Helmet Detection Project

This project uses the YOLO (You Only Look Once) object detection algorithm to detect helmets in images and videos. The project includes a web application for uploading and processing images to identify whether a person is wearing a helmet.

Features

• Image Upload: Allows users to upload images for helmet detection.
• YOLOv3 Integration: Utilizes YOLOv3 for accurate helmet detection.
• Flask Web Application: Provides an easy-to-use web interface for uploading and viewing results.
• Real-time Processing: Processes images in real-time and displays results immediately.
• Customizable: Easy to modify for detecting other objects by changing the YOLO configuration and weights.

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/helmet-detection.git
   cd helmet-detection

2. Create and activate a virtual environment (optional but recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows use `venv\Scripts\activate`

3. Install the required dependencies:

   pip install -r requirements.txt

4. Download YOLO weights:

   Download the pre-trained YOLO weights file from the official YOLO website and place it in the project directory.

Usage

1. Run the Flask application:

   python app.py

2. Interact with the application:

   • Open your web browser and navigate to http://127.0.0.1:5000/.
   • Upload an image for helmet detection.
   • View the processed image with detected helmets highlighted.

File Structure

The application is designed with a modular structure for scalability and ease of maintenance:

Helmet_detection/
│
├── __pycache__/         # Python bytecode files
├── outputs/             # Directory to save processed output files
├── templates/           # HTML templates for the web application
│   └── index.html       # Main page of the web application
├── uploads/             # Directory where uploaded images are stored
├── _config.yml          # Configuration file for the web application
├── app.py               # Main application code
├── helmet.names         # File containing names of the objects/classes for helmet detection
├── yolov3-helmet.cfg    # YOLO configuration file customized for helmet detection
├── yolov3.weights       # YOLO pre-trained weights file
├── requirements.txt     # List of required Python packages
└── README.md            # This readme file

Libraries Used

• Flask: For building the web application.
  • Library: flask
• OpenCV: For image processing and YOLO integration.
  • Library: opencv-python-headless
• NumPy: For numerical operations.
  • Library: numpy
• os: For accessing environment variables and handling other OS-level operations.
  • Library: os (standard Python library)

Flowchart

Output

test_video1.mp4

License

This project is licensed under the MIT License. See the LICENSE file for details.

Acknowledgments

• Flask for providing an excellent framework for building web applications.
• OpenCV for powerful image processing capabilities.
• YOLO for the robust object detection algorithm.