Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/maturk/image-denoising
Image denoising algorithms in Python
https://github.com/maturk/image-denoising
computer-vision denoising denoising-images python
Last synced: 7 days ago
JSON representation
Image denoising algorithms in Python
- Host: GitHub
- URL: https://github.com/maturk/image-denoising
- Owner: maturk
- Created: 2023-02-03T18:15:56.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2023-02-05T19:49:57.000Z (almost 2 years ago)
- Last Synced: 2024-10-28T07:19:23.389Z (about 2 months ago)
- Topics: computer-vision, denoising, denoising-images, python
- Language: Python
- Homepage: https://maturk.github.io/page/2023/02/03/image_denoising.html
- Size: 1.74 MB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
A small collection of Image Based Denoising algorithms written in Python. Small writeup and demo images can be seen here: [link to demo.](https://maturk.github.io/page/2023/02/03/image_denoising.html) All algorithms are hard coded. Requirements are only numpy and Pillow (see requirements.txt).
## Algorithms:
- [x] Gaussian blur
- [x] Bilateral filter
- [ ] TODO: optimize bilateral
- [ ] Non-local means filter
- [ ] Neural network based denoising
- [ ] Intel® Open Image Denoise## Download
```
git clone [email protected]:maturk/image-denoising.git
cd image-denoising/
pip install -r requirements.txt
```### Gaussian Blur
Gaussian blur is one of the simplest denoising algorithms and it amounts to estimating
at each pixel position a local average of intensities and corresponds to low-pass filtering.```
python ./gaussian-blur/gaussian-blur.py --size 5 --sigma 1 --show True --save True
```
### Bilateral Filter
The bilateral filter is technique to smooth images while preserving edges. It consists of a weighted product of two gaussian kernels, one for pixel distances (sigma_space) and another for pixel intensities (sigma_color).```
python ./bilateral-filter/bilateral-filter.py --size 10 --sigma_color 10 --sigma_space 10 --show True --save True
```