HuGaDB: Human Gait Database for Activity Recognition from Wearable Inertial Sensor Networks

Chereshnev, Roman; Kertész‐Farkas, Attila

doi:10.1007/978-3-319-73013-4_12

Cited by 67 publications

(53 citation statements)

References 22 publications

(17 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Human Gait Database (HuGaDB) [ 32 ] was selected as a corpus to meaningfully and repeatably demonstrate the performances of various compression methods. HuGaDb is a public dataset of six-axis IMU signals collected from six different body segments (right and left foot, right and left shank, right and left thigh) of 18 healthy subjects performing various movement activities (including walking, running, sitting, standing, and biking) sampled at 60 Hz ( Figure 2 ).…”

Section: Methodsmentioning

confidence: 99%

Lossless Compression of Human Movement IMU Signals

Chiasson

Shull

2020

Sensors

View full text Add to dashboard Cite

Real-time human movement inertial measurement unit (IMU) signals are central to many emerging medical and technological applications, yet few techniques have been proposed to process and represent this information modality in an efficient manner. In this paper, we explore methods for the lossless compression of human movement IMU data and compute compression ratios as compared with traditional representation formats on a public corpus of human movement IMU signals for walking, running, sitting, standing, and biking human movement activities. Delta coding was the highest performing compression method which compressed walking, running, and biking data by a factor of 10 and compressed sitting and standing data by a factor of 18 relative to the original CSV formats. Furthermore, delta encoding was shown to approach the a posteriori optimal linear compression level. All methods were implemented and released as open source C code using fixed point computation which can be integrated into a variety of computational platforms. These results could serve to inform and enable human movement data compression in a variety of emerging medical and technological applications.

show abstract

Section: Methodsmentioning

confidence: 99%

Lossless Compression of Human Movement IMU Signals

Chiasson

Shull

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Many gait databases are publicly accessible. These range from databases collected from healthy people [25], [26] to databases that contain gait data of patients affected with Parkinsons's disease [27], [28] or gait data collected after surgeries [29]. For this work, we have selected the HuGaDB because it provides human gait data in great detail compared to other published datasets [25].…”

Section: A Human Gait Databasementioning

confidence: 99%

“…In this work, we propose a novel TRNG based on human gait data that can be used for generating cryptographic keys for an Internet of You application. To that end, we will make use of a public dataset containing human gait information [25]. The contributions of this paper are innovative in several ways.…”

Section: Introductionmentioning

confidence: 99%

A True Random Number Generator Based on Gait Data for the Internet of You

et al. 2020

View full text Add to dashboard Cite

The Internet of Things (IoT) is more and more a reality, and every day the number of connected objects increases. The growth is practically exponential-there are currently about 8 billion and expected to reach 21 billion in 2025. The applications of these devices are very diverse and range from home automation, through traffic monitoring or pollution, to sensors to monitor our health or improve our performance. While the potential of their applications seems to be unlimited, the cyber-security of these devices and their communications is critical for a flourishing deployment. Random Number Generators (RNGs) are essential to many security tasks such as seeds for key-generation or nonces used in authentication protocols. Till now, True Random Number Generators (TRNGs) are mainly based on physical phenomena, but there is a new trend that uses signals from our body (e.g., electrocardiograms) as an entropy source. Inspired by the last wave, we propose a new TRNG based on gait data (six 3-axis gyroscopes and accelerometers sensors over the subjects). We test both the quality of the entropic source (NIST SP800-90B) and the quality of the random bits generated (ENT, DIEHARDER and NIST 800-22). From this in-depth analysis, we can conclude that: 1) the gait data is a good source of entropy for random bit generation; 2) our proposed TRNG outputs bits that behave like a random variable. All this confirms the feasibility and the excellent properties of the proposed generator. INDEX TERMS True random number generator, Internet of Things, gait data, entropy, randomness.

show abstract

“…The proposed method was implemented and evaluated in Matlab R2017b on a PC, and the data for training and testing the method was from two publicly available gait databases, HuGaDB [10] and MAREA [11]. HuGaDB gait database, collected by Chereshnev and Kertesz-Farkas, consists of many human activity recordings.…”

Section: B Experimental Setupmentioning

confidence: 99%

An artificial neural network framework for lower limb motion signal estimation with foot-mounted inertial sensors

Sun

Yang

2018

2018 IEEE 15th International Conference on Wearable and Implantable Body Sensor Networks (BSN)

View full text Add to dashboard Cite

This paper proposes a novel artificial neural network based method for real-time gait analysis with minimal number of Inertial Measurement Units (IMUs). Accurate lower limb attitude estimation has great potential for clinical gait diagnosis for orthopaedic patients and patients with neurological diseases. However, the use of multiple wearable sensors hinder the ubiquitous use of inertial sensors for detailed gait analysis. This paper proposes the use of two IMUs mounted on the shoes to estimate the IMU signals at the shin, thigh and waist for accurate attitude estimation of the lower limbs. By using the artificial neural network framework, the gait parameters, such as angle, velocity and displacements of the IMUs can be estimated. The experimental results have shown that the proposed method can accurately estimate the IMUs signals on the lower limbs based only on the IMU signals on the shoes, which demonstrates its potential for lower limb motion tracking and real-time gait analysis.

show abstract

HuGaDB: Human Gait Database for Activity Recognition from Wearable Inertial Sensor Networks

Cited by 67 publications

References 22 publications

Lossless Compression of Human Movement IMU Signals

Lossless Compression of Human Movement IMU Signals

A True Random Number Generator Based on Gait Data for the Internet of You

An artificial neural network framework for lower limb motion signal estimation with foot-mounted inertial sensors

Contact Info

Product

Resources

About