A5102735163- Remote Sensing and LiDAR Applications
- Remote-Sensing Image Classification
- Remote Sensing and Land Use
- Advanced Neural Network Applications
- Automated Road and Building Extraction
- Medical Imaging Techniques and Applications
- Advanced Image and Video Retrieval Techniques
- Advanced X-ray and CT Imaging
- Advanced MRI Techniques and Applications
- Medical Image Segmentation Techniques
- Robotics and Sensor-Based Localization
- Satellite Image Processing and Photogrammetry
- Video Surveillance and Tracking Methods
- Fault Detection and Control Systems
- Anomaly Detection Techniques and Applications
- Domain Adaptation and Few-Shot Learning
- Remote Sensing in Agriculture
- Industrial Vision Systems and Defect Detection
- Wildlife-Road Interactions and Conservation
- Adversarial Robustness in Machine Learning
- Geological Modeling and Analysis
- Forest ecology and management
- Image and Signal Denoising Methods
- Radiomics and Machine Learning in Medical Imaging
- Augmented Reality Applications
Wuhan University
2022-2025
Gauhati University
2018
Distributed acoustic sensor (DAS) technology leverages optical fiber cables to detect signals, providing cost-effective and dense monitoring capabilities. It offers several advantages including resistance extreme conditions, immunity electromagnetic interference, accurate detection. However, DAS typically exhibits a lower signal-to-noise ratio (S/N) compared geophones is susceptible various noise types, such as random noise, erratic level long-period noise. This reduced S/N can negatively...
Magnetotelluric (MT) sounding is a vital geophysical exploration method, renowned for its ability to investigate deep geological structures, detect low-resistivity anomalies, and support diverse applications. It widely used in mineral geothermal resource exploration, hydrogeological surveys, structural studies, effectively delineating subsurface structures from hundreds of meters kilometers. However, MT inversion, essential quantitatively analyzing electrical faces challenges due the...
Intelligent interpretation of remote sensing images using deep learning is heavily reliant on large datasets, and models trained in one domain often struggle with cross-domain application.Pre-training the backbone network via masked image modeling can effectively diminish this reliance extensive sample data, thereby reducing transfer obstacles.However, current typically employ a pure Transformer architecture, which may not fully capitalize lowlevel features.To address these issues, paper...
Road topology extraction from satellite images, which has long been of interest, is an essential task in remote sensing. The graph representation road networks one the most challenging aspects extraction. Most existing approaches cast as binary segmentation and then use postprocessing, such skeletonization, to infer pixelwise prediction. In our work, we believe that a network can be represented by undirected denoted <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML"...
Semantic segmentation of airborne laser scanning (ALS) point clouds is a valuable yet challenging task in remote sensing. When processing large-scale ALS scenes, it necessary to partition them into smaller blocks for ease handling. However, this partitioning introduces challenge capturing the ample spatial context within each block adequately recognize objects with significant span. This limitation becomes particularly pronounced when relying solely on 3D representations as input nerual...
Automatic extraction of vector polygons buildings from remotely sensed images is an important but difficult task. Recent existing methods based on deep learning usually adopt a multi-stage solution semantic segmentation, contour detection, and polygon simplification. Such long processing chain may lead to unreliable results as the boundary regularization optimization processes are ultimately completed by utilizing low-level features, which ignores potential features in generation. In this...
Building subclass segmentation, aimed at predicting classes of buildings (high-rise zone, low-rise single high-rise, and low-rise) from satellite images, is beneficial in numerous applications, including human geography, urban planning, humanitarian aid. However, problems such as complex scenes similar characteristics different building categories make it difficult for general models to balance the accuracy localization classification segmentation. Therefore, this paper proposes a novel...
Rapid generation of tree models from point clouds canopies holds wide-ranging applications in the field earth sciences, including forest ecology research, environmental monitoring, and management. Traditional modeling methods rely on procedural to simulate growth, which are time-consuming due their extensive manual parameterization. Furthermore, existing deep learning struggle generate visually realistic because complex branch structures specific natural patterns trees. To address these...
In the era of increasingly advanced Earth Observation (EO) technologies, extracting pertinent information (such as water-bodies) from Earth's surface has become a crucial task. Deep Learning, especially via pre-trained models, currently offers highly promising approach for semantic segmentation Remote Sensing Imagery (RSI). However, effectively adapting these models to RSI tasks remains challenging. Typically, undergo fine-tuning specialized tasks, involving modifications their parameters or...