ECG-Based Driver's Pressure Location Utilizing Profound Exchange Learning And Fluffy Rationale Approaches

DOI

10.2991/978-94-6463-471-6_142How to use a DOI?

Keywords

Darknet53; Xception; Google Net; Convolutional Neural Network (CNN); and Electrocardiogram (ECG)

Abstract

The stress experienced by drivers is modelled in this article utilising multiple pre-trained networks. Seven pre-trained networks were utilised to extract features from ECG-based scalogram images in order to automatically increase the detection performance: Google Net, DarkNet-53, ResNet-101, InceptionResNetV2, Xception, DenseNet-201, and InceptionV3. To lessen the likelihood of car accidents and health problems caused by drivers’ stress, driver stress detection has emerged as a major area of study. The majority of the prior research in this field relies on feature extraction techniques to manually classify the driver's stress levels using typical machine learning models. Finding the best characteristics using these methods is never easy. More recently, deep learning methods have been developed to automatically build trustworthy features and classify data with great accuracy. However, issues with gradient disappearing or growing are encountered by large deep learning models. Building a large dataset from beginning might also be a challenge when training a full network. In order to circumvent these issues and save computational time and money, this study is constructed on top of the deep transfer learning technique. Using electrocardiogram (ECG) readings, seven different approaches are suggested for determining the stress levels of drivers in real-world scenarios. In order to categorise the driver's stress level, three separate CNNs are trained in advance. We looked at the ECG signal data from seven different algorithms—InceptionV3, Xception, Resnet101, Google Net, InceptionResnetV2, Densenet201, and Darknet53—to see how well the pre-trained algorithms predicted stress from DRIVERS. Xception provides algorithms with a high level of accuracy. Several machine learning algorithms have been created to attempt to alert drivers who are stressed, as they are a known accident-causing factor. One potential issue with these algorithms’ prediction accuracy is that they are trained using features that are developed manually instead of utilising exact calculations.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Gunamani Jena
AU  - P. Ramesh
AU  - Uma Kumari
AU  - V. Sravallika
AU  - K. S. P. Anjaneyulu
AU  - P. S. V. S. Srinivas
PY  - 2024
DA  - 2024/07/30
TI  - ECG-Based Driver's Pressure Location Utilizing Profound Exchange Learning And Fluffy Rationale Approaches
BT  - Proceedings of the International Conference on Computational Innovations and Emerging Trends (ICCIET- 2024)
PB  - Atlantis Press
SP  - 1462
EP  - 1469
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-471-6_142
DO  - 10.2991/978-94-6463-471-6_142
ID  - Jena2024
ER  -