Wearable Inertial Sensor-Based Hand-Guiding Gestures Recognition Method Robust to Significant Changes in the Body-Alignment of Subject

Abstract

The accuracy of the wearable inertia-measurement-unit (IMU)-sensor-based gesture recognition may be significantly affected by undesired changes in the body-fixed frame and the sensor-fixed frame according to the change in the subject and the sensor attachment. In this study, we proposed a novel wearable IMU-sensor-based hand-guiding gesture recognition method robust to significant changes in the subject’s body alignment based on the floating body-fixed frame method and the bi-directional long short-term memory (bi-LSTM). Through comparative experimental studies with the other two methods, it was confirmed that aligning the sensor-fixed frame with the reference frame of the human body and updating the reference frame according to the change in the subject’s body-heading direction helped improve the generalization performance of the gesture recognition model. As a result, the proposed floating body-fixed frame method showed a 91.7% test accuracy, confirming that it was appropriate for gesture recognition under significant changes in the subject’s body alignment during gestures.

Bi-Directional Long Short-Term Memory-Based Gait Phase Recognition Method Robust to Directional Variations in Subject’s Gait Progression Using Wearable Inertial Sensor

“…As shown in Figure 3 (left), even if the subject performs the same action while changing pose, the same pattern of data is not always collected, and the features that affect phase recognition change, resulting in low overall recognition accuracy. Therefore, this study solves the aforementioned problem by creating the float body-fixed frame ( ), as proposed in previous our research [ 20 ].…”

“…This alignment ensures that the IMU sensor data remain consistent, even when the user changes posture or moves, as illustrated on the right of Figure 3 . The formation and detailed information regarding the are available in our previous research [ 20 ]. Algorithm 1.…”

“…In this study, we used bi-directional LSTM based on previous research showing that bi-directional LSTM is capable of recognizing not only the instantaneous mode of repetitive and regular motion but also the long-term dependency between modes [ 20 ], and the detailed model structure is shown in Figure 8 . The 22 gait data for each lower extremity segment consisting of orientation, acceleration, and angular rate collected through the FBF method and IMU are captured at 1 ms intervals through standardization and a sliding window overlapping method with a window of size.…”

“…In this study, to address the identified limitations, the floating body-fixed frame ( ) concept from previous research was adopted [ 20 ]. This approach ensures consistent IMU sensor data collection, irrespective of attachment variability or changes in the wearer’s posture.…”

Air-GR: An Over-the-Air Handwritten Character Recognition System Based on Coordinate Correction YOLOv5 Algorithm and LGR-CNN

“…Xiaoliang Zhang et al [ 11 ] who detected gestures by combining a wearable armband and a smart glove made from a customizable array of pressure sensors, achieved the recognition of 10 gestures. Haneul Jeon et al [ 12 ] proposed a novel gesture recognition method based on wearable IMU sensors and experimentally demonstrated that the method is appropriate for gesture recognition under significant changes in the subject’s body alignment during gestures. In addition, to improve the accuracy of wearable sensor-based gesture recognition algorithms, Haneul Jeon et al [ 13 ] proposed a DCGAN structure with a mode switcher for data enhancement of time-series sensor data.…”