Inertial sensor fusion for gait recognition with symmetric positive definite Gaussian kernels analysis
- PDF / 1,587,599 Bytes
- 28 Pages / 439.37 x 666.142 pts Page_size
- 51 Downloads / 171 Views
Inertial sensor fusion for gait recognition with symmetric positive definite Gaussian kernels analysis Jessica Permatasari 1 & Tee Connie 1
& Thian Song Ong
1
Received: 2 December 2019 / Revised: 22 June 2020 / Accepted: 28 July 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
Wearable sensor-based gait recognition has received much interest because it is unobtrusive and is user friendly. Many research has been carried out in this area but conventional gait recognition methods are not free from drawbacks. In this paper, accelerometer and gyroscope signals representing gait movements are encoded using covariance matrices. The covariance matrices provide a compact and descriptive representation for the accelerometer and gyroscope signals. Nonsingular covariance matrices are inherently Symmetric Positive Define (SPD) matrices. Interpreting such SPD matrices as points on the Riemannian manifold leads to increased performance. However, direct geodesic distance calculation for the matrix manifold may yield a suboptimal result. The proposed method solves this issue by embedding the manifold valued points to a higher dimensional Reproducing Kernel Hilbert Space (RKHS) via Positive Definite Gaussian Kernel functions. Extensive experiments have been conducted on three challenging benchmark datasets and a self-collected dataset. Experiment results testify the performance of the proposed RKHS embedding approach. Keywords Covariance matrices . Extreme learning machines (ELMs) . Gaussian kernel . Riemannian manifold . Sensor fusion
* Tee Connie [email protected] Jessica Permatasari [email protected] Thian Song Ong [email protected]
1
Faculty of Information Science and Technology, Multimedia University, Jalan Ayer Keroh Lama, 75450 Melaka, Malaysia
Multimedia Tools and Applications
1 Introduction With the rapid development of microelectromechanical system (MEMS), mobile devices have been equipped with more sensors. Study in [27] has shown that human gait is unique and human identification based on gait recognition has been a growing interest ever since that. One of the emerging topics is the use of inertial measurement unit (IMU) sensors (e.g. accelerometer and gyroscope) embedded in the smartphone for gait analysis since it can capture individuals’ daily activities and gestures unobtrusively. Most of the previous studies investigate the use of IMU sensors for human activity recognition (HAR) [37, 39], biometric authentication [2, 18, 26], etc. The acceleration and angular rates measures by IMU when smartphone is carried during daily activities contain characteristic signal information that can be used to detect gait events. Gait parameters such as step length and number of steps can be estimated from the accelerometer [1, 9, 13], whereas, the gyroscope can help to estimate the sensor orientation [13]. Gait recognition using different sensor modalities has become a revolutionary technology for real-time and autonomous monitoring in various domains such as activity of dai
Data Loading...
