sawyercharlton/image-dehazing-learning

This project is about the learning and testing of a dehazing method.

Python

image-dehazing

The method is from Effective Single Image Dehazing by Fusion by Codruta Orniana Ancuti, Cosmin Ancuti and Philippe Bekaert.

To group members

I write something helpful to presentation and report in quantitative_comparison.
sots is downloaded from Kaggle.
sots_oudoor is the output of our adapted method.

Pre-requisites

Package requirements:
- matplotlib
- numpy
- opencv-python
- scikit-image

Usage

Once the pre-requisites are satisfied, the inference can be run by running the following command, it produces dehazed images:
```
python inference.py
```
cal_metrics calculates PSNR and SSIM:
```
python cal_metrics.py
```