This project presents an AI-based ground surveillance system using Raspberry Pi 4 for real-time human and weapon detection. The system enhances security monitoring by integrating computer vision techniques with an embedded platform. A Pi Camera module continuously captures live video, and a lightweight YOLOv5s model is deployed on the Raspberry Pi 4 to identify humans and weapons in real time.

When a human or weapon is detected, the system immediately triggers multiple alert mechanisms. The detection result is displayed on an LCD screen through an I2C interface, and a buzzer is activated to provide an audible warning. Simultaneously, the system captures an image of the detected object and sends an email notification to the registered user along with the captured photo as evidence for remote monitoring and quick response.

The system operates using an SD card for storage and is powered through an adapter supply. This compact, cost-effective, and intelligent surveillance solution is suitable for ground security applications such as border security, restricted areas, military zones, and smart surveillance systems.

  The main objectives of this project are:

• To implement real-time video capture using Pi Camera.
• To deploy a lightweight YOLOv5s model on Raspberry Pi 4.
• To detect humans and weapons using deep learning.
• To display detection results on an LCD screen.
• To trigger a buzzer for immediate local alerts.
• To capture and store detected images.
• To send email notifications with image evidence.
• To develop a cost-effective embedded surveillance solution.

  The major building blocks of this project are:  

1. Adapter power supply.
2. Raspberry pi4.
3. SD card.
4. Pi Camera.
5. Buzzer.
6. I2C along with LCD display.

    Software’s used:  

• Raspbian OS.
• Express SCH for Circuit design.

• lightweight YOLOv5s model for Object detection.
• IoT technology for sending mail.

     

video: