PDF Optimizer

PDF Optimizer is a cross-platform application built with Flutter and Python, designed to optimize text-based PDF files. It uses Python scripts to enhance images, reduce noise, apply super-resolution, and recombine the optimized images into a PDF file.

---

Features

• PDF Upload: Upload PDF files and extract images from them.
• Image Optimization: Offers multiple image optimization options, including:
  • Text enhancement (enhance_text)
  • Median filtering (median_filt)
  • Bilateral filtering (bilateral_filt)
  • Gaussian blur (gaussian_blur)
  • Sharpening (sharpen)
  • Super-resolution (super_resolve)
• Custom Optimization Pipeline: Drag and drop to customize the optimization workflow.
• Optimization Preview: Compare images before and after optimization in real-time.
• Export Optimized PDF: Save the optimized images as a new PDF file.

---

Tech Stack

• Flutter: For building the cross-platform user interface.
• Python: For image processing and PDF optimization logic.
• PyMuPDF (fitz): For extracting images from PDFs.
• OpenCV: For image processing (filtering, sharpening, etc.).
• Real-ESRGAN: For image super-resolution (model source: Real-ESRGAN License: BSD 3-Clause License).

---

Installation

1. Clone the Repository

git clone https://github.com/blue-soda/pdf-optimizer.git
cd pdf-optimizer

2. Install Flutter Dependencies

Ensure you have the Flutter SDK installed, then run the following command to install dependencies:

flutter pub get

3. Install Python Dependencies

install Python dependencies:

pip install -r requirements.txt

---

Running the Project

1. Run the Flutter App

In the project root directory, run the following command to start the Flutter app:

flutter run

2. Using the App

Upload PDF:

Click the "Upload PDF" button to select a PDF file for optimization.

Select Optimization Options:

Drag and drop optimization options into the pipeline area and configure parameters.

Optimize Images:

Click the "Optimize Image" button to view the before-and-after comparison of the images.

Output PDF:

Click the "Output PDF" button to save the optimized images as a new PDF file.

---

Optimization Options

Option Name	Description	Example Parameters
enhance_text	Enhances text contrast for better readability.	factor=2.0
gaussian_blur	Applies Gaussian blur to smooth images and reduce Gaussian noise.	ksize=3
sharpen	Sharpens images to enhance edges and details.	kernel=1
median_filt	Applies median filtering to remove noise (e.g., salt-and-pepper).	ksize=5
bilateral_filt	Applies bilateral filtering to reduce noise while preserving edges.	d=9, sigmaColor=75, sigmaSpace=75
super_resolve	Applies super-resolution using Real-ESRGAN to increase image resolution.	No parameters required

License

This project is open-source under the BSD 3-Clause License. See the LICENSE file for details.