Abstract
The recognition of gait pattern variation is of high importance to various industrial and commercial applications, including security, sport, virtual reality, gaming, robotics, medical rehabilitation, mental illness diagnosis, space exploration, and others. The purpose of this paper is to study the nature of gait variability in more detail, by identifying gait intervals responsible for gait pattern variations in individuals, as well as between individuals, using cognitive demanding tasks. This work uses deep learning methods for sensor fusion of 116 plastic optical fiber (POF) distributed sensors for gait recognition. The floor sensor system captures spatiotemporal samples due to varying ground reaction force (GRF) in multiples of up to 4 uninterrupted steps on a continuous 2×1 m area. We demonstrate classifications of gait signatures, achieving up to 100% F1-score with Convolutional Neural Networks (CNN), in the context of gait recognition of 21 subjects, with imposters and clients. Classifications under cognitive load, induced by 4 different dual tasks, manifested lower F1-scores. Layer-Wise Relevance Propagation (LRP) methods are employed to decompose a trained neural network prediction to relevant standard events in the gait cycle, by generating a “heat map” over the input used for classification. This allows valuable insight into which parts of the gait spatiotemporal signal have the heaviest influence on the gait classification and consequently, which gait events, such as heel strike or toe-off, are mostly affected by cognitive load.
Highlights
G AIT recognition has been intensively studied in recent years for the best achievable accuracy in distinguishing a certain target case in a plethora of applications, e.g. in healthcare, biometrics [1] and authentication for surveillance and forensics [2]
The deep convolutional neural networks (CNN) model is applied on the dataset in order to test the validity of the algorithms for identifying gait signatures
We have shown that floor sensors and gait under cognitive load can be used for subject’s identifications by capturing the changes in the individual’s unique gait signature due to the need to process additional cognitive information in performing additional (“dual”) tasks
Summary
G AIT recognition has been intensively studied in recent years for the best achievable accuracy in distinguishing a certain target case in a plethora of applications, e.g. in healthcare, biometrics [1] and authentication for surveillance and forensics [2]. Previous studies proposed cameras to record video sequence of the body motion [4], wearable sensors to acquire the limbs’ trajectories and body posture [5], or microphones to capture the sound of footsteps [3]. These methods generally perform well, the quality of data embodying the spatial and temporal aspects of gait is affected by clothing’s. The deep CNNs utility is based on the ability to model complex relationships between inputs and outputs and find patterns in divergent data with high background levels. Since gait is characterized by events occurring naturally in all healthy humans, a comparatively small number of weakly manifested features can be picked up to make a classification in biometrics or indicate a health condition by detecting spatiotemporal deviations from gait previously categorized as “normal” for a specific individual
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
