Image Upscaler Web Application

This web application allows users to upload an image and upscale it using either traditional image interpolation techniques or Deep Neural Network (DNN) based super-resolution models. Built with Django, it provides a user-friendly interface for enhancing the resolution of images directly in the browser.

Features

• Traditional Upscaling: Offers several classic interpolation methods for image upscaling:
  • Lanczos
  • Linear
  • Cubic
• AI-Powered Upscaling: Integrates pre-trained DNN super-resolution models for more advanced upscaling:
  • EDSR (Enhanced Deep Residual Networks for Single Image Super-Resolution)
  • FSRCNN (Fast Super-Resolution Convolutional Neural Network)
  • ESPCN (Efficient Sub-Pixel Convolutional Network)
  • LapSRN (Laplacian Pyramid Super-Resolution Network)
• Scale Factor Selection: Users can choose the desired upscaling factor (2x, 3x, or 4x).
• File Handling: Supports uploading common image formats and downloads the upscaled image as a file.
• Temporary File Management: Cleans up temporary input and output files after processing.

Prerequisites

Before running the application, ensure you have the following installed:

• Python: Version 3.x
• pip: Python package installer
• Django: Version 5.x (or the version specified in requirements.txt)
• pillow Version 11.x
• OpenCV (cv2): Python library for computer vision
• Pre-trained Super-Resolution Models: These are expected to be located in a directory specified by MODEL_ROOT in your Django settings.

Installation

1. Clone the repository (if you have the code in a repository):

   git clone https://github.com/LPfontes/upImage.git
   cd upImage

2. Create a virtual environment (recommended):

   python -m venv venv
   source venv/bin/activate  # On macOS/Linux
   venv\Scripts\activate  # On Windows

3. Install the dependencies:

   pip install -r requirements.txt

4. Configure Django Settings:

   • Ensure you have MEDIA_ROOT and MODEL_ROOT defined in your settings.py file.
   • MEDIA_ROOT should point to the directory where uploaded and processed images will be temporarily stored.
   • MODEL_ROOT should point to the directory containing the pre-trained .pb files for the DNN super-resolution models (e.g., edsr_x2.pb, fsrcnn_x3.pb, etc.).
   • Make sure these directories exist and are writable by the Django application.

5. Apply migrations:

   python manage.py migrate

6. Create a superuser (optional, but recommended for admin access):

   python manage.py createsuperuser

7. Run the development server:

   python manage.py runserver

   Open your web browser and navigate to http://127.0.0.1:8000/ to access the application.

Usage

1. Navigate to the application in your web browser.
2. You will see two forms: one for traditional upscaling and another for AI-powered upscaling.
3. For Traditional Upscaling:
   • Click the "Choose File" button to upload an image.
   • Select the desired "Scale" (2x, 3x, or 4x).
   • Choose an "Interpolation" method (Lanczos, Linear, or Cubic).
   • Click the "Upscale Image" button.
4. For AI-Powered Upscaling:
   • Click the "Choose File" button to upload an image.
   • Select the desired "Scale" (the available scales depend on the chosen model).
   • Choose a "Model" (EDSR, FSRCNN, ESPCN, or LapSRN).
   • Click the "Upscale Image (AI)" button.
5. The upscaled image will be processed and automatically downloaded to your computer.

Contributing

Contributions to this project are welcome. Please feel free to submit pull requests or open issues for bug fixes, feature requests, or improvements.