NITHISHM2410 / driver-drowsiness-detection

The driver's drowsiness detection system using TensorFlow and CNN models employs computer vision to analyze real-time images , allowing it to identify signs of driver fatigue and alertness and providing the driver with timely warnings.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

A study of machine learning-based driver drowsiness detection systems

The application of machine learning in driver drowsiness systems is crucial in enhancing road safety and mitigating accidents resulting from driver weariness. These systems has the capability to promptly identify signs of drowsiness by examining facial expressions, eye movements, and drowsiness indicators such as yawning. By disseminating prompt notifications, they aid in the prevention of accidents and offer drivers early indications, allowing them to manage fatigue prior to it reaching a critical state.

Usage :
Run the detector.py file to start the drowsiness detector. Two .h5 files that are accessible within the same directory are required in order to use the drowsiness detection module.The input to the detector module is an image. Videos must transmitted to the detector in sequence of frames so that real-time detection is possible and result for each frame is produced.

Dataset
The dataset is a standard image dataset for both Eye state and Yawn Detection consisting 600 images per class.
Access the dataset : https://www.kaggle.com/datasets/serenaraju/yawn-eye-dataset-new

About

The driver's drowsiness detection system using TensorFlow and CNN models employs computer vision to analyze real-time images , allowing it to identify signs of driver fatigue and alertness and providing the driver with timely warnings.


Languages

Language:Jupyter Notebook 99.9%Language:Python 0.1%