🧠 Crowd Vision

Crowd Vision is an intelligent real-time crowd monitoring system that uses computer vision to estimate the number of people in a video feed and detect overcrowding conditions.
It is designed to enhance public safety, space management, and event monitoring using AI and OpenCV.

---

🚀 Features

• 🔹 Real-time crowd counting using a webcam or CCTV feed
• 🔹 Overcrowding alert system — triggers a warning when the crowd exceeds a threshold
• 🔹 Automatic logging of crowd count with timestamps in a CSV file
• 🔹 Simple visualization using a Tkinter GUI
• 🔹 Lightweight and local — no cloud dependencies

---

🧩 Tech Stack

• Programming Language: Python 3.x
• Libraries: OpenCV, Tkinter, CSV, datetime
• Environment: Local (works on Windows, macOS, or Linux)

---

⚙️ How It Works

1. The system captures a video stream using OpenCV.
2. Each frame is analyzed to estimate the number of people in view (using a pretrained model / contour-based detection).
3. The count is displayed on the GUI and logged into a CSV file with a timestamp.
4. If the number of people exceeds the predefined limit (e.g., 50), a ⚠️ Overcrowding Alert is printed to the terminal.

---

🖥️ GUI Preview

The Tkinter-based interface displays:

• Current crowd count
• Real-time video feed
• Logging indicator

---

📂 Project Structure

Crowd_Vision/ │ ├── crowd_vision.py # Main Python script ├── crowd_log.csv # Auto-created log file ├── requirements.txt # Python dependencies └── README.md # Documentation

---

🧠 Installation & Usage

1. Clone this repository

   git clone https://github.com/has257/Crowd-Vision.git
   cd Crowd-Vision
   
   

---

🧠 Installation & Usage

1. Clone this repository

   git clone https://github.com/has257/Crowd-Vision.git
   cd Crowd-Vision
   

Create a virtual environment: python3 -m venv venv source venv/bin/activate # macOS/Linux venv\Scripts\activate # Windows

Install dependencies pip install -r requirements.txt

Run the project python main.py

📊 Output Example Timestamp Crowd Count 2025-11-04 14:12:23 34 2025-11-04 14:13:05 57 ⚠️ Alert

🧩 Future Enhancements

Integrate a deep learning-based YOLO detector for improved accuracy

Add email/SMS alert system for overcrowding

Deploy on Raspberry Pi or Jetson Nano for edge AI use cases

Dashboard visualization using Streamlit or Flask

👩‍💻 Author

Hasmitha Siddani B.Tech CSE | AI & Computer Vision Enthusiast 📍 Gokaraju Rangaraju Institute of Engineering and Technology 🔗 GitHub | LinkedIn