Identity Recognition by Walking Outdoors Using Multimodal Sensor Insoles

Abstract

Recently, gait attracts attention as a practical biometric for devices that naturally possess walking pattern sensing. In the present study, we explored the feasibility of using a multimodal smart insole for identity recognition. We used sensor insoles designed and implemented by us to collect kinetic and kinematic data from 59 participants that walked outdoors. Then, we evaluated the performance of four neural network architectures, which are a baseline convolutional neural network (CNN), a CNN with a multi-stage feature extractor, a CNN with an extreme learning machine classifier using sensor-level fusion and CNN with extreme learning machine classifier using feature-level fusion. The networks were trained with segmented insole data using 0%, 50%, and 70% segmentation overlap, respectively. For 70% segmentation overlap and both-side data, we obtained mean accuracies of 72.8% ±0.038, 80.9% ±0.036, 80.1% ±0.021 and 93.3% ±0.009, for the four networks, respectively. The results suggest that multimodal sensor-enabled footwear could serve biometric purposes in the next generation of body sensor networks.

[Retracted] Multimodal Sensor Motion Intention Recognition Based on Three‐Dimensional Convolutional Neural Network Algorithm

“…CNN (2D-CNN) performs well in the recognition field. An important reason is that its convolution operation and the convolved image are two-dimensional, so 2D-CNN can effectively extract the spatial features of the image [26][27][28]. However, when dealing with video classification tasks such as target recognition, 2D-CNN can only extract the features of each frame of the video independently.…”

Plantar Biometrics for Edge Computing

“…Describing the concept of a Body Sensory Network (BSN), Ivanov et al [12] cite an expectation for growing need of biometric footwear to help facilitate this. By using a segmented design, they propose a two-part system, with one part being fastened to the shoe sole and transmitting the data to a central processing system.…”

“…To provide a fair comparison between the introduced biometric system and the state of the art, the most similar system(s) will be used. These are the Raspberry Pi 2B-based system by Yeh et al [19], the custom board of Ivanov et al [12] and Huang et al's microcontroller-based proposal. To quantify the different aspects of the three systems, table 2 presents a number of differences between the three.…”

“…A less common area of biometrics is the utilisation of foot biometrics. The plantar pressure, gait pattern and other foot related biomarkers adhere the main properties of biometrics, namely they are characteristics that (almost) every user has, there is a uniqueness in these characteristics [12,19], they are quantifiable and they are obtained without the user's notice to avoid their influence. The biometric systems, which are designed around foot biomarkers, can be classified into three main classes: sole-based, mat-based and photography-based.…”

Development of a Self-Powered Piezo-Resistive Smart Insole Equipped with Low-Power BLE Connectivity for Remote Gait Monitoring

“…Recent advances in the biomedical field have led to the development of new advanced wearable devices, including those intended for gait analysis [ 15 , 16 ]. The most widespread applications regard the monitoring of diabetes ulcers [ 17 ], the evaluation of sports performance [ 18 ], workers’ conditions [ 19 ], and biometric recognition [ 20 ]. The assessment of plantar pressure distribution is essential to understand the lower limbs’ functioning, design footwear, and prevent injuries [ 21 ].…”