Important: This repository contains only the material used to produce The Sand Atlas website. The actual particle data is located on a different server. On the website you can navigate to the page for the sand you are interested in and download the particle data from there.
If you would like to help us improve the Sand Atlas, you should:
- Clone this repository
- Install ruby if it is not already on your computer (https://www.ruby-lang.org/en/documentation/installation/).
- In the terminal, inside the package folder run
bundle install
. This will install the requried packages so that we can build the website. - To run a local server to be able to see the Sand Atlas, run
bundle exec jekyll serve
. Open the browser tohttp://localhost:4000/
to see the website.
The code used to process and access data is available via
pip install sand-atlas
You can access the code itself and documentation here.
When someone wants to give us data, they start off with either a direct email or by filling out the web form on the homepage of the Sand Atlas. From there:
- We send a link to them requesting the upload of raw and labelled tiffs. We also ask for as much as possible of the following information:
- Title
- Short description
- Photo of the sand
- Latitude and longitude of source location
- Source data URL (if already hosted somewhere)
- Publication URL (if already published somewhere)
- Source license (if source data already licensed). If not previously licensed, we recommend CC BY 4.0.
- Mineralogy
- Resolution (mm per voxel)
- Once received, we download the tiffs
- The following is performed using the
sand-atlas
pip package:- Use openvdb (inside blender) to convert labelled image to meshes
- Use blender to render each mesh to a video
- We then host the meshes and videos on our server at
https://data.scigem-eng.sydney.edu.au/
. - Put all of this information into the sand atlas (metadata is stored in a json file at
_data/sands/
) and push to the repository to update the website.