CamVisioTech is an AI-powered IoT-based security system project, collaboratively developed to implement advanced surveillance features with iterative enhancements. Each version builds upon its predecessor, integrating cutting-edge technologies for home and office security.
CamVisioTech aims to deliver comprehensive smart security solutions by leveraging:
- Facial Recognition: For automated user authentication.
- Object Detection: Real-time analysis using advanced detection algorithms.
- Automated Alerts: Notifications via Telegram and email.
- Live Streaming: Video feed and control interfaces through web apps.
- Hardware Integration: ESP32-CAM microcontroller for embedded AI applications.
With every iteration, CamVisioTech integrates improved hardware and software capabilities, ranging from basic facial recognition (MK-0) to advanced object detection and web-based control (MK-1).
The foundational version focuses on facial recognition for access control and intruder alerts:
- Face Recognition: Controls a solenoid lock to unlock doors for known individuals.
- Intruder Alerts: Sends images of unauthorized users via Telegram and activates a buzzer.
- Web Interface: Streams video feed and provides control options.
- Python GUI: Desktop application for managing the system with both high and low latency options.
- Solenoid lock control via a relay.
- Hardware enhancements like heat sinks for improved reliability.
- Python-based GUI and Telegram integration for real-time alerts.
- ESP32-CAM microcontroller
- Solenoid lock, relay module, buzzer
- IC-7805 voltage regulator
- Additional components: diode, transistor, resistors, and capacitors.
- Arduino IDE for programming ESP32-CAM.
- Python with libraries like
opencv-python,face_recognition, andcustomtkinter.
The enhanced version introduces real-time object detection and a Flask web application:
- Haar Cascade: Utilized for object and face detection.
- Alerts via Email & Telegram: Notifications for security breaches.
- Flask Web App: Streams live video feed with overlays.
- Hardware Improvements: Focused on scalability and performance.
- Understand and implement the Haar Cascade method.
- Develop Flask-based live monitoring and control systems.
- Integrate AI and IoT seamlessly for enhanced surveillance.
- Combines face recognition and object detection.
- Real-time alerts for unauthorized access.
- Web-based video streaming with detection overlays.
- ESP32CAM module
- Similar components as MK-0 with enhanced integration.
- Python and Flask for backend and web streaming.
- Libraries for detection:
opencv,face_recognition.
The MK-2 version of CamVisioTech moves beyond conventional surveillance by leveraging Edge AI for on-device processing. Unlike earlier iterations, it focuses on executing models locally, enhancing privacy, reliability, and efficiency, even in low-bandwidth environments.
- YOLOv2 Integration: Enables precise object detection and activity recognition directly on the device.
- Edge AI Processing: All inference operations are performed on the hardware itself (Maixduino), eliminating the need for cloud-based computations.
- Multi-Connectivity Support: Offers Wi-Fi or GSM for sending real-time alerts and notifications.
- Actuator Integration: Supports on-site physical responses such as buzzer alerts or relay-controlled actions.
- Alerts are dispatched via Wi-Fi or GSM, with extensibility to third-party apps like Telegram using platforms such as IFTTT or PipeDream.
- Maixduino RISC-V + AI Kit: AI-capable development board with integrated ESP32 for Wi-Fi and Bluetooth capabilities.
- OV2640 Camera Module: High-quality imaging for real-time object detection.
- Buzzer: For audio alerts.
- Breadboard & Jumper Wires: For modular connections.
- 2.4-inch TFT Display: For on-device status monitoring.
- Type-C Data Cable: For power and data transfer.
- MicroPython: Lightweight scripting language for development.
- MaixPy IDE: To build and deploy applications on the Maixduino hardware.
- kflash_gui: For uploading firmware and pre-trained models.
- uPyLoader: For accessing, updating, and managing files on the device.
- YOLOv2 Model Files: Optimized for real-time inference on the Maixduino platform.
- Integration of advanced AI models for object detection and activity recognition.
- Scalability for multi-camera setups.
- Mobile app for real-time control and notifications.
- Energy-efficient hardware for long-term deployments.
This project is licensed under the Apache 2.0 License - see the LICENSE file for details.
By following the documentation, users can deploy a robust AIoT-based security system with advanced features and modular enhancements.