Cross Teager-Kaiser Energy Operator based Feature Extraction Method for Gait Recognition from Cumulative Foot Pressure Images

The human footprint is a convenient biometric to be used for recognition purposes as it is universal, easy to acquire and does not change much over time. The cumulative foot pressure image (CFPI) records the spatial and temporal changes of the ground reaction force over a single gait cycle, thus offering more information than an ordinary footprint. This enables us to distinguish between various human gait aspects like limb movement. In this paper a scheme is proposed for gait recognition, that exploits features extracted from the cumulative foot pressure images (CFPIs) collected from 88 subjects, using the Cross Teager-Kaiser Energy Operator (CTKEO). This scheme exhibits a consistency in performance across major classifiers, while delivering a maximum recognition accuracy of 97.3%.

[1]  Tomomasa Sato,et al.  Person Recognition Method Using Sequential Walking Footprints via Overlapped Foot Shape and Center-of-Pressure Trajectory , 2004 .

[2]  K. Abed-Meraim,et al.  Cross Psi(B)-energy operator-based signal detection. , 2008, The Journal of the Acoustical Society of America.

[3]  Patricia Scully,et al.  Temporal Pattern Recognition in Gait Activities Recorded With a Footprint Imaging Sensor System , 2016, IEEE Sensors Journal.

[4]  James F. Kaiser,et al.  Some useful properties of Teager's energy operators , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[5]  J. F. Kaiser,et al.  On a simple algorithm to calculate the 'energy' of a signal , 1990, International Conference on Acoustics, Speech, and Signal Processing.

[6]  Masao Masugi,et al.  Gait Classification of Young Adults, Elderly Non-Fallers, and Elderly Fallers Using Micro-Doppler Radar Signals: Simulation Study , 2017, IEEE Sensors Journal.

[7]  Karim Abed-Meraim,et al.  Cross B-energy operator-based signal detection a ) , 2008 .

[8]  Firat Hardalaç,et al.  Application of an adaptive neuro‐fuzzy inference system for classification of Behcet disease using the fast Fourier transform method , 2007, Expert Syst. J. Knowl. Eng..

[9]  Marimuthu Palaniswami,et al.  Support vector machines for automated gait classification , 2005, IEEE Transactions on Biomedical Engineering.

[10]  Jaakko Astola,et al.  Teager energy and the ambiguity function , 1999, IEEE Trans. Signal Process..

[11]  Robert B. Dunn,et al.  Detection of transient signals using the energy operator , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[12]  V. D. Ambeth Kumar,et al.  Footprint Recognition using Modified Sequential Haar Energy Transform (MSHET) , 2010 .

[13]  Saibal Dutta,et al.  An automated hierarchical gait pattern identification tool employing cross‐correlation‐based feature extraction and recurrent neural network based classification , 2009, Expert Syst. J. Knowl. Eng..

[14]  Tieniu Tan,et al.  Evaluation framework on translation-invariant representation for cumulative foot pressure image , 2011, 2011 18th IEEE International Conference on Image Processing.