The program aims to determine subject space and projection method of a map image or graphic. The project was first a part of 2016 Wolfram Summer School, then Wolfram Mentorships through summer of 2017, under supervision of Dr. Giorgia Fortuna. An overview at the end of the Mentorships program is here.
Wolfram Mathematica 10.1 or later. Program was last compiled on 11.2. The program heavily relies on Wolfram's TextRecognize function, whose dependability varies by version.
A que7 network can be trained in the code, with several tweakable parameters, including network structure, loss function, training rounds, etc. After que7 is trained, the final function can be tested on any given image, with the image as the argument.
- Wolfram Mathematica - The platform used
- Junghyun Min - Initial work - Aatlantise
See also the list of contributors who participated in this project.
This project is licensed under the GNU GPL v3 - see the LICENSE.md file for details
- Much thanks to Giorgia Fortuna for guidance and mentorship,
- Stephen Wolfram for many meaningful discussions, and
- Karan Shah and other members of the Summer School for support, interest, and beer, and
- Billie Thompson PurpleBooth for the README template.