Master Thesis MSTR-2011-05

BibliographySlieter, Daniel: Development and Evaluation of a Prototype Metrology System for Robust Driver Attention Monitoring within Automotive Environments.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Master Thesis No. 5 (2011).
111 pages, english.
Abstract

Within this master-thesis, a prototype metrology tool is developed in order to quantify the quality of video-based driver attention monitoring systems. This tool allows to detect and track the driverís eyes as well as the estimation of the pose of the driverís face. To realize this system, several algorithms are identified, evaluated and merged. Accurate eye-tracking and gaze-detection implies the detection of small pupil movements, which may happen when taking a short look to the instrument panel. Therefore, an innovative approach is shown in this thesis, which adapts to different environmental conditions and tracks the driverís pupils. A comparison to a competitorís prototype has shown that the developed system still has further room for improvement, but already provides robust results. With the detected pose of the driverís head and the eye-positions, other attention monitors can be evaluated and compared in the future. Furthermore, this system enables driverindependent data recording within driver-studies, because instead of video-files just the headís rotations and the eye-positions are needed to be stored. This recorded data can be analyzed regarding the driverís attentiveness.

Department(s)University of Stuttgart, Institute of Software Technology, Software Engineering
Superviser(s)Levi, Prof. Paul, Schanz, Dr. Michael; Gebhard; Dr. Matthias
Entry dateMay 24, 2019
   Publ. Computer Science