Skip to content

README.md

Latest commit

 

History

History
55 lines (39 loc) · 1.51 KB

README.md

File metadata and controls

55 lines (39 loc) · 1.51 KB

🖼️ Image Denoising with Autoencoders

This project implements an Image Denoising Autoencoder using TensorFlow/Keras.
The model takes noisy images as input and learns to reconstruct clean images, improving image quality by removing noise.

🚀 Features

  • Autoencoder model for denoising.
  • Comparison of different architectures:
    • Fully connected autoencoder
    • Convolutional autoencoder with Conv2DTranspose.
  • Visualization of results before and after denoising.

🛠️ Installation & Setup

  1. Clone this repository:

    git clone https://github.com/yourusername/Image-Denoising-Autoencoder.git
cd Image-Denoising-Autoencoder

  2. Install dependencies:

    pip install tensorflow numpy matplotlib

  3. Run the notebook:

    jupyter notebook Image_Denoising.ipynb

📊 Results

Standard Autoencoder Output

Image

Convolutional Autoencoder Output (with Conv2DTranspose)

Image

🔮 Future Improvements

  • Experiment with deeper architectures.
  • Test on larger and more complex datasets.
  • Add comparison with other denoising methods (e.g., traditional filters).

🙌 Acknowledgments

This project was built as a practical exploration of image denoising using deep learning.