Datasets:
license: cc-by-4.0
task_categories:
- feature-extraction
pretty_name: IndLands
size_categories:
- 1M<n<10M
tags:
- geospatial
- remote-sensing
- spatial-analysis
- benchmarking
IndLands : A Spatiotemporal dataset for region-aware landslide analysis from Multi-Source Remote Sensing Imagery
This repository contains the complete workflow and supporting files for generating a landslide-prone area dataset using remote sensing and machine learning techniques. The dataset has been prepared for the following Indian states:
|
|
The final output is a multi-modal dataset containing terrain, spectral, and texture-based features extracted from satellite data, along with manually annotated landslide zones for training machine learning models.The complete workflow and steps can be accessible from here
Structure of the ML Ready Dataset
Each ZIP archive corresponds to a single state and contains multi-modal data derived from remote sensing sources, including:
- Raw Sentinel-2 image tiles
- Digital Elevation Model (DEM) data
- GLCM-based texture features
- Spectral index feature maps
- Spatial subsets
- Manually annotated landslide regions
dataset.csv, providing a consolidated tabular view of all extracted features, annotations, and associated latitude–longitude coordinates
Note: The full dataset, including intermediate preprocessing outputs generated at each stage of the pipeline, is available upon request.