Welcome to Image Dehazer, a Python tool for removing haze from images. This repository provides a comprehensive solution for enhancing the clarity and quality of hazy images.
- Removal of haze from images using a guided filter-based algorithm.
- Graphical user interface (GUI) for easy selection and processing of images.
- Sample input images provided for testing the tool.
The repository is organized as follows:
-
gf.py: Contains functions for guided filtering.
-
haze_remover.py: Implements the haze removal algorithm using guided filtering.
-
main.py: Provides a graphical user interface (GUI) for selecting and processing images.
-
Sample Input Images: Provides the images for testing and demonstration.
-
img/: Directory containing sample input images (src.jpg, dark.jpg, trans.jpg, gtrans.jpg) and the resulting dehazed image (dst.jpg).
- src.jpg: Original hazy image, serving as input for haze removal.
- dark.jpg: Dark channel prior of the original hazy image.
- trans.jpg: Estimated transmission map of the original hazy image.
- gtrans.jpg: Alternative transmission map estimation for comparison.
- dst.jpg: Resulting dehazed image after applying Image Dehazer algorithm.
-
dehazed.jpg: Resulting dehazed image after processing using the tool.
-
requirements.txt: To specify the dependencies required by the project.
Getting started with Image Dehazer is simple. Follow these steps to set up the tool on your machine:
- Clone Repository: Clone this repository to your local machine:
git clone https://github.com/Danish-Jamil-01/Image-Dehazer.git
- Navigate to Directory: Move into the cloned repository directory:
cd Image-Dehazer
- Install Dependencies: Install the required dependencies using pip:
pip install -r requirements.txt
- radius=7
- omega=0.95
- t0=0.1
- r=20
- eps=0.001
After installing the dependencies, you can use the Image Remover Tool as follows:
- Run Script: Execute the main.py script to launch the graphical user interface (GUI):
python main.py
-
Select Image: Use the GUI to select an image for haze removal. Alternatively, you can manually enter the image path.
-
Process Image: Click the "Submit" button to initiate the haze removal process. The tool will display the dehazed image in the GUI.
-
Save Output: The dehazed image will be saved as dehazed.jpg in the repository directory.
Contributions to this project are welcome. Feel free to open an issue or submit a pull request.
- Single Image Haze Removal using Dark Channel Prior
- Guided Image Filtering
- Single Image Dehazing via Conditional Generative Adversarial Network
- Single image dehazing based on learning of haze layers
- Dark channel prior in low frequency domain for time-efficient single image dehazing
This project is licensed under the MIT License.