Skip to content

Latest commit

 

History

History
21 lines (14 loc) · 954 Bytes

README.md

File metadata and controls

21 lines (14 loc) · 954 Bytes

Satellite Levels

About

Turns satellite images into level files, parsed in Unity. Uses the TensorFlow library in Python for image recognition. Created during my third year of university for an Advanced Technologies module.

Instructions

The project as-is includes a few example levels that have already been parsed from images. These can be read with the Unity project. To create more, modify the 'classify_image' script and change the 'curImage' and 'saveFolder' variables. Running the script will then scan the given image if it can find it, and save the resulting file.

This process can take a fairly long time to complete, often up to an hour if not longer. Heightmaps were hard to obtain, and only one example is supplied in 'map1'.

Examples

Comparison video 1

Comparison video 2