This repository contains code developed for super-resolution by a deep spatial interpolation model of the mapping of lunar surface elemental abundances obtained from Chandrayaan-2 CLASS instrument XRF data. The detailed description of the project is available in this Report. The folder data_processing_and_mapping contains the codes for the extraction and processing of XRF data from the FITS file using XSPEC, XSMDAS, and GDL, and Information related to the visualization of the abundances on the Lunar albedo map using QGIS.

1. data_processing_and_mapping:

   • gdl: GNU Data Language (GDL) for processing and analyzing FITS (Flexible Image Transport System) files containing data on lunar surface elemental abundances.
   • xspec_xsmdas: XSPEC is a software designed for the analysis of astronomical X-ray spectra. It enables users to fit theoretical models to observational data, providing insights into the physical properties of celestial objects. XSM Data Analysis Software (XSMDAS) processes XSM data step by step to create usable results. It converts raw data (level-0) into intermediate data (level-1) and then into final products (level-2). These final products include calibrated spectra and light curves, which can be customized with different settings.
   • mapping: Quantum Geographic Information System (QGIS) software for Visualizing the Lunar Abundances on Lunar Albedo Basemap at good sub-pixel resolution. The task is automated using PyQGIS codes. QGIS an open-source GIS platform that is freely available and offers a range of features suitable for various types of geospatial data analysis and visualization.

2. file_initialisation.py:

   • Initializes the 64 subregion CSV files for the Moon's surface.
   • For each subregion, compute the latitudes and longitudes of 2 km x 2 km pixels.
   • Extracts features from image tiles using a pre-trained ResNet50 model.
   • Populates the CSV files with selected features, mare/highland classifications, and metadata.

3. top_k_features.ipynb:

   • Analyzes feature importance from ResNet50 extractions.
   • Identifies the top 300 features for further processing based on variance thresholding.

4. Final.py:

   • Combines all parts of the project.
   • Constructs subgraphs for abundance maps.
   • Train the GNN-CNN combined model on these subgraphs.
   • Returns the final high-resolution abundance maps.

• Python 3.10+
• GPU with CUDA support (optional but recommended).
• Clone the repository:

  git clone https://github.com/who-else-but-arjun/ISRO_XRF_SR.git
  cd ISRO_XRF_SR

• Install dependencies:

  pip install -r requirements.txt

• Shapefile for Mare Regions: LROC_GLOBAL_MARE_180.SHP (https://drive.google.com/file/d/12J4m65JyD-cekRiMmBzjrCBWQKoxNWYq/view?usp=drive_link).
• Lunar Surface Image: Lunar_LRO_LROC-WAC_Mosaic_global_100m_June2013.tif (https://planetarymaps.usgs.gov/mosaic/Lunar_LRO_LROC-WAC_Mosaic_global_100m_June2013.tif).
• CSV Files: Intermediate outputs from file_initialisation.py.

Place the downloaded files in their respective directories as outlined below:

• LROC_GLOBAL_MARE_180.SHP: ./Mare Classification/
• Lunar_LRO_LROC-WAC_Mosaic_global_100m_June2013.tif: Root directory.

.
├── data_processing_and_mapping/
        | - gdl/
        | - xspec_xsmdas/
        | - mapping_codes/
        | - pre_processing/
        | - Final.csv                                        # Final extracted abundances. 
        | - README.md                                        # Instructions for data extraction.
├── file_initialisation.py                                   # Initialization and feature extraction.
├── top_k_features.ipynb                                     # Feature analysis and selection.
├── Final.py
├── Final_notebook.ipynb                                     # Full pipeline integration.
├── Lunar_LRO_LROC-WAC_Mosaic_global_100m_June2013.tif                
├── requirements.txt                                         # Dependency file.
├── Mare Classification/                                     # Directory for shapefiles.
        | - LROC_GLOBAL_MARE_180.DBF
        | - LROC_GLOBAL_MARE_180.PRJ
        | - LROC_GLOBAL_MARE_180.SHP
        | - LROC_GLOBAL_MARE_180.SHP.XML
        | - LROC_GLOBAL_MARE_180.SHX
        | - LROC_GLOBAL_MARE_README.TXT  
├── input.csv
├── output.csv
├── PART2Output.npz                                        # Stores updated subregions
├── graphs/
├── masks/
├── models/ 
└── regions/                                                  # Outputs of file_initialisation.py.

Final.csv could be used as the input to further steps involving super-resolution.

• regions CSVs:

  • Files named subregion_<i>_<j>.csv.
  • Columns include:
    • lat_center and lon_center: Geographic coordinates of the pixel centers.
    • Elemental abundances: Features like Fe, Ti, Ca, etc.
    • mareOrHighland: Binary classification indicating whether the region is Mare or Highland.
    • Top 300 spatial features: Feature columns extracted from lunar rgb images and top 300 selected features based on variance thresholding.

• Graphs and their corresponding masks:

  • Structured inputs for graph neural network models.
  • Encodes spatial and feature-based relationships within subregions.
  • Graphs stored in graphs\graphs_subregion_<i>_<j>\subgraph_{i}_{j}_{row_idx}_{col_idx}.pt.
  • Masks for the graphs stored in masks\masks_subregion_<i>_<j>\mask_{i}_{j}_{row_idx}_{col_idx}.pt.

• Final High-Resolution Maps:

  • Enhanced elemental abundance data for each subregion.
  • Saved as updated subregion_<i>_<j>.csv files for mapping.

This script initializes and populates the subregion CSV files.

python PART1.py

The final.py script combines all components and generates the high-resolution abundance maps.

python Final.py --mode 1 <input_data.csv>

mode = 1 for population of initialised csv files with the input elemental abundances data. Assuming input.csv contains the data to be added. Headers required in input.csv (similar to the Final.csv) :

This will populate all the regions csv and create a file by the name of PART2Output.npz in the current directory, which is required for Abundance Generalisation to run. It contains information regarding the number of entries added in each region and subregion, along with the updated indices of each region file. This is required for mask creation during Abundance Generalisation.

To train on subregion i, j using parameters mode = 2, todo = run, and num_iterations :

python Final.py --mode 2 <i> <j> <num_iterations>

mode = 2 for creating the graphs for each subregion and training, and interpolation for final high-resolution abundance mapping. The pretrained models are automatically saved and loaded from models/

The first execution for any region will take longer and also create masks. Do not delete these masks as they are used to store number of old runs which is required for future runs.

Ensure the following files are downloaded on your system:

1. https://planetarymaps.usgs.gov/mosaic/Lunar_LRO_LROC-WAC_Mosaic_global_100m_June2013.tif
2. https://drive.google.com/file/d/1Ig5nXZqwscWYRLWgD22a76N5AsIDF2cv/view
3. https://drive.google.com/file/d/1RgGrRBntdAb6aMf7mvAgD23WunY2uqv8/view
4. https://drive.google.com/file/d/1LqYxaY08nyZqGlK0Udd28MzekwzwLsCz/view
5. https://drive.google.com/file/d/11794ulttX9D46Onumwvg6ToBJCGd9M05L/view
6. https://drive.google.com/file/d/15DUqZp716VV60jBM8V0CCg2oQdlBXlDU/view
7. https://drive.google.com/file/d/1Ltm2PjQvXCQ-NiJjARWrJGsr74AxWv3g/view
8. https://drive.google.com/file/d/1agqOartoVDRJ65yG_gxmtF6Z6oyk7Xd4/view
9. https://drive.google.com/file/d/1ul-A8zHvUUOl62F0fzDqm8wL49GNuDiL/view

All these files must be located in the root directory of the folder