Recognition of Preoccupied Drivers on the Road Using Deep Learning Approach

Main Article Content

A.Nithya, Dr.S.Muthusundari, L.Sherin Beevi , J. Ashok Kumar

Abstract

 There are several reasons behind road accidents like violating road rules and regulations, over speed driving, etc. but one among those is drowsiness in driver. A sleepy/drowsy driver causes 40% of road accidents says the Central Road Research Institute (CRRI). Every year we lose nearly 1.5 lakh people in accidents caused due to the drivers dozing off on the wheels.  This issue can be solved by detecting the drowsy drivers and alerting them whenever they seemed to be dozing off using Convolutional Neural Network (CNN) algorithms. We capture the images of the person on the wheels and classify it as open or close using the CNN algorithm implemented on TensorFlow andKeras framework. When comparing the existing algorithm with the CNN algorithm the current one works fine with the binary classification process so that the accuracy of this system is higher than that of existing system.

Article Details

Section
Articles