Extraction and Classification of Human Body Parameters for Gait Analysis

  • PDF / 3,910,383 Bytes
  • 19 Pages / 595.276 x 790.866 pts Page_size
  • 59 Downloads / 297 Views

DOWNLOAD

REPORT


Extraction and Classification of Human Body Parameters for Gait Analysis Alana de M. e Souza1

· Marcelo R. Stemmer1

Received: 17 January 2018 / Revised: 3 July 2018 / Accepted: 20 July 2018 © Brazilian Society for Automatics–SBA 2018

Abstract Human gait analysis is considered a new biometric tool for the ability to obtain metrics from the body at a distance. Biometric identifiers have properties that can technologically measure and analyze the characteristics of the human body and can be used as a form of identification and access control for security applications. Recognition through proper interpretation of gait parameters has become a relevant pattern classification problem. This work aims to develop an image processing system, with the use of the Microsoft Kinect sensor, which is capable to extract movement patterns for gait analysis and to present a comparative study of different pattern recognition methods for human identification. The image processing system, developed in C#, allowed the acquisition of three-dimensional data from several volunteers and made it possible to identify the human skeleton and automatically extract the kinetic and kinematic parameters of the body. For data analysis, different classification methods were compared. Among them, the algorithms that presented better performance were probabilistic neural networks, deep neural networks and k-nearest neighbors, with nearly 99% correct recognition rate. The obtained results demonstrate the efficiency of gait analysis as a biometric method. They also show the viability of gait parameter extraction using the Kinect sensor and the good performance of pattern recognition methods applied to the acquired gait kinetic and kinematic parameters. Keywords Gait analysis · Pattern recognition · Biometric identifiers · Kinect sensor

1 Introduction Gait analysis has evolved to be a very reliable way to identify people. The analysis of the walking movement requires the measurement and evaluation of biomechanical characteristics that are associated with various tasks (Batista and Pereira 2015). There has been significant progress on gait analysis, accompanied by the development of motion capture systems. Among the attractive aspects of this biometric approach is the ability to recognize targets at a distance and the low data processing time (Best and Begg 2006). The human and animal movement has been studied for a long time in history. The Greek philosopher Aristotle (384 B.C.–322 B.C.) published a text on the march of animals,

B

Alana de M. e Souza [email protected] Marcelo R. Stemmer [email protected]

1

Department of Automation and Systems, Federal University of Santa Catarina, Florianopolis, SC, Brazil

beside other studies that complemented the subject. His study discusses why men and birds, although being both bipedal, have an opposite curvature in the legs and introduce some of the basic geometric knowledge used for calculations and analyses (Rosenhahn et al. 2008; Winter 2009). Leonardo da Vinci (1452–1519), in the Renaissance