Vision-Activated
Control System

Facial Recognition in Embedded Systems

Developed by Ing. Anfilove

ICIR2025 Success Published Academic Edition IoT & Accessibility

The Dawn of FICS

Facial Interaction Control Systems (FICSs) represent a new generation of human-machine interfaces. They are fast, intuitive, and completely hands-free. This breakthrough project explores the design and development of a Vision-Activated Control System that interprets micro-expressions, eye movements, and facial gestures to control physical hardware in real time.

Relying heavily on modern Computer Vision, Convolutional Neural Networks (CNNs), and Human-Computer Interaction (HCI) principles, this prototype allows users to toggle devices using nothing but a smile, a blink, or a spoken phrase.

The Prototype in Action

Operating the system is seamless and interactive:

  • Booting Up: The ESP8266 microcontroller powers on, connects to Wi-Fi, and displays its IP address on the integrated I2C LCD.
  • Activation: The user smiles at the webcam. The system detects the expression, switches off the standby LED, and illuminates the "System Ready" indicator.
  • Gaze Control: Looking left or right selectively controls physical LEDs (D5 and D6).
  • Complex Gestures: A combination of a smile and three quick blinks toggles a 12V DC motor via an isolated relay (D4).
  • Voice Overrides: Saying "Stop please" pauses all operations instantly, while "Go on please" resumes the workflow.

Split-Stack Architecture

The system is elegantly divided between high-performance computing and low-latency embedded processing:

  • Python Vision Logic (PC): Leverages OpenCV, dlib, GazeTracking, and SpeechRecognition to extract facial signals and voice inputs in threaded processes.
  • Signal Relay Module (PC): Continuously maps recognized events (like a blink or gaze) to URLs (e.g., /S, /SB) and fires HTTP GET requests over the local network.
  • ESP8266 Hardware Control (C++): Acts as an embedded web server running Arduino code, parsing incoming HTTP requests to manipulate GPIO pins and provide real-time LCD feedback.
  • Component Tester Utility: A standalone web-interface sketch for the ESP8266 used during prototyping to manually validate all hardware endpoints.

Hardware Blueprint

At the physical level, the FICS prototype is an Internet-of-Things (IoT) powerhouse built on affordable, accessible hardware:

Visualizing the Data & Setup

Future Trajectory & Industry Applications

The implications of this vision-driven controller extend far beyond a lab demo. As detailed in the ICIR2025 presentation, this adaptable architecture paves the way for:

Smart Wheelchair Control Contactless Access Systems Factory Machine Safety Toggles Hospital Bedside Utilities Interactive Museum Exhibits Automotive Fatigue Systems Voice-Triggered Alerts Educational IoT Kits

By prioritizing accessibility, low latency, and robust cross-platform communication, this model serves as a foundational blueprint ready to evolve into countless critical real-life applications. This isn’t the future — it’s already here.