Face Mask and Social Distance Detection Using Deep Learning Models

This project implements deep learning models to detect face mask usage and monitor social distancing in real-time. It is designed to assist in mitigating the spread of communicable diseases by ensuring compliance with health guidelines in crowded places such as airports, hospitals, and workplaces. This work is associated with a research paper published in Springer, accessible here: Face Mask and Social Distance Detection Using Deep Learning Models.

Features

• Face Mask Detection: Identifies individuals wearing or not wearing masks with high accuracy.
• Social Distance Monitoring: Measures the distance between individuals and flags violations in real-time.
• Deep Learning Models: Utilizes MobileNet, YOLO, and ResNet-50 architectures for efficient detection.
• Real-Time Processing: Processes video feeds to provide immediate feedback on compliance.

Technologies Used

• Programming Language: Python 3.x
• Deep Learning Frameworks: TensorFlow, Keras
• Computer Vision Library: OpenCV
• Development Environment: Anaconda
• Model Architectures: MobileNet, YOLO, ResNet-50

Installation & Setup

1. Clone the repository:

   git clone https://github.com/SaumyaGupta907/FaceMaskandSocialDistanceDetectionusingDLLmodels.git
   cd FaceMaskandSocialDistanceDetectionusingDLLmodels



2. Set up the Anaconda environment:

   conda create --name facemask_detection python=3.8
   conda activate facemask_detection



3. Install dependencies:

   pip install -r requirements.txt



4. Download pre-trained model weights: Ensure that the pre-trained weights for MobileNet, YOLO, and ResNet-50 are downloaded and placed in the appropriate directories as specified in the project structure.

5. Run the application:

   python live_inference.py



This script initiates real-time detection using your webcam.

Usage

• Real-Time Detection: Execute live_inference.py to start the webcam feed and perform live detection of face masks and social distancing violations.

• Training Models: Use train_mask_detector.py to train the face mask detection model on a custom dataset. Ensure that your dataset is structured appropriately and the paths are updated in the script.

• Testing: Utilize test_video.mp4 or any other video file to test the detection algorithms on pre-recorded footage.

Results & Performance

The models have demonstrated high accuracy in detecting face masks and monitoring social distancing. Detailed performance metrics, including precision, recall, and F1-score, are discussed in the associated research paper.

Research Contribution

This project is part of a research initiative aimed at leveraging deep learning for public health safety. The methodologies and findings are elaborated in the research paper published in Springer: Face Mask and Social Distance Detection Using Deep Learning Models.

Future Enhancements

• Integration with Surveillance Systems: Enhance the system to integrate seamlessly with existing CCTV infrastructures.
• Edge Deployment: Optimize models for deployment on edge devices to enable real-time processing without reliance on cloud services.
• Expanded Detection Capabilities: Extend the system to detect other compliance measures, such as hand hygiene practices.

Contributing

Contributions are welcome! If you have suggestions or improvements, please fork the repository and submit a pull request. Ensure that your contributions align with the project's objectives and maintain code quality standards.

Contact

• Authors: Saumya Gupta, N. Anusha, Y. Nikitha Naidu, M. Ruchitha & Richal Pandey
• LinkedIn: Saumya Gupta