How siddheshpohare/landslide-risk-mapping-himachal-pradesh works
This repository supports a Himachal Pradesh landslide-risk research workflow: acquire pre/post-event satellite imagery and point-level features, prepare inventory and training data, build susceptibility products, validate and refine them, and present results. It also documents both Earth Engine and local-download routes. The graph preserves the separate ML and AHP analysis branches. Some connections are documented in the README or visible excerpts; unsampled modules are represented only where their role is supported by the tree or documentation.
Main entry points: download_landslide_images.py Local image download · run_ml_pipeline_v3.py ML training pipeline · gee_extract_features_ml.js Point feature extraction · gee_landslide_hp.js Image collection
Components 18
Data acquisition
- Image collection
gee_landslide_hp.js - Local image download
download_landslide_images.py - Point feature extraction
gee_extract_features_ml.js
Data preparation
- Inventory conversion
convert_landslide_inventory.py - Report conversion
convert_landslide_report.py - Earth Engine inputs
prepare_gee_inputs.py
Risk mapping
- ML training pipeline
run_ml_pipeline_v3.py - ML evaluation plots
generate_plots_v3.py - AHP susceptibility
ahp_susceptibility.py - Susceptibility mosaic
phase2_susceptibility_map.py
Validation and refinement
- Raster validation
phase3_validation.py - Mask-fix functions
phase3_fix_funcs.py - Mask refinement
phase3_mask_fixes.py - Mask-fix utility
gen_mask_fixes.py
Reporting
- Midterm presentation
create_midterm_pptx.py
Other components
- Researcher
- Google Earth Engine
- Sentinel-2 data
Connections 25
- Researcher → Image collection: runs export
README.md - Image collection → Google Earth Engine: requests imagery
gee_landslide_hp.js - Google Earth Engine → Sentinel-2 data: queries collection
- Image collection → Local image download: alternative route
README.md - Researcher → Local image download: runs download
README.md - Local image download → Google Earth Engine: downloads composites
download_landslide_images.py - Local image download → Sentinel-2 data: queries collection
download_landslide_images.py - Point feature extraction → Google Earth Engine: samples features
gee_extract_features_ml.js - Point feature extraction → Sentinel-2 data: samples imagery
gee_extract_features_ml.js - Researcher → Inventory conversion: runs conversion
- Researcher → Report conversion: runs conversion
- Researcher → Earth Engine inputs: prepares inputs
- Point feature extraction → ML training pipeline: exports features
- ML training pipeline → ML evaluation plots: evaluates models
- ML training pipeline → ML evaluation plots: plots comparisons
- Researcher → ML training pipeline: runs analysis
- Researcher → AHP susceptibility: runs weighting
- AHP susceptibility → Susceptibility mosaic: creates input rasters
- Researcher → Susceptibility mosaic: builds map
- Susceptibility mosaic → Raster validation: supplies susceptibility
- Researcher → Raster validation: runs assessment
- Raster validation → Mask refinement: writes metrics
phase3_mask_fixes.py - Mask refinement → Mask-fix functions: uses functions
- Researcher → Mask-fix utility: runs utility
- Researcher → Midterm presentation: generates slides