A5100767171- Remote Sensing and LiDAR Applications
- Solar Radiation and Photovoltaics
- Remote-Sensing Image Classification
- Synthetic Aperture Radar (SAR) Applications and Techniques
- 3D Surveying and Cultural Heritage
- Groundwater and Watershed Analysis
- Landslides and related hazards
- Energy and Environment Impacts
- Soil Moisture and Remote Sensing
- Photovoltaic System Optimization Techniques
- Automated Road and Building Extraction
- Remote Sensing in Agriculture
- Domain Adaptation and Few-Shot Learning
- Advanced Neural Network Applications
- Advanced Optical Sensing Technologies
- Advanced Image and Video Retrieval Techniques
- Remote Sensing and Land Use
Nanjing University of Posts and Telecommunications
2015-2024
Nanjing University
2023
Remote sensing images play a critical role in urban planning, land resources and environmental monitoring. Land cover classification is one of the straightforward applications remote sensing. However, anomalous data challenges reliability results. Deep learning has been widely used image analysis, but it remains sensitive to data. To address this issue, we re-evaluate map high-noise scenarios with propose novel network architecture solve problem. A new proposed Our focuses on decoupling...
Over exploitation of groundwater in Changzhou city, China can cause land deformation, which turn proves detrimental to the urban infrastructure. In this study, multi-band synthetic aperture radar (SAR) data sets (C-band Envisat ASAR, L-band ALOS PALSAR, and X-band COSMO-SkyMed) acquired from 2006 2012 were analysed using interferometry (InSAR) time-series method investigate relationship between spatial–temporal distribution deformation exploitation. Annual rate inferred interferograms ranges...
The importance of renewable energy has been steadily increasing. Efficiently obtaining geographical distribution information through remote sensing is essential for managing and developing photovoltaic power, which a significant aspect energy. This study introduces novel framework detecting the power plants in large-scale areas. Initially, our employs frozen pre-trained Vision Transformer (ViT) model as encoder incorporates decoder to align textual feature representation with original visual...
We aim to improve the efficiency of traditional deep learning methods for remote sensing by reducing reliance on annotated data and minimizing training time. Instead using large-scale unimodal image datasets pre-training, we propose use multimodal (text-image pairs), which believe be more effective. To enhance model's generalization performance in domain achieve accurate scene classification, employ Feature Adaptive Embedding Module. For this purpose, introduce a cross-modal comparison...
With the development of social economy, wind turbines are taking up a larger and share new energy sources. The detection number spatial distribution in remotely sensed images holds great scientific significance. Wind difficult to identify remote sensing therefore, fast method based on deep learning is proposed. First, we extract potential turbine candidate regions from speed, slope, land use data. Second, YOLO v5 model was trained using our labeled dataset. Finally, were used for optimal...
Mobile Laser Scanning (MLS) systems provide fast and easy access to dense accurate point clouds of roadways. Currently, semantic segmentation urban roadway objects in these data plays a significant role modelling, autonomous driving, intelligent transportation, etc. However, the large-scale is still challenging due unstructured disordered nature cloud enormousness points acquired during practical applications. Based on that, we propose an automated target approach based RandLA-Net. With this...