Study on Fuzzy Logic Decision Support System for Hypovigilance Detection Based on CNN Feature Extractor and WN Classifier | Chapter 13 | Recent Advances in Mathematical Research and Computer Science Vol. 1

Excessive speed and drinking are the leading causes of road accidents, followed by fatigue and drowsiness. It is critical to regularly check the driver's alertness level in order to improve their capacity to drive safely and efficiently. This paper is about the issue of road safety. It makes an attempt to present a video-based driver vigilance monitoring system. The goal of this project is to develop an assistance driving application that uses eye closure length and head position estimate as effective indicators of alertness control. The suggested system may be broken down into three steps: video eye recognition and tracking, categorization of eye states, and integration of both sub-systems based on blinking and head position. Because of its efficiency in real-time applications, we employed the Viola and Jones algorithm for interest area recognition to complete the prior tasks. We used two innovative architectures of transfer learning classifiers based on fast wavelet transform and separator wavelet networks for the classification stage, which is the paper's key contribution. This unique design outperforms the standard version of transfer learning based on SVM classifiers as well as our previous classifier based solely on fast wavelet networks without a deep learning component. The goal of our research is to compare the performance of CNNs with wavelet networks in the classification phase. We also want to highlight the value of fuzzy logic as a tool for combining multiple inputs, allowing us to create a more accurate vigilance control system.

Author (S) Details

Ines Teyeb
RTIM: Research Team in Intelligent Machines, University of Gabes, National Engineering School of Gabes (ENIG), Tunisia.

Ahmed Snoun
RTIM: Research Team in Intelligent Machines, University of Gabes, National Engineering School of Gabes (ENIG), Tunisia.

Olfa Jemai
RTIM: Research Team in Intelligent Machines, University of Gabes, National Engineering School of Gabes (ENIG), Tunisia.

Mourad Zaied
RTIM: Research Team in Intelligent Machines, University of Gabes, National Engineering School of Gabes (ENIG), Tunisia.

View Book :-  https://stm.bookpi.org/RAMRCS-V1/article/view/4343