IJFANS International Journal of Food and Nutritional Sciences

ISSN PRINT 2319 1775 Online 2320-7876

DRIVER SAFETY ENHANCEMENT WITH REAL-TIME EMOTION RECOGNITION AND DROWSINESS DETECTION USING DEEP LEARNING AND COMPUTER VISION

Main Article Content

Dr V Surya Narayana, Sai Krishna Pothini, Nikitha Muthineni, Sailaja Gandipudi

Abstract

A key factor that contributes to traffic accidents is driver weariness. In this research, we present a complete Multimodal Driver Tiredness Detection System that analyses yawning and eye closure in real-time using computer vision and deep learning algorithms. To improve the precision and resilience of tiredness diagnosis, the system incorporates facial feature identification, emotion recognition, and sleepiness prediction. The system utilizes Haar Cascade classifiers for face and eye detection, coupled with deep learning models to predict yawning and eye closure events. A Convolutional-Neural-Networkis employed for yawning detection, while a pre-trained model is utilized for eye closure analysis. Additionally, the system leverages OpenCV and TensorFlow for efficient computer vision and deep learning implementations. Real-time feedback is provided to the driver through a graphical user interface, displaying the status of yawning, eye closure, and frame count. An audible alarm is triggered in the event of prolonged eye closure, serving as an additional safety measure. The combination of these features results in a versatile and effective system for mitigating driver fatigue, ultimately contributing to road safety. This project showcases the integration of multiple technologies to address a crucial safety concern and underscores the potential of multimodal approaches in enhancing, the accuracy and responsiveness of fatigue detection systems

Article Details