💤 Drowsiness Detection Web App

A real-time drowsiness detection prototype that uses your device camera (laptop/mobile/tablet) to capture an image, process it using deep learning models, and predict whether the driver is alert or drowsy.

---

🚀 Features

• Uses your webcam to take a snapshot.
• Detects eye status (open/closed) using a CNN-based model.
• Detects yawning using a separate image classifier.
• Combines both predictions with weighted logic to infer drowsiness.
• Models are preloaded to avoid lag during prediction.

---

🧰 Tech Stack

• Frontend: Streamlit + HTML + JS Webcam Interface
• Backend: Python
• Models:
  • eye_classifier_model.keras
  • yawn_detection_model.keras

---

Multiple Implementations

• The combinedModel.py code has various different implementation that can be used for testing, real applications that best fits the use case of the user.