Scientists Propose Gait-based Biometric Identification Method for the Old with Wearable Devices

    Human gait is a unique feature that could be used for robust identity  recognition. Gait-based identity recognition method combines several  advantages, such as high fraud-resistance, secure data collection, no  need for explicit user interaction, and continuous and long-distance  authentication. This combination makes gait a very suitable biometric  parameter for user verification when associated with wearable devices. 

    However,  intra-subject gait fluctuation in older adults is more significant than  in young people due to physical strength changes associated with aging.  As a result, gait-based identity recognition of older adults is more  challenging. 

    Prof. LI Ye and his colleagues Dr. SUN  Fangmin and Dr. ZANG Weilin at the Shenzhen Institutes of Advanced  Technology (SIAT) of the Chinese Academy of Sciences, in cooperation  with colleagues from the University of Calabria in Italy, have proposed a  gait-based identification method for elderly users.  

    The current work, published in Information Fusion, is based on the team’s a previous study which was published in IEEE Internet of Things Journal  in 2018 and extends gait identification to elderly users. In the  previous study, LI’s team has proposed a speed-adaptive gait cycle  segmentation method and an individualized matching threshold generation  method to improve the gait authentication of young adults when walking  at various speeds.  

    In the current report, the  researchers introduced a gait-based identity recognition method used for  the access control of elderly people-centered wearable healthcare  devices. A gait template synthesis algorithm was proposed to alleviate  the problem of intra-subject gait fluctuation in elderly older people.  To further improve the identity recognition rate, the researchers  designed an arbitration-based score-level fusion algorithm: two matching  algorithms are used to make preliminary decisions; if there are  inconsistencies in the preliminary decisions, the third matching  algorithm is used to provide the final decision. Such procedure is shown  to improve the recognition rate of older adults compared with the  existing histogram similarity based method. 

    The  feasibility of the proposed method was verified using a public dataset  containing acceleration signals from three IMUs (inertial measurement  units) worn by 64 older users ranging in age from 50 to 79 years which  was shared by Osaka University in 2011. The experimental results  obtained prove that the average recognition rate reached 96.7%,  indicating the proposed method was quite suitable for robust gait-based  identification of elderly users. 

Fig. (a) Placement of acceleration sensor nodes; (b) Impact of the number of templates in Temps on recognition rate; (c) and (d) Comparison of identity recognition of older adults using the proposed method (Image by Dr. SUN Fangmin )

CONTACT:

ZHANG Xiaomin

Email: xm.zhang@siat.ac.cn

Tel: 86-755-86585299

http://english.siat.cas.cn/Research2017/RP2017/201907/t20190701_212310.html

https://www.eurekalert.org/pub_releases/2019-06/caos-spg062819.php

https://phys.org/news/2019-07-scientists-gait-based-biometric-identification-method.html 

http://7thspace.com/headlines/886490/scientists_propose_gait_based_biometric_identification_method_for_the_old_with_wearable_devices.html