Neural Computing and Applications

, Volume 26, Issue 7, pp 1621–1629 | Cite as

Use of the image and depth sensors of the Microsoft Kinect for the detection of gait disorders

  • Aleš ProcházkaEmail author
  • Oldřich Vyšata
  • Martin Vališ
  • Ondřej Ťupa
  • Martin Schätz
  • Vladimír Mařík
Original Article


This paper presents a novel method of gait recognition that uses the image and depth sensors of the Microsoft (MS) Kinect to track the skeleton of a moving body and allows for simple human–machine interaction. While video sequences acquired by complex camera systems enable very precise data analyses and motion detection, much simpler technical devices can be used to analyze video frames with sufficient accuracy in many cases. The experimental part of this paper is devoted to gait data acquisition from 18 individuals with Parkinson’s disease and 18 healthy age-matched controls via the proposed MS Kinect graphical user interface. The methods designed for video frame data processing include the selection of gait segments and data filtering for the estimation of chosen gait characteristics. The proposed computational algorithms for the processing of the matrices acquired by the image and depth sensors were then used for spatial modeling of the moving bodies and the estimation of selected gait features. Normalized mean stride lengths were evaluated for the individuals with Parkinson’s disease and those in the control group and were determined to be 0.38 and 0.53 m, respectively. These mean stride lengths were then used as features for classification. The achieved accuracy was >90 %, which suggests the potential of the use of the image and depth sensors of the MS Kinect for these applications. Further potential increases in classification accuracy via additional biosensors and body features are also discussed.


Image and depth sensors Gait disorders Motion features MS Kinect Video processing Parkinson’s disease 



Authors would like to thank all patients who signed the informed consent to participate in the project with all the procedures approved by the local ethics committee as stipulated by the Helsinki Declaration. All data were kindly provided by the Movement Disorders Center of the Faculty Hospital in Hradec Králové, Czech Republic.


  1. 1.
    Brscic D, Kanda T, Ikeda T, Miyashita T (2013) Person tracking in large public spaces using 3-D range sensors. IEEE Trans Hum Mach Syst 43(6):522–534CrossRefGoogle Scholar
  2. 2.
    Han J, Shao L, Xu D, Shotton J (2013) Enhanced computer vision with Microsoft Kinect sensor—a review. IEEE Trans Cybern 43(5):1318–1344CrossRefGoogle Scholar
  3. 3.
    Procházka A, Vyšata O, Schätz M, Ťupa O, Yadollahi M, Vališ M (2014) The MS Kinect image and depth sensors use for gait features detection. In: IEEE international conference on image processing (ICIP)Google Scholar
  4. 4.
    Tupa O (2014) Multi-dimensional data modelling and analysis using MS Kinect. Master thesis, Institute of Chemical Technology in PragueGoogle Scholar
  5. 5.
    Ringer M, Lasenby J (2002) Multiple hypothesis tracking for automatic optical motion capture. Lecture Notes in Computer Science 2350, Springer-Verlag Berlin, pp 524–536Google Scholar
  6. 6.
    Lasenby J, Stevenson A (2000) Using geometric algebra for optical motion capture. In: Bayro-Corrochano E, Sobcyzk G (eds) Applied Clifford algebras in computer science and engineering. Birkhauser, BostonGoogle Scholar
  7. 7.
    Procházka A, Kubíček M, Pavelka A (2006) Multicamera systems in the moving body recognition. In: IEEE 48th international symposium on ELMARGoogle Scholar
  8. 8.
    Fortino G, Giannantonio R, Gravina R, Kuryloski P, Jafari R (2013) Enabling effective programming and flexible management of efficient body sensor network applications. IEEE Trans Hum Mach Syst 43(1):115–133CrossRefGoogle Scholar
  9. 9.
    Camplani M, Mantecon T, Salgado L (2013) Depth-color fusion strategy for 3-D scene modeling with kinect. IEEE Trans Cybern 43(6):1560–1571CrossRefGoogle Scholar
  10. 10.
    Lu J, Tan YP (2013) Ordinary preserving manifold analysis for human age and head pose estimation. IEEE Trans Hum Mach Syst 43(2):249–258CrossRefGoogle Scholar
  11. 11.
    Yanik P, Manganelli J, Merino J, Threatt A, Brooks J, Green K, Walker I (2014) A gesture learning interface for simulated robot path shaping with a human teacher. IEEE Trans Hum Mach Syst 44(1):41–54CrossRefGoogle Scholar
  12. 12.
    Li Y (2012) Multi-scenario gesture recognition using kinect. In: 17th international conference on computer games, pp 126–130Google Scholar
  13. 13.
    Strbac M, Markovic M, Popovic DB (2012) Kinect in neurorehabilitation: computer vision system for real time hand and object detection and distance estimation. In: IEEE 11th symposium on neural network application in electrical engineering (NEUREL 2012), USA, pp 127–132Google Scholar
  14. 14.
    Milovanovic M, Minovic M, Starkevic D (2012) New gait recognition method using Kinect stick figure and CBIR. In: Proceedings of the telecommunications forum (TELFOR), pp 1323–1326Google Scholar
  15. 15.
    Shum H, Ho E, Jiang Y, Takagi S (2013) Real-time posture reconstruction for Microsoft Kinect. IEEE Trans Cybern 43(5):1357–1369CrossRefGoogle Scholar
  16. 16.
    Choudry M, Beach T, Callaghan J, Kulic D (2013) A stochastic framework for movement strategy identification and analysis. IEEE Trans Hum Mach Syst 43(3):314–327CrossRefGoogle Scholar
  17. 17.
    Zhang C, Xu J, Xi N, Jia Y, Li W (2012) Development of an omni-directional 3D camera for robot navigation. In: 2012 IEEE/ASME international conference on advanced intelligent mechatronics, USA, pp 262–267Google Scholar
  18. 18.
    Wang Y, Yang C, Wu X, Xu S, Li H (2012) Kinect based dynamic hand gesture recognition algorithm research. In: 4th international conference on intelligent human–machine systems and cybernetics, pp 274–279Google Scholar
  19. 19.
    Machida E, Cao M, Murao T, Hashimoto H (2012) Human motion tracking of mobile robot with Kinect 3D sensor. In: SICE annual conference, Akita University, Japan, pp 2207–2211Google Scholar
  20. 20.
    Clark RA, Pua YH, Fortin K, Ritchie C, Webster KE, Denehy L, Bryant AL (2012) Validity of the Microsoft Kinect for assessment of postural control. Gait Posture 36:372–377CrossRefGoogle Scholar
  21. 21.
    Gabel M, Renshaw E, Schuster A, Gilad-Bachrach R (2012) Full body gait analysis with Kinect. In: International conference of the IEEE engineering in medicine and biologyGoogle Scholar
  22. 22.
    Preis J, Kessel M, Werner M, Linnhoff-Popien C (2012) Gait recognition with Kinect. In: 1st international workshop on kinect in pervasive computing, pp P1–P4Google Scholar
  23. 23.
    Weber I, Koch J, Meskemper J, Friedl K, Heinrich K, Hartmann U (2012) Is the MS Kinect suitable for motion analysis? Biomed Eng 57(8):664Google Scholar
  24. 24.
    Lee S, Shin S (2014) Gait signal analysis with similarity measure. Sci World J 2014:1–8Google Scholar
  25. 25.
    Gelb D, Oliver E, Gilman S (2011) Diagnostic criteria for Parkinson disease. JAMA Neurol 56(1):33–39Google Scholar
  26. 26.
    Venkat I, De Wilde P (2011) Robust gait recognition by learning and exploiting sub-gait characteristics. Int J Comput Vis 91(1):7–23CrossRefzbMATHGoogle Scholar
  27. 27.
    Yogev G, Giladi N, Peretz C, Springer C, Simon ES, Hausdorff JM (2005) Dual tasking, gait rhythmicity, and Parkinson’s disease: which aspects of gait are attention demanding? Eur J Neurosci 22(5):1248–1256CrossRefGoogle Scholar
  28. 28.
    Zhang Z (2012) Accuracy and resolution of Kinect depth data for indoor mapping applications. IEEE Multimed 19(2):4–10CrossRefGoogle Scholar
  29. 29.
    Qin S, Zhu X, Yang Y (2014) Real-time hand gesture recognition from depth images using convex shape decomposition method. J Signal Process Syst 74:47–58CrossRefGoogle Scholar
  30. 30.
    Image Acquisition Toolbox: User's Guide (2013) The MathWorks, NatickGoogle Scholar
  31. 31.
    Khoshelham K, Elberink SO (2012) Accuracy and resolution of Kinect depth data for indoor mapping applications. Sensors 12:1437–1454CrossRefGoogle Scholar
  32. 32.
    Obdrzalek S, Kurillo G, Ofli F, Bajcsy R, Seto E, Jimison H, Pavel M (2012) Accuracy and robustness of Kinect pose estimation in the context of coaching of elderly population. In: 34th annual international conference of the IEEE EMBS, IEEE, USA, pp 1188–1193Google Scholar
  33. 33.
    Smisek J, Jancosek M, Pajdla T (2012) 3D with kinect. In: IEEE international conference on computer vision, IEEE, USA, pp 1154–1160Google Scholar
  34. 34.
    Essmaeel K, Gallo L, Damiani E, De Pietro G, Dipanda A (2012) Temporal denoising of Kinect depth data. In: Eighth international conference on signal image technology and internet based systems. IEEE Computer Society, IEEE, USA, pp 47–52Google Scholar
  35. 35.
    Fawcett T (2006) An introduction to ROC analysis. Pattern Recognit Lett 27:861–874CrossRefGoogle Scholar
  36. 36.
    BenAbdelkader C, Cutler R, Davis L (2002) Person identification using automatic height and stride estimation. In: Proceedings of the 16th international conference on pattern recognition, vol 4, pp 377–380Google Scholar
  37. 37.
    Munsell B, Temlyakov A, Qu C, Wang S, (2012) Person identification using full-body motion and anthropometric biometrics from Kinect videos. In: ECCV 2012 ARTEMIS workshopGoogle Scholar
  38. 38.
    Dockstader S, Bergkessel K, Tekalp A (2002) Feature extraction for the analysis of gait and human motion. In: Proceedings of the 16th international conference on pattern recognition, vol 1, pp 5–8Google Scholar
  39. 39.
    Vysata O, Schatz M, Valis M, Prochazka A, Simunek L (2013) Quantitative gait analysis in Parkinson’s disease using MS Kinect: entropy and fractal dimension. J Neurol Sci 333(Suppl. 1):e67CrossRefGoogle Scholar
  40. 40.
    Chen Y, Cho C, Lin S, Lai H, Lo Y, Chen S, Chang Y, Huang W, Chen C, Jaw F, Tsang S, Tsai S (2012) A vision-based regression model to evaluate Parkinsonian gait from monocular image sequences. Expert Syst Appl 39:520–526CrossRefGoogle Scholar
  41. 41.
    Muro-de-la-Herran A, Garcia-Zapirain B, Mendez-Zorrilla A (2014) Gait analysis methods: an overview of wearable and non-wearable systems, highlighting clinical applications. Sensors 14:3362–3394CrossRefGoogle Scholar
  42. 42.
    Galna B, Barry G, Jackson D, Mhiripiri D, Olivier P, Rochester L (2014) Accuracy of the Microsoft Kinect sensor for measuring movement in people with Parkinson’s disease. Gait Posture 39:1062–1068CrossRefGoogle Scholar

Copyright information

© The Natural Computing Applications Forum 2015

Authors and Affiliations

  • Aleš Procházka
    • 1
    • 3
    Email author
  • Oldřich Vyšata
    • 2
  • Martin Vališ
    • 2
  • Ondřej Ťupa
    • 1
  • Martin Schätz
    • 1
  • Vladimír Mařík
    • 3
  1. 1.Department of Computing and Control EngineeringInstitute of Chemical Technology in PraguePrague 6Czech Republic
  2. 2.Department of Neurology, Faculty of Medicine in Hradec KrálovéCharles University in PragueHradec KrálovéCzech Republic
  3. 3.Czech Institute of Informatics, Robotics and CyberneticsCzech Technical University in PraguePrague 6Czech Republic

Personalised recommendations