Year: 2025
Type: article
Abstract: Landslides in hilly areas can be triggered by natural factors like heavy rainfall and earthquakes, or by human activities such as unplanned construction. These events often result in significant loss of life and property. Machine learning (ML) and deep learning (DL) algorithms have been increasingly used for automatic landslide detection from satellite images. While there has been progress in semiautomatic detection, fully automatic systems with high accuracy are still limited. One of the biggest challenges is the lack of appropriate training datasets. This study reviews various ML and DL techniques for landslide classification and identifies research gaps. It proposes a novel prototype using a modified ResNet101 deep learning model, achieving an accuracy of 96.88% on an augmented Beijing satellite dataset. The findings offer valuable insights for future research in landslide detection and classification using satellite imagery
Authors: B Kamala Deepthi, Boya Sravani, Meedinti Pavithra, Yashaswini C, Dr. K. Pavan Kumar
Language: English
Subfield: Management, Monitoring, Policy and Law
Field: Environmental Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4410272475