IMU Sensor-Based Hand Gesture Recognition for Human-Machine Interfaces.

Minwoo Kim,Yunho Jung,Jaechan Cho,Seongjoo Lee

doi:10.3390/s19183827

Minwoo Kim, Yunho Jung + Show 2 more

Open Access

https://doi.org/10.3390/s19183827

Copy DOI

Abstract

We propose an efficient hand gesture recognition (HGR) algorithm, which can cope with time-dependent data from an inertial measurement unit (IMU) sensor and support real-time learning for various human-machine interface (HMI) applications. Although the data extracted from IMU sensors are time-dependent, most existing HGR algorithms do not consider this characteristic, which results in the degradation of recognition performance. Because the dynamic time warping (DTW) technique considers the time-dependent characteristic of IMU sensor data, the recognition performance of DTW-based algorithms is better than that of others. However, the DTW technique requires a very complex learning algorithm, which makes it difficult to support real-time learning. To solve this issue, the proposed HGR algorithm is based on a restricted column energy (RCE) neural network, which has a very simple learning scheme in which neurons are activated when necessary. By replacing the metric calculation of the RCE neural network with DTW distance, the proposed algorithm exhibits superior recognition performance for time-dependent sensor data while supporting real-time learning. Our verification results on a field-programmable gate array (FPGA)-based test platform show that the proposed HGR algorithm can achieve a recognition accuracy of 98.6% and supports real-time learning and recognition at an operating frequency of 150 MHz.

Highlights

A human-machine interface (HMI) presents information to a user regarding the state of a process, accepts commands, and operates associated devices [1]
Recognition performance greatly depends on the quality of the selected templates. Because these algorithms require finding optimal templates from a prepared dataset, it is hard to perform real-time learning with high accuracy, which is one of the most important features for hand gesture recognition (HGR) because users want devices to learn their hand gestures immediately
We address the following question: how can we make a high-performance HGR algorithm that supports real-time learning? We propose an efficient HGR algorithm that uses a simple learning algorithm based on an restricted coulomb energy (RCE) neural network and employ the distance measurement method of dynamic time warping (DTW)

Summary

Introduction

A human-machine interface (HMI) presents information to a user regarding the state of a process, accepts commands, and operates associated devices [1]. HMIs designed for human convenience are aimed at allowing users to freely control devices via simple operations without requiring the user’s full attention [2]. Hand gesture recognition (HGR) is an essential feature of HMIs because it allows users to efficiently control devices with simple hand gestures. HGR can be categorized into vision-based gesture recognition (VGR) and sensor-based gesture recognition (SGR) [3]. VGR is a method of recognizing gestures using camera images, and various technologies have been proposed [4,5,6]. VGR accuracy degrades in light-sensitive application scenarios because camera images are affected by lighting conditions [7]. SGR methods have relatively few limitations because they use various sensors that are not affected

Objectives

Results

Conclusion