Obscurra

Obscurra is a Python-based image processing tool designed to detect faces in images and apply a blur effect to anonymize them. The project aims to enhance privacy by automatically processing images to obscure identifiable facial features through a user-friendly graphical interface.

This repository contains the source code, documentation, and models used by the tool.

Features

• Detects faces in images using Haar cascades.
• Blurs detected faces to anonymize them.
• Supports multiple image formats: JPG, JPEG, PNG, and WEBP.
• Modular design for easy maintenance and extensibility.

Objectives

• Develop a tool to detect and blur faces in images.
• Provide a simple GUI for image processing.
• Compile the application into a portable executable.

Timeline

Phase 1: Planning (January - February 2024)

• Research: Investigated face detection models and image processing techniques.
• Requirements: Defined project scope and objectives.

Phase 2: Development (March - June 2024)

• Core Features: Implemented face detection using Haar Cascade and MTCNN.
• GUI: Developed a Tkinter-based user interface.

Phase 3: Testing and Optimization (July 2024)

• Testing: Identified and resolved bugs.
• Optimization: Enhanced performance and accuracy.

Phase 4: Deployment (August 2024)

• Executable: Packaged application using PyInstaller.
• Documentation: Created user and developer guides.

Achievements

• Integrated multiple face detection models.
• Developed a portable executable with a comprehensive GUI.
• Enhanced user experience with real-time image processing updates.

Challenges and Bugs

• Model Integration: Ensured compatibility between different face detection models.
• Performance: Required optimization for processing speed and accuracy.
• GUI Usability: Improved interface design based on user feedback.

Methodology

The project followed an agile development process with iterative cycles of development, testing, and refinement, focusing on usability and performance.

Technologies Used

• Python: Core language
• Tkinter: GUI framework
• OpenCV: Image processing
• TensorFlow: MTCNN model
• PyInstaller: Executable packaging

Project Structure

Obscurrra/
├── src/
│   ├── Obscurrra.py        # Main script
│   ├── icon.ico            # Application icon
│   ├── haarcascade_frontalface_default.xml # Frontal face detection model
│   ├── haarcascade_profileface.xml # Profile face detection model
│   ├── mtcnn_weights.npy   # MTCNN model weights
└── dist/                   # Compiled executable

User Guide

1. Download the Executable: Obtain from the release section.
2. Run the Application: Launch the GUI executable.
3. Process Images:
   • Select input and output folders.
   • Choose detection models and set preferences.
   • Click "Start Processing."

Developer Guide

Setup

1. Clone the Repository:

   git clone https://github.com/jakujobi/obscurrra.git
   cd obscurrra

2. Create a Virtual Environment and Install Dependencies:

   python -m venv venv
   source venv/bin/activate  # On Windows use `venv\Scripts\activate`
   pip install -r requirements.txt

3. Run the Application:

   python src/Obscurrra.py

Building the Executable

1. Install PyInstaller:

   pip install pyinstaller

2. Compile the Application:

   pyinstaller --onefile --windowed src/Obscurrra.py

Future Improvements

• Explore advanced face detection models.
• Add real-time video processing capabilities.
• Enhance cross-platform compatibility.

Contributors

• John Akujobi- Project Lead, UI Designer, Software Engineer

---

This documentation provides a comprehensive yet concise overview of the Obscurrra project, highlighting key features, challenges, and future directions for both users and developers.