◷
Glacial Lake Outburst Flood Risk Assessment in High Mountain Asia
A logistic-regression approach using remote-sensing parameters to assess GLOF risk, presented at IGARSS 2024 in Athens.
Publications
Peer-reviewed research.
Our Earth-observation work is published in leading IEEE venues. Here are selected papers, the full, continuously updated record lives on Google Scholar.
Featured papers
Selected peer-reviewed work.
IEEE · Peer-reviewed
PKLandSeg: Annotation-Guided Hybrid CNN–Vision Transformer for Landslide Segmentation
A hybrid CNN–Vision Transformer architecture, guided by annotation strategy, for accurate landslide segmentation from satellite imagery across Pakistan's northern mountains.
Read on IEEE Xplore ↗IEEE · Peer-reviewed
LADI: Landslide Activity Detection Index Based on InSAR Coherence and Sensitivity
A novel index for detecting landslide activity from satellite radar interferometry, combining InSAR coherence and sensitivity analysis for hazard monitoring.
Read on IEEE Xplore ↗IEEE · Peer-reviewed
Remote Sensing for Hazard Monitoring and Modeling in High Mountain Asia
An overview of remote-sensing methods for monitoring and modeling natural hazards across the High Mountain Asia region.
Read on IEEE Xplore ↗IEEE · Peer-reviewed
Oil Spill Segmentation in SAR Imagery with a Deep Encoder-Decoder Network
A deep encoder-decoder network for automatically segmenting marine oil spills from synthetic-aperture-radar imagery.
Read on IEEE Xplore ↗Journal · 2025
Glacial Lake Mapping using a Remote-Sensing Geo-Foundation Model
Published in the International Journal of Applied Earth Observation and Geoinformation, applying a remote-sensing Geo-Foundation Model to glacial-lake mapping.
Read on ScienceDirect ↗More research & talks
Presented at leading venues.
◷
Glacial Lake Mapping using a Remote-Sensing Geo-Foundation Model
Published in the International Journal of Applied Earth Observation and Geoinformation, developed with ETH Zurich and the Chinese Academy of Sciences.
◷
Oil-Spill Detection from SAR Imagery using Deep Learning
An invited tutorial delivered at the 9th Asia-Pacific Conference on Synthetic Aperture Radar (APSAR 2025), Matsue, Japan.