Image-Enhancement

Super-Resolution Convolutional Neural Network (SRCNN) Project

Project Link

For more details, visit the project repository.

Overview

This project focuses on enhancing low-quality images using a Super-Resolution Convolutional Neural Network (SRCNN). The SRCNN model is trained to take low-resolution images as input and output high-resolution images. The dataset consists of pairs of low-quality and high-quality images.

Project Structure

normalize.py:

Contains functions for image preprocessing such as loading, resizing, and normalizing images.

plot.py:

Contains functions for plotting original and enhanced images side by side.

train.py:

Main script for training the SRCNN model.

srcnn_model.h5:

Pre-trained SRCNN model.

low_quality_images/:

Directory containing low-quality images.

high_quality_images/:

Directory containing corresponding high-quality images.

How to Use

Preprocess the Images: Ensure all low-quality and high-quality image pairs are processed and normalized.
Train the Model: Run train.py to train the SRCNN model using the preprocessed image pairs.
Test the Model: Use a low-quality test image, preprocess it, and use the trained SRCNN model to enhance it. Plot the original and enhanced images side by side.

Requirements

Python 3.x
TensorFlow
NumPy
Pillow
Matplotlib

Installation

Install the required packages using pip:
pip install tensorflow numpy pillow matplotlib

Steps to Execute

Ensure the low_quality_images and high_quality_images directories contain the appropriate image pairs.
Execute the train.py script to train and save the model.
Use the trained model to enhance new low-quality images and visualize the results using the plot_image function.

Acknowledgements

This project uses the SRCNN architecture as described in the paper “Image Super-Resolution Using Deep Convolutional Networks” by Chao Dong et al.