The number of automobiles around the world is increasing day by day. With the number of increasing automobiles, there is also a surge in road accidents annually. The primary causes of most of these accidents are drivers’ drowsiness and distraction. Even Today’s Autonomous vehicles need drivers’ attention to avoid accidents at times of emergency or system failure. So a real-time detection of drivers’ drowsiness level plays a pivotal role in reducing road accidents. In this paper, we present two algorithms for performing real-time detection of drivers’ drowsiness levels. The algorithms used are Multi-Contrast Convolution Neural Networks (MC-CNN) and Single Shot Multibox Detector (SSD). Our study can also be extrapolated into various other applications, such as real-time customer satisfaction detection through facial expressions in shopping malls.
Multi-level Drowsiness Detection using Multi-Contrast Convolutional Neural Networks and Single Shot Detector
25.06.2021
630870 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
DROWSINESS DETECTOR AND PROGRAM FOR DROWSINESS DETECTION
Europäisches Patentamt | 2024
|DROWSINESS DETECTOR AND PROGRAM FOR DROWSINESS DETECTION
Europäisches Patentamt | 2024
|Prediction of Driver Drowsiness Level Using Recurrent Neural Networks and Multi-Time-Scale Fusion
British Library Conference Proceedings | 2021
|