited by this paper
• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

  2020influential reference
• Response of plant traits of Stipa breviflora to grazing intensity and fluctuation in annual precipitation in a desert steppe, northern China

  2020cited by this paper
• Desertification Glassland Classification and Three-Dimensional Convolution Neural Network Model for Identifying Desert Grassland Landforms with Unmanned Aerial Vehicle Hyperspectral Remote Sensing Images

  2020influential reference
• Variation of vegetation fractional coverage and its relationship with climate in a desert steppe: optimization of farmland layout in a farming–pastoral ecotone using the ecological suitability index

  2020cited by this paper
• Impact of climate change on primary production of Inner Mongolian grasslands

  2020cited by this paper
• Estimation of the Yield and Plant Height of Winter Wheat Using UAV-Based Hyperspectral Images

  2020cited by this paper
• Resilience of revegetated grassland for restoring severely degraded alpine meadows is driven by plant and soil quality along recovery time: A case study from the Three-river Headwater Area of Qinghai-Tibetan Plateau

  2019cited by this paper
• Spectral-Spatial Attention Networks for Hyperspectral Image Classification

  2019cited by this paper
• Grazing induced changes in plant diversity is a critical factor controlling grassland productivity in the Desert Steppe, Northern China

  2018cited by this paper
• Land-cover change in the Wulagai grassland, Inner Mongolia of China between 1986 and 2014 analysed using multi-temporal Landsat images

  2018cited by this paper
• The diversity and co-occurrence patterns of diazotrophs in the steppes of Inner Mongolia

  2017cited by this paper
• Pixel-Level Spatiotemporal Analyses of Vegetation Fractional Coverage Variation and Its Influential Factors in a Desert Steppe: A Case Study in Inner Mongolia, China

  2017cited by this paper
• Effects of Government Grassland Conservation Policy on Household Livelihoods and Dependence on Local Grasslands: Evidence from Inner Mongolia, China

  2016cited by this paper
• Climate adaptation, local institutions, and rural livelihoods: A comparative study of herder communities in Mongolia and Inner Mongolia, China

  2013cited by this paper
• Spatial vegetation patterns as early signs of desertification: a case study of a desert steppe in Inner Mongolia, China

  2010cited by this paper

• Transformers meet hyperspectral imaging: A comprehensive study of models, challenges and open problems

  2026cites this paper
• Hyperspectral Image Synthesis Through Blind Unmixing Dictionary and Deep Diffusion Models

  2026cites this paper
• Large-Scale Stitching of Hyperspectral Remote Sensing Images Obtained from Spectral Scanning Spectrometers Mounted on Unmanned Aerial Vehicles

  2025cites this paper
• Semantic-Guided Transformer Network for Crop Classification in Hyperspectral Images

  2025cites this paper
• ResMamba: A state-space model approach and benchmark dataset for precise forage identification in desert rangelands

  2025cites this paper
• CenterMamba: Enhancing semantic representation with center-scan Mamba network for hyperspectral image classification

  2025cites this paper
• Deep Diffusion Models and Unsupervised Hyperspectral Unmixing for Realistic Abundance Map Synthesis

  2025cites this paper
• A Novel Remote Sensing Framework Integrating Geostatistical Methods and Machine Learning for Spatial Prediction of Diversity Indices in the Desert Steppe

  2025cites this paper
• Cross-domain adversarial learning for forage mapping and alpha-diversity assessment from UAV hyperspectral imagery in desert rangelands

  2025cites this paper
• SmartLite-YOLO: a smart classroom lightweight YOLO for student behavior recognition

  2025cites this paper
• Synthesis of Abundance Maps Through Blind Hyperspectral Unmixing and Deep Diffusion Models

  2025cites this paper
• Integrating UAV visible and multispectral imagery to assess grazing-induced vegetation responses in sandy grasslands

  2025cites this paper
• Integrated Remote Sensing Evaluation of Grassland Degradation Using Multi-Criteria GDCI in Ili Prefecture, Xinjiang, China

  2025cites this paper
• An efficient and precise dynamic neighbor graph network for crop mapping using unmanned aerial vehicle hyperspectral imagery

  2025cites this paper
• Mangrove Species Classification from Unmanned Aerial Vehicle Hyperspectral Images Using Object-Oriented Methods Based on Feature Combination and Optimization

  2024cites this paper