Identification System Based on Resolution Adjusted 2D Spectrogram of Driver’s ECG for Intelligent Vehicle

Abstract

Recently, traditional vehicles are being developed into intelligent vehicles as information is exchanged among various devices inside and outside the vehicles. In the connected car environment, the need for vehicle security is growing due to vehicle hacking accidents and possible threats to human life. Driver identification technology using electrocardiogram (ECG) signals has been studied to address vehicle security issues and driver-specific services. Existing driver identification systems tried to address the issues using a multidimensional feature extraction method. However, there are remaining issues, including accuracy concerns, because the resolution was adjusted without considering the ECG’s P, QRS Complexes, and T waves feature when analyzing the time-frequency multidimensional features. In this paper, we propose a driver identification system using a 2D spectrogram. It identifies a section where the resolution is optimally adjusted using a spectrogram that can simultaneously analyze the time-frequency features of an ECG. The experimental results show that the proposed method improved the identification performance compared to the existing multidimensional feature extraction methods such as EEMD and MFCCs. Besides, with a 2D spectrogram of 1/4 image size, the recognition performance is maintained in a CNN network and the training time is significantly reduced.

Comparative Analysis of Machine Learning Algorithms With Advanced Feature Extraction for ECG Signal Classification

“…The y-axis displays the intensity values of the signal inside each interval, and the x-axis divides the ECG signal into discrete time intervals (bins). The bin's colour represents the number of 10 to Figure 14 for each class, evaluate the signal's symmetry or variability, and highlight any outliers or odd intensity patterns as done in the research study [34].…”

A Comprehensive Review: Analysis of Machine Learning, Deep Learning, and Large Language Model Techniques for Revolutionizing Driver Identification

“…The pre-processing begins with filtering to remove noise [1,5]. Data cleaning [30,31,32]. Feature scaling [33,32,5].…”

“…Therefore, while physiological sensors offer the potential for driver identification, carefully considering user comfort, signal quality, and privacy safeguards is necessary for their successful implementation. [23,30] proposed a driver identification system by employing ECG morphological features. [67] demonstrate that including driverrelated information like heart rate and driving pattern-related data can significantly improve the effectiveness of driver identification.…”

“…[36] have introduced a method to optimize driver detection using data extracted from GPS and in-car sensors. To improve identification accuracy, [30] suggest a driver recognition system that concurrently uses both electrocardiogram (ECG) and electromyogram (EMG) data in various driving states, accounting for the actual driving environment. Compared to utilizing either a single EMG or ECG signal alone, the results show that combining both EMG and ECG data increases accuracy.…”

“…[13] implemented a transformation of 1D accelerometer signals into a 2D format, which were subsequently inputted into a discriminative end-to-end deep learning model. [30] put forward a driver identification system that employs a 2D spectrogram. The proposed method involves converting a segmented ECG cycle into a spectrogram as part of the process.…”