Landslide Detection and Mapping using Remote Sensing U-Net Model In Leh-Ladakh Region

Landslides pose significant risks to communities and infrastructure, making detection, monitoring, and prediction crucial for effective risk management. Remote sensing techniques (RSTs) play a vital role in this regard by offering tools to observe, monitor, and predict landslides. Observation of Earth from space provides a broad perspective, allowing for the detection of changes in terrain morphology that may indicate landslide activity. Satellite imagery and data can be analyzed to identify areas prone to landslides and monitor changes over time. Laser scanning, also known as LiDAR (Light Detection and Ranging), offers high-resolution elevation data that can be used to create detailed terrain models. These models help in identifying slope instability and potential landslide risk areas. Ground-based interferometry involves the use of sensors deployed on the ground to measure ground deformation with high precision. This technique can detect subtle movements in the Earth’s surface, providing early warning signs of potential landslides. By integrating data from these remote sensing techniques, scientists and authorities can develop effective strategies for landslide risk assessment, early warning systems, and disaster mitigation measures. This holistic approach is essential for safeguarding lives and infrastructure in landslide-prone areas. This paper’s output provides an overview of landslip debris utilizing satellite imagery techniques and deep learning techniques based on the UNET model.