This is a TensorFlow implementation of paper:
Unsupervised Single Image Dehazing using Dark Channel Prior Loss
Based on :repository
Abstract: Single image dehazing is a critical stage in many modern-day autonomous vision applications. Early prior-based methods often involved a time-consuming minimization of a hand-crafted energy function. Recent learning-based approaches utilize the representational power of deep neural networks (DNNs) to learn the underlying transformation between hazy and clear images. Due to inherent limitations in collecting matching clear and hazy images, these methods resort to training on synthetic data; constructed from indoor images and corresponding depth information. This may result in a possible domain shift when treating outdoor scenes. We propose a completely unsupervised method of training via minimization of the well-known, Dark Channel Prior (DCP) energy function. Instead of feeding the network with synthetic data, we solely use real-world outdoor images and tune the network's parameters by directly minimizing the DCP. Although our "Deep DCP" technique can be regarded as a fast approximator of DCP, it actually improves its results significantly. This suggests an additional regularization obtained via the network and learning process. Experiments show that our method performs on par with large-scale supervised methods.
This section will consists of a research based on Single Image Haze Removal Using Dark Channel Prior paper.
We have been assuming that the atmospheric light is known. In the previous works, the color of the most haze-opaque region is used as initial guess. However, little attention has been paid to the detection of the “most haze-opaque” region. The brightest pixels in the hazy image are considered to be the most haze-opaque. This is true only when the weather is overcast and the sunlight can be ignored. In this case, the atmospheric light is the only illumination source of the scene.
Estimating the atmospheric light. (a) Input image. (b) Dark channel and the most haze-opaque region. (c) The patch from where our method automatically obtains the atmospheric light. (d), (e) Two patches that contain pixels brighter than the atmospheric light.
We tried to Estimate the Atmospheric Light. For this we downloaded (partially) relevant pictures of ImageNet, cleaned it up a bit and ran the best network of the article on it. On figure below one can find comparison of pictures in terms of PSNR. In middle is the result of the paper model and in the right our improvement. We can see that the PSNR value increases.
This repository contains:
- Test function for each application:
Test_Dehaze.py - Implementations of all energy functions:
Dehazing_Loss.py - Neural network model we use throughout all experiments:
Models.py - Training function to train your own model:
Deep_Energy.py - Implementations of Our Contribution to the Dehazing Algorithm:
our_dehaze.py - Utilities:
train_utils.py,Utils.py - Configuration parameters file:
params.ini
To perform test on saved models, you need the following:
- TensorFlow
- Numpy
- Scipy
- configargparse
- Matplotlib
To perform simple test, open Test_Dehaze.py in you IDE and run the script. Alternatively, from the \Code folder in the main repository, enter the following in the command:
python Test_Dehaze.py
The model parameters used in our papers are located in the \Results folder, under each application.
Some applications, for example single image dehazing contain several checkpoints: 27,30,33. For mild haze use 27 and for heavier amounts of haze use 30 and 33.
When new training will be performed, other sub folders in \Results will be created with updated timestamps.
The datasets we use are all linked below: HSTS, SOTS indoor and SOTS outdoor can be found in the following link.
.
├─ README.md
├─ Code
├─ params.ini (Configuration parameters file)
│ └── report.pdf
├─ Datasets (source images)
│ └──hazy
│ └── ... (input images)
├─ Results (the results)
│ ├─ checkpoints (saved models)
│ └── Qual_output (output images)
│ └── paper ...
│ └── our ...
└─ Code (the python source code)
├── Test_Dehaze.py (dehazing using the dark channel prior, generate the results for the report)
├── Models.py (Neural network model we use throughout all experiments)
├── Dehazing_Loss.py (Implementations of Dehazing energy function)
├── our_dehaze.py (Implementations of Our Contribution to the Dehazing Algorithm)
├── Utils.py (utilities)
└── train_utils.py (utilities)