This ReadMe.txt file was generated on 20230306 by Bickel, Valentin Tertius
This ReadMe.txt file was updated on 20230717 by Bickel, Valentin Tertius
-------------------
GENERAL INFORMATION
-------------------
1. Title of dataset: Global Mapping of Fragmented Rocks on the Moon with a Neural Network: Implications for the Failure Mode of Rocks on Airless Surfaces

2. Contributor information: 
Data creator
      Name: Bickel, Valentin Tertius
      Role/Function: Date creator
      Institution: Center for Space and Habitability, University of Bern
      Address: G6, Gesellschaftsstrasse 6, 3012 Bern, Switzerland
      Email: valentin.bickel@unibe.ch

Contributor
      Name: Ruesch, Ottaviano
      Role/Function: Contributor
      Institution: University of Muenster
      
3. Date of data collection: 
2022-03-01 to 2022-05-01

4. Geographic location of data collection:
Moon, 60°N to 60°S

5. Keywords describing the subject of your dataset:
Moon; boulder fragmentation; landscape evolution; machine learning; object detection

6. Information about funding sources that supported the collection of the data:
      Funding agency name: Center for Space and Habitability, University of Bern
      Funding agency name: ETH Zurich

-------------------
SHARING/ACCESS INFORMATION
-------------------
1. Licenses/restrictions placed on the data:
Attribution-NonCommercial-ShareAlike (CC BY-NC-SA 4.0)

2. Links to publications that cite or use the data: 
https://iopscience.iop.org/article/10.3847/PSJ/acd1ef

3. Recommended citation for this dataset: 
Ruesch, O., & Bickel, V.T. (2023). Global Mapping of Fragmented Rocks on the Moon with a Neural Network: Implications for the Failure Mode of Rocks on Airless Surfaces. The Planetary Science Journal 4 (126). DOI 10.3847/PSJ/acd1ef

---------------------
DATA & FILE OVERVIEW
---------------------
1. File List:
      Filename: Ruesch_and_Bickel_2023_fractured_CT04_cluster.csv
      File format: .csv
      Short description: File with meta information about all detections

      Filename: Ruesch_and_Bickel_2023_fractured_CT04_cluster_Part1.zip
      File format: .zip, .tif
      Short description: Zipped archive of .tif thumbnail images (first part)

      Filename: Ruesch_and_Bickel_2023_fractured_CT04_cluster_Part2.zip
      File format: .zip, .tif
      Short description: Zipped archive of .tif thumbnail images (second part)

--------------------------
METHODOLOGICAL INFORMATION
--------------------------
1. Description of methods used for collection/generation of data: 
Supervised convolutional neural network (RetinaNet with ResNet50 backbone), see associated publication for details

2. Methods for processing the data: 
Neural network deployed on Lunar Reconnaissance Orbiter Narrow Angle Camera RDR (calibrated "C") PTIFF images, see associated publication for details

3. Describe any quality-assurance procedures performed on the data:
Expert-based review of a substantial fraction of the dataset, see associated publication for details

-----------------------------------------
DATA-SPECIFIC INFORMATION FOR: Ruesch_and_Bickel_2023_fractured_CT04_cluster.csv
-----------------------------------------
1. Number of variables/columns: 19

2. Number of cases/rows: 187,788

3. Variable List:
nac_run_ID; nac_ID; center_LON; center_LAT; upper_left_LON; upper_left_LAT; boulder_diameter_meter; x_length_pixel; y_length_pixel; x_length_meter; y_length_meter; upper_left_x_img; upper_left_y_img; lower_right_x_img; lower_right_y_img; confidence; nac_resolution; cluster_ID; thumbnail_ID

4. Variable description:
internal detection ID; NAC image ID; detection box center latitude (°); detection box center longitude (°); detection box top left latitude (°); detection box top left longitude (°); estimated fractured boulder diameter (not calibrated, meter); x length of detection box (pixel); y length of detection box (pixel); x length of detection box (meter); y length of detection box (meter); detection box upper left x (pixel); detection box upper left y (pixel); detection box lower right x (pixel); detection box lower right y (pixel); CNN confidence (%); NAC image resolution (m/pixel); cluster ID; .tif thumbnail ID