FFTrix

FFTrix is a high-performance video surveillance and analysis platform utilizing classical Computer Vision (CV) and pattern recognition techniques for real-time monitoring and event detection.

Features

• Real-time Monitoring: Multi-camera streaming with low-latency processing via vidgear.
• Classical CV Analytics:
  • Motion Detection: Background subtraction using the MOG2 (Mixture of Gaussians) algorithm.
  • Person Detection: Histogram of Oriented Gradients (HOG) with Linear SVM.
  • Face Detection: Local Binary Patterns (LBP) and Haar Feature-based Cascade Classifiers.
  • LPR (License Plate Recognition): Morphology-based plate localization combined with Tesseract OCR.
  • Zone Monitoring: User-defined exclusion and inclusion zones for targeted event triggering.
• 24/7 Recording: Continuous background recording with automatic retention management.
• Interactive Dashboard: Modern web-based interface built with NiceGUI for camera management, live viewing, and event playback.
• Automated Alerts: Event-driven notifications via Email (SMTP) and JSON Webhooks.
• PTZ & Discovery: ONVIF-compatible device discovery and Pan-Tilt-Zoom control.
• Privacy Controls: Dynamic privacy masking with Gaussian blurring.

Architecture

FFTrix is built on a "Classic CV" architecture, prioritizing deterministic behavior and low resource overhead:

• Vision Engine: Pure OpenCV-based pipeline (no GPU/Deep Learning dependencies).
• OCR Engine: Tesseract (via pytesseract) for text and plate recognition.
• Database: SQLite for configuration, user management, and event logging.
• GUI: NiceGUI (FastAPI + Vue.js + Quasar) for a responsive management interface.

Installation

FFTrix can be installed directly from PyPI:

pip install fftrix

Or using uv as a tool:

uv tool install fftrix

Prerequisites

• Python 3.14 or higher.
• Tesseract OCR: Required for LPR and OCR features. Install it via your system package manager (e.g., apt install tesseract-ocr or brew install tesseract).
• Docker (for secure deployment): NOTE: Although you can replicate the deployment process without Docker, it will not be supported by me or anyone else from my team.
• ngrok: Can be implemented for remote access. While not officially supported, you can contact the developer for guidance.

Quick Start

Launch the FFTrix surveillance server:

fftrix serve --ui 

The dashboard will be available at http://localhost:8080.

Default Credentials

• Username: admin
• Password: admin (Note: A password change is required upon first login)

CLI Usage

FFTrix includes a Click-based CLI for server control and user management.

# Start headless server (no local browser launch)
fftrix serve --no-ui  # ***NOTE: The UI is REQUIRED for configuration of watermarking and zone configuration but once configured the server can be run headless***

# Provision a new operator
fftrix user add <username> --role [admin|viewer]

# Emergency reset (wipes all users, restores admin:admin)
fftrix user reset-admin

Configuration

Configure the notification engine via environment variables:

Tech Stack

• Framework: NiceGUI
• CLI: Click
• Core Vision: OpenCV
• OCR: Tesseract
• Networking: VidGear, ONVIF-Zeep

Development

If you want to contribute to FFTrix start by forking the repository then checking the requirements listed in the CONTRIBUTING.md file.

# Clone the repository
git clone https://github.com/yourusername/fftrix.git
cd fftrix

# Sync dependencies using uv
uv sync

License

MIT License