mdh.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
An Intelligent Non-Contact based Approach for Monitoring Driver’s Cognitive Load
Mälardalen University, School of Innovation, Design and Engineering, Embedded Systems. (Intelligent Future Technology)ORCID iD: 0000-0002-1547-4386
2018 (English)Licentiate thesis, comprehensive summary (Other academic)
Abstract [en]

The modern cars have been equipped with advanced technical features to help make driving faster, safer and comfortable. However, to enhance transport security i.e. to avoid unexpected traffic accidents it is necessary to consider a vehicle driver as a part of the environment and need to monitor driver’s health and mental state. Driving behavior-based and physiological parameters-based approaches are the two commonly used approaches to monitor driver’s health and mental state. Previously, physiological parameters-based approaches using sensors are often attached to the human body. Although these sensors attached with body provide excellent signals in lab conditions it can often be troublesome and inconvenient in driving situations.  So, physiological parameters extraction based on video images offers a new paradigm for driver’s health and mental state monitoring. This thesis report presents an intelligent non-contact-based approach to monitor driver’s cognitive load based on physiological parameters and vehicular parameters. Here, camera sensor has been used as a non-contact and pervasive methods for measuring physiological parameters.

The contribution of this thesis is in three folds: 1) Implementation of a camera-based method to extract physiological parameters e.g., heart rate (HR), heart rate variability (HRV), inter-bit-interval (IBI), oxygen saturation (SpO2) and respiration rate (RR) considering several challenging conditions e.g. illumination, motion, vibration and movement. 2) Vehicular parameters e.g. lateral speed, steering wheel angle, steering wheel reversal rate, steering wheel torque, yaw rate, lanex, and lateral position extraction from a driving simulator. 3) Investigation of three machine learning algorithms i.e. Logistic Regression (LR), Support Vector Machine (SVM) and Linear Discriminant Analysis (LDA) to classify driver’s cognitive load. Here, according to the results, considering the challenging conditions, the highest correlation coefficient achieved for both HR and SpO2 is 0.96. Again, the Bland Altman plots shows 95% agreement between camera and the reference sensor. For IBI, the quality index (QI) is achieved 97.5% considering 100 ms R-peak error. For cognitive load classification, two separate studies are conducted, study1 with 1-back task and study2 with 2-back task and both time domain and frequency domain features are extracted from the facial videos. Finally, the achieved average accuracy for the classification of cognitive load is 91% for study1 and 83% for study2. In future, the proposed approach should be evaluated in real-road driving environment considering other complex challenging situations such as high temperature, complete dark/bright environment, unusual movements, facial occlusion by hands, sunglasses, scarf, beard etc.

Place, publisher, year, edition, pages
Västerås: Mälardalen University , 2018.
Series
Mälardalen University Press Licentiate Theses, ISSN 1651-9256 ; 268
National Category
Computer Sciences
Research subject
Computer Science
Identifiers
URN: urn:nbn:se:mdh:diva-39187ISBN: 978-91-7485-389-6 (print)OAI: oai:DiVA.org:mdh-39187DiVA, id: diva2:1205169
Presentation
2018-06-19, Kappa, Mälardalens högskola, Västerås, 13:15 (English)
Opponent
Supervisors
Projects
SafeDriver:Available from: 2018-05-11 Created: 2018-05-11 Last updated: 2018-05-23Bibliographically approved

Open Access in DiVA

fulltext(2124 kB)224 downloads
File information
File name FULLTEXT05.pdfFile size 2124 kBChecksum SHA-512
694db5dcab232085c6521c36351880ae029c5ae0b05b58fc5d1f425f385098abd7c6d83a5e976317f41f22ecf5381cd41abcfbcd7d6bc2cd865ea898dfbb5891
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Rahman, Hamidur
By organisation
Embedded Systems
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 262 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 201 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf