Image-Colorization

This project aims at colorization of Astronomical images

Background:

Astronomical images are generally grey-scale due to lack of suffiecient light that enters the telescope, the color images are genrated by ong exposure and by adding filters to the telescopes, long exposures are time consuming, typically about 20-40 hours. This project aims at colorizing the grey-scale images using Generative-Adversiaral-Networks

Note

This project is still in the early stages of development

How to use

Check out the project into your favourite IDE, open 'colorizer_tester.py' and change the 'dir_path' variable to a test image oath of your choice, and train it, once its trained model files will be created in Images/Model folder, while testing just comment the train part of the code and run it again

I strongly acknowledge the below resources that lead to the development of the project

1. https://arxiv.org/abs/1803.05400

2. https://github.com/ImagingLab/Colorizing-with-GANs

3. http://bamos.github.io/2016/08/09/deep-completion/