Gait identification using fractal analysis and support vector machine
- 156 Downloads
This paper presents the development of wearable sensing system that can be used to study the gait dynamics of human. A tester wearing sensing shoes participates in this study. Human gait information about standing, jumping and walking is obtained as prior probability based on the train movement model setup theory. For feature extraction of gait, five kinds of features are extracted from foot pressure signals, which are subsequently used for motion analysis. We employ support vector machine and fractal analysis for gait recognition and test the identification performance. Testing outcomes indicate an overall accuracy of 93.57% via radial basis function kernel function. These results demonstrate considerable potential in gait identification.
KeywordsGait identification Foot pressure signal Feature extraction Fractal analysis Support vector machine
Project is supported by Shanghai municipal commission of health and family planning key developing discipline (No. 4122015ZB0401); Natural Science Foundation of Hunan Province, China (No. 2018JJ2023); the Natural Science Foundation of Shanghai (CN) (Nos. 14ZR1429800, 15ZR1430000), Ministry of Education of the People’s Republic of China (No. EIA140412).
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Informed consent was obtained from all individual participants included in the study.
- Alqasemi R, Dubey R (2010) A 9-DoF wheelchair-mounted robotic arm system: design, brain-computer interfacing and testing. Adv Robot Manip, Control, pp 51–78Google Scholar
- Carlet M (1872a) Etude de la marche. Ann Nat Sci No. 15, Article 6Google Scholar
- Carlet M (1872b) Sur la locomotion humaine, Etude de la marche. Ann Sci Nat No. 5, 16Google Scholar
- Chiang HS, Sangaiah AK, Chen MY, Liu JY (2018) A novel artificial bee colony optimization algorithm with SVM for bio-inspired software-defined networking. Int J Parallel Program 8:1–19Google Scholar
- Han T, Zeng M, Zhang L, Sangaiah AK (2018) A channel-aware duty cycle optimization for node-to-node communications in the internet of medical things. Int J Parallel Program 7:1–16Google Scholar
- Hong L, Jain AK, Pankanti S (1999) Can multiple biometrics improve performance. In: IEEE workshop on automatic identification advanced technologies, pp 1–8Google Scholar
- Ives TE (2004) Aging and balance. Adv Dir Rehabil 13:12Google Scholar
- Lasrsen RK, Simonsen EB, Lynnerup N (2007) Gait analysis in forensic medicine, article no. 64910m, Video metrics IX, vol 6491Google Scholar
- Liikavainio T (2010) Biomechanics of gait and physical function in patients with knee osteoarthritis thigh muscle properties and joint loading assessment. Doctor dissertation, University of Eastern FinlandGoogle Scholar
- Moon SB, Ji Y-H, Jang H-Y, Hwang S-H, Shin D-B, Lee S-C, Han J-S, Han C-S, Lee YG, Jang SH, Park SB, Kim MJ (2017) Gait analysis of hemiplegic patients in ambulatory rehabilitation training using a wearable lower-limb robot: a pilot study. Int J Precis Eng Manuf 18(12):1773–1781CrossRefGoogle Scholar
- Murata A, Iwase H (1998) Chaotic analysis of body sway, proceedings of the 20th annual international conference of the IEEE engineering in medicine and biology society, vol 3, pp 1557–1560Google Scholar
- Sangaiah AK, Samuel OW, Li X, Abdel-Basset M, Wang H (2017) Towards an efficient risk assessment in software projects: fuzzy reinforcement paradigm. Comput Electr Eng 8:1–12Google Scholar
- Sazonov ES, Bumpus T, Zeigler S, Marocco S (2005) Classification of plantar pressure and heel acceleration patterns using neural networks. Int Jt Conf Neural Netw 5:3007–3010Google Scholar
- Sodhro AH, Luo Z, Sangaiah AK, Baik SW (2018) Mobile edge computing based QoS optimization in medical healthcare applications. Int J Inf Manag 9:45–56Google Scholar
- Su H, Yang M, Xu H (2008) A novel method of gait recognition based on kernel fisher discriminant analysis. In: 2008 IEEE international conference on systems, man and cybernetics, pp 830–835Google Scholar
- Sugimoto C, Tsuji M, Lopez G, Hosaka , Sasaki K, Hirota T, Tatsuta S (2006) Development of a behavior recognition system using wireless wearable information devices. In: 2006 1st international symposium on CD-ROM wireless pervasive computing (ISWPC), pp 1–5Google Scholar
- Suxia XING, Tianhua CHEN, Jingxian LI (2010) Image fusion based on regional energy and standard deviation. Signal Process Syst 1:739–743Google Scholar
- Vaughan CL, Davis BL, O’Connor JJ (1999) Dynamics of human gait. Kiboho Publisher, Cape TownGoogle Scholar
- Whittle MW (2007) Gait analysis: an introduction. Elsevier, AmsterdamGoogle Scholar
- Yang Y, Wang J, Yang Y, (2012) Exploiting rotation invariance with SVM classifier for micro-calcification detection. In: Biomedical imaging, pp 590–593Google Scholar
- Yang G, Tan W, Jin H, Zhao T, Tu L (2018) Review wearable sensing system for gait recognition. Clust Comput 1:1–9Google Scholar
- Yong F, Yunjian GE, Quanjun S (2001) A human identification method based on dynamic plantar pressure distribution. In: International conference on information and automation, pp 329–332Google Scholar
- Yuan G, Zhang M, Wang Z, Zhang J (2004) The distribution of foot pressure and its influence factors in Chinese people. Chin J Phys Med Rehabil 26(3):156–159Google Scholar
- Zoss A, Kazerooni H, Chu A (2005) On the mechanical design of the Berkeley lower extremity exoskeleton (BLEEX). In: International conference on intelligent robots and systems, pp 3465–3472Google Scholar