Drowsiness detection system

This project is a smart desktop application designed to improve road safety by preventing accidents caused by fatigue. Using Deep Learning and Computer Vision, the system monitors a driver or operator in real-time through a camera to detect early signs of drowsiness.

The application works by using a Convolutional Neural Network (CNN), which is an AI model trained to recognize whether a person's eyes are open or closed. When the camera captures a face, the software isolates the eye area and analyzes it frame by frame. If the system detects that the eyes remain closed for too long, a "drowsiness score" increases. Once this score reaches a dangerous limit, the app immediately triggers a loud sound alarm and displays a red visual warning to alert the user.

Technically, the project is built using Python. It uses OpenCV for video processing, TensorFlow to run the AI model, and Dash to create the visual interface. To make sure the video stays smooth and the detection is fast, the app uses multithreading. This means the computer does two things at the same time: one part of the program analyzes the video, while the other part updates the screen. This ensures a professional and reliable safety tool that can work on any standard laptop with a webcam.