Dissertation/Thesis Abstract

Drowsiness detection using HRV analysis
by Padmanabhan, Sivasankar, M.S., California State University, Long Beach, 2015, 89; 1596988
Abstract (Summary)

The field of drowsiness detection is gaining more attention these days. An estimate by National Highway Traffic Safety Administration states that the total number of people falling asleep on the wheel is increasing day by day. If there is an effective way to monitor this condition and alert the drivers, many fatal accidents can be prevented. This thesis work elaborates on one such simple, yet effective drowsiness detection algorithm, the HRV - Heart Rate Variability analysis. Many psychological researchers have found out that when a person becomes drowsy, there is a variation in their heart signal. Monitoring this physiological variation would be more efficient than monitoring their facial movements such as blinking, eye brow contraction, and yawning, which are said to happen after much longer time when compared to the immediate changes in the heart rate. Hence, an algorithm that detects drowsiness based on HRV analysis is developed and implemented by analyzing heart signals. Simple hardware setups were used to collect the ECG data, and digital filters were used to remove noise and extract the desired information for further analysis. The developed algorithm was implemented successfully and the results obtained were more precise and satisfactory. This approach of monitoring drowsiness is more reliable and accurate and when implemented with its necessary features, it can monitor drowsiness more effectively and save hundreds of lives every day.

Indexing (document details)
Advisor: Mozumdar, Mohammad
Commitee: Chassiakos, Anastasios, Khoo, I-Hung
School: California State University, Long Beach
Department: Electrical Engineering
School Location: United States -- California
Source: MAI 55/01M(E), Masters Abstracts International
Subjects: Electrical engineering
Keywords: Drowsiness, Heart rate variability analysis
Publication Number: 1596988
ISBN: 978-1-321-99192-5
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy