Engineeering & IT Projects and Resources: AI Mask Detection System

AI Mask Detection System

1. Introduction

The AI Mask Detection System is designed to detect and classify whether individuals are wearing a face mask in real-time. This system uses deep learning techniques and computer vision libraries such as OpenCV and TensorFlow/Keras. It is particularly useful in ensuring compliance with mask mandates in public spaces.

2. Prerequisites

• Python: Install Python 3.x from the official Python website.
• Required Libraries:
- opencv-python: Install using pip install opencv-python
- numpy: Install using pip install numpy
- tensorflow: Install using pip install tensorflow
- keras: Install using pip install keras
- pretrained mask detection model (if available).
• Webcam or camera-enabled device for real-time detection.

3. Project Setup

1. Create a Project Directory:

- Name your project folder, e.g., `AI_Mask_Detection`.
- Inside this folder, create the Python script file (`mask_detection.py`).

2. Install Required Libraries:

Ensure TensorFlow, Keras, and OpenCV are installed using `pip`.

4. Writing the Code

Below is an example code snippet for the AI Mask Detection System:

import cv2
import numpy as np
from tensorflow.keras.models import load_model

# Load pre-trained model for mask detection
model = load_model('mask_detection_model.h5')

# Load face detector (Haar cascades or DNN model)
face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')

# Function to predict mask usage
def detect_mask(face_img):
    face_img = cv2.resize(face_img, (128, 128))
    face_img = np.expand_dims(face_img, axis=0)
    face_img = face_img / 255.0
   prediction = model.predict(face_img)[0]
    return "Mask" if prediction[0] > 0.5 else "No Mask", prediction

# Main function for video stream processing
def main():
    video_capture = cv2.VideoCapture(0)

    while True:
        ret, frame = video_capture.read()
        if not ret:
            break

        gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
        faces = face_cascade.detectMultiScale(gray, scaleFactor=1.1, minNeighbors=5, minSize=(60, 60))

        for (x, y, w, h) in faces:
            face = frame[y:y+h, x:x+w]
            label, prediction = detect_mask(face)

            color = (0, 255, 0) if label == "Mask" else (0, 0, 255)
            cv2.rectangle(frame, (x, y), (x+w, y+h), color, 2)
            cv2.putText(frame, label, (x, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.8, color, 2)

        cv2.imshow('AI Mask Detection', frame)

        if cv2.waitKey(1) & 0xFF == ord('q'):
            break

    video_capture.release()
    cv2.destroyAllWindows()

if __name__ == "__main__":
    main()

5. Key Components

• Face Detection: Detects faces using Haar cascades or a pre-trained deep learning model.
• Mask Classification: Classifies detected faces as 'Mask' or 'No Mask' using a deep learning model.
• Real-Time Processing: Captures and processes live video streams.

6. Testing

1. Ensure the mask detection model is trained and saved as `mask_detection_model.h5`.

2. Run the script:

python mask_detection.py

3. Verify the detection and classification of faces in real-time.

7. Enhancements

• Accuracy Improvements: Train on a larger dataset for better model accuracy.
• Multi-Class Detection: Include additional classifications (e.g., improper mask usage).
• Notifications: Integrate alerts for 'No Mask' detections.

8. Troubleshooting

• Detection Issues: Ensure proper lighting and clear video input.
• Model Loading Errors: Verify the path and compatibility of the mask detection model.
• Performance Lag: Optimize the model or reduce input resolution.

9. Conclusion

The AI Mask Detection System is a practical application for promoting public safety in health-sensitive environments. With further improvements, it can be deployed in public areas, workplaces, and transportation hubs.

Pages