November 21, 2025 BY
IN Raspberry Pi Projects

🔍 Project Overview

This project uses a Raspberry Pi, Pi Camera, and MobileNet-SSD deep learning model to detect the number of people in real time. It marks people with bounding boxes, displays the count, and saves automatic snapshot logs at regular intervals.

The system runs independently on Raspberry Pi and is suitable for both indoor and outdoor monitoring.

⚙️ How It Works (Short & Simple)

  1. Camera captures live video
    RPi Camera streams frames at 640×480 resolution.
  2. AI model detects humans
    MobileNet-SSD identifies “Person” class in each frame.
  3. People are counted
    Bounding boxes highlight each detected person.
  4. Auto-snapshot
    Every fixed interval (10 seconds by default), the system saves:
    • Image of the crowd
    • Timestamp
    • Detected count
  5. Excel logging (if enabled)
    Each record is stored in an Excel sheet for analysis.

🛠️ Hardware Used

  • Raspberry Pi 4 / Pi 3
  • Raspberry Pi Camera Module (any version)
  • Micro SD card
  • Power supply
  • Internet for installation (optional later)

✨ Key Features

✔ AI-based person detection
✔ Real-time crowd counting
✔ Bounding box visualization
✔ Auto image saving with timestamp
✔ Optional Excel data logging
✔ Lightweight model works smoothly on Raspberry Pi
✔ No cloud needed — runs fully offline

📈 Applications

  • Mall & shop crowd monitoring
  • School/college safety
  • Industrial workplace monitoring
  • Public event analysis
  • Smart security systems
  • Queue management
  • Temple/church crowd analysis

🚀 Future Scope

  • Email alert integration
  • Live dashboard via Flask web server
  • IoT cloud syncing (Firebase / AWS / Thingspeak)
  • SMS alerts using GSM module
  • Thermal camera support
  • Face mask + helmet detection
  • Real-time crowd density heatmap

👍 Advantages

  • Works offline
  • Low power consumption
  • Low-cost hardware
  • Accurate for close-to-medium range
  • Easy image and data export
  • Highly customizable

⚠️ Precautions

  • Ensure proper lighting for camera
  • Keep lens clean for best accuracy
  • Install camera firmly; avoid vibrations
  • Use heat-sink on Raspberry Pi to prevent throttling
  • Avoid exposing Pi Camera to sunlight for long durations