Gait Feature Extraction Research Articles

Gait Recognition of Acceleration Sensor for Smart Phone Based on Multiple Classifier Fusion

Gait, as a kind of biological feature, has a profound value in personnel identification. This paper analyzes gait characteristics based on acceleration sensors of smart phones and proposes a new gait recognition method. First, in view of the existing methods in the process of extraction of gait features, a large number of redundant calculations, cycle detection error, and the phase deviation issue during the week put forward the Shape Context (SC) and Linear Time Normalized (LTN) combining SCLTN calibration method of gait cycle sequence matching, to represent the whole extract typical gait cycle gait. In view of the existing extracted gait features are still some conventional features; the velocity change of relatively uniform acceleration and the change of acceleration per unit time are proposed as new features. Secondly, combining new features with traditional features to form a new feature is set for training alternative feature set, from which the training time and recognition effect of multiple classifiers are screened. Finally, a new multiclassifier fusion method, Multiple Scale Voting (MSV), is proposed to fuse the results of Multiple classifiers to obtain the final classification results. In order to verify the performance of the proposed method, gait data of 32 testers are collected. The final experimental results show that the new feature has good separability, and the recognition rate of fusion feature set after MSV algorithm is 98.42%.

Experimental study on the evading behavior of individual pedestrians when confronting with an obstacle in a corridor

In this paper, controlled experiments have been conducted to make deep analysis on the obstacle evading behavior of individual pedestrians affected by one obstacle. Results of Fourier transform show that with the increase of obstacle width, the frequency and amplitude of body sway would barely be affected while the lateral deviation of walking direction would largely increase. On the one hand, the relation among the extracted gait features including body sway, stride length, frequency and speed has been illustrated. On the other hand, the walking direction can be featured by three critical evading points where apparent change of walking direction could be observed. Gaussian function has been used to fit the walking direction, thus allowing to estimate the three critical points and examine their variation with the increase of obstacle size. Furthermore, the direct–indirect evading and left–right turning preference as well as the possible psychological motivations behind have been analyzed. It is indicated that direct-evading pedestrians have a higher walking efficiency and right-turning pedestrians have a stronger tendency to behave ‘direct’. Results of this paper are expected to provide practical evidence for the modeling of pedestrian dynamics affected by obstacles.

Human Gait Feature Extraction based-on Silhouette and Center of Mass

When someone walks, there is a repetitive movement or coordinated cycle that forms a gait. Gait is different, unique and difficult to imitate. This characteristic makes gait one of the biometrics to find out one's identity. Gait analysis is needed in the development of biometric technology, such as in the field of security surveillance and the health sector to monitor gait abnormalities. The center of mass is the unique point of every object that has a role in the study of humans walking. Each person has a different center of mass. In this research, through a series of processes in image processing such as video acquisition, segmentation, silhouette formation, and feature extraction, the center of mass of the human body can be identified using a webcam with the resolution of 640 x 480 pixels and the frame rate of 30 frames/second. The results obtained from this research were gait frames of 510 frames from 17 pedestrian videos. Segmentation process using background subtraction separates the pedestrian object image from the background. Silhouette gait was produced from a series of image enhancement processes to eliminate noise that interferes the image quality. Based on the silhouette, feature extraction provides the center of mass to distinguish each individual's gait. The sequence of center of mass can be further processed for characterizing human gait cycle for various purposes.

Gait Study of Parkinson's Disease Subjects Using Haptic Cues with A Motorized Walker.

Gait abnormalities are one of the distinguishing symptoms of patients with Parkinson’s disease (PD) that contribute to fall risk. Our study compares the gait parameters of people with PD when they walk through a predefined course under different haptic speed cue conditions (1) without assistance, (2) pushing a conventional rolling walker, and (3) holding onto a self-navigating motorized walker under different speed cues. Six people with PD were recruited at the New York Institute of Technology College of Osteopathic Medicine to participate in this study. Spatial posture and gait data of the test subjects were collected via a VICON motion capture system. We developed a framework to process and extract gait features and applied statistical analysis on these features to examine the significance of the findings. The results showed that the motorized walker providing a robust haptic cue significantly improved gait symmetry of PD subjects. Specifically, the asymmetry index of the gait cycle time was reduced from 6.7% when walking without assistance to 0.56% and below when using a walker. Furthermore, the double support time of a gait cycle was reduced by 4.88% compared to walking without assistance.

Gait recognition method of temporal–spatial HOG features in critical separation of Fourier correction points

To solve the problem of low accuracy of gait recognition algorithm in video surveillance, a multi-shot gait recognition algorithm based on two-sided Fourier correction motion point estimation is proposed, which is based on temporal–spatial HOG feature template matching. First, in view of the features of video surveillance, the image point estimation reconfiguration algorithm is used, and a gait feature extraction algorithm is designed for the unmarked class separation temporal–spatial HOG feature template matching, which realizes the automatic recognition and extraction of the gait. Secondly, aiming at extracting the difference of feature motion angle under different lens and different angle of view, the angle correction is realized based on two-sided Fourier series, and the search algorithm of adaptive sequential forward floating selection is designed for high-dimensional feature space. Finally, the simulation results on Southampton test library show that the correct gait classification rate of the proposed algorithm can reach 96.3%, and the gait motion angle can be basically consistent in different gait cycles after angle correction.

Individualized Gait Pattern Generation for Sharing Lower Limb Exoskeleton Robot

The development of sharing technology makes it possible for expensive lower limb exoskeleton robots to be extensively employed. However, due to the uniqueness of gait pattern, it is challenging for lower limb exoskeleton robot to adapt to different wearers’ gait patterns. Studies have shown that the gait pattern is affected by many physical factors. This paper proposes an individualized gait pattern generation (IGPG) method for sharing lower limb exoskeleton (SLEX) robot. First, the gait sequences are parameterized to extract gait features. Then, the Gaussian process regression with automatic relevance determination is used to establish the mapping relationships between the body parameters and the gait features, and the weights of each body parameters on gait pattern are also given. The gait features of an unknown subject can be predicted based on the training set. Finally, the individualized gait pattern is reconstructed by autoencoder neural network and scaling process based on predicted gait features. The experimental results show that the gait pattern predicted by IGPG is very similar to the subject’s actual trajectory and has been successfully applied on the SLEX robot. With the help of sharing technology, the training set will be increased, and the prediction accuracy of individualized gait pattern will also be improved. Note to Practitioners —The main purpose of this paper is to solve the gait pattern mismatch problem when different people wear an lower limb exoskeleton robot. The gait patterns are different for each individual, and the main gait-related factors include body parameters and walking speed (WS). Therefore, the suitable gait pattern for the wearer is predicted according to their body parameters and target WS in this paper. The detailed prediction process and a full analysis of experimental results are also given. Finally, the generated gait patterns are successfully verified on the lower limb exoskeleton robot.

Kinect-Based Human Gait Recognition Using Locally Linear Embedded and Support Vector Machine

Recognition of human gait could be performed effectively provided that significant gait features are well extracted along with effective recognition process. Thus, the gait features should be selected or optimized appropriately for optimal accuracy during recognition. Therefore, in this research, optimization of gait features for both oblique and frontal view are evaluated for recognition purpose using Locally Linear Embedded (LLE) along with multi-class Support Vector Machine (SVM). Firstly, dynamic gait features for one gait cycle are extracted from each subject’s walking gait that is acquired using Kinect sensor. Next, the extracted gait features were then optimized using LLE known as DG-LLE and further classified by multi-class SVM with Error Correcting Output Code (ECOC) algorithm. Further, to validate the effectiveness of LLE as optimization technique, the proposed method is then compared with another two gait features namely the original gait features known as DG and optimization using Principal Component Analysis labeled as DG-PCA. Results showed that the optimization based on DG-LLE outperformed the other two methods namely DG and DG-PCA for both oblique and frontal views. In addition, DG-LLE method contributed as the highest recognition rate for both frontal and oblique views. Results also confirmed that the accuracy rate for frontal view is higher specifically 98.33% as compared to oblique view with 94.67%.

The Research on Shape Context Based on Gait Sequence Image

A gait feature extraction method based on resampling shape context is proposed in this article. First, the moving target detection is carried out to obtain the target area of the human body. Second, the gait cycle is measured, and the contour points and the lower limb joints are selected as sampling points. Then, the different sampling points is placed in the polar coordinates of the origin, the number of sampling points in different cells is counted as the shape context; Finally, the feature vectors are constructed according to the shape context, and the minimum distance is used for classification and recognition. Simulation experiments based on resampling shape context are tested in CASIA gait Database A and Database B. The experimental results show that the method proposed in this article has a lower computational complexity and higher recognition rate when compared with the original shape context method, which can be used for gait recognition.

Optimal Foot Location for Placing Wearable IMU Sensors and Automatic Feature Extraction for Gait Analysis

Our aim is to maximize the interpretable information for gait analysis. To achieve this, it is important to find the optimal sensor placement and the parameters that influence the extraction of automatic gait features. We investigated the effect of different anatomical foot locations on inertial measurement unit (IMU) sensor output. We designed and developed an android app to collect real-time synchronous sensor output. We selected a set of five anatomical foot locations covering most of the foot regions to place wearable IMU sensors for data collection. Each participant performed a trial in a straight corridor comprising 25 strides of normal walking, a turn-around, and another 25 strides. We proposed an automatic gait features extraction method to analyze the data for stride number, distance, speed, length and period of stride, stance, and swing phases during walking. The highest accuracy for detecting stride number was in location 1 (first cuneiform) followed by location 5 (Achilles Tendon) and 4 (Talus). Location 1 was the closest to correlate estimate to the measured distance travelled. The accuracy of detecting number of strides on average is 95.47% from accelerometer data and 93.60% from gyroscope data and closest to the 60:40% split for average stance and swing for 15 subjects. To validate our results, we conducted trials using the Qualisys motion capture instrument and from our sensors concurrently. The average accuracy of the result is 97.77% with 95% confidence interval 0.767 for estimated and 99.01% with 95% confidence interval 0.266 for period.

Gait Recognition Based on Silhouettes Sequences and Neural Networks for Human Identification

Human gait recognition is moving ahead by the need for automated person identification and verification at a distance in many applications. In this paper our system presented silhouette videos and neural networks based system for human identification. For each sequence of silhouette images, an automated Region of Interest (ROI) algorithm applied to reduce dimensionality, extract gait features has been attempted with 3level 2Dimension Discrete Wavelet Transform (3L-2D-DWT), edges detection and gait cycle were used to extract relevant feature. Back propagations neural networks used as pattern recognition. The MATLAB software graphics user interface is designed to display result and simplify the use of the system. The developed system has been evaluated using (TUM-IITKGP) which contains three different type of walking categories and the results demonstrate that the proposed system achieves 98.8%, 95.87% and 88% for normal walk, walking with backpack and walking with hand in pocket category respectively of correct recognition. We concluded that the different side of view movement increase reliability of the key extracted feature and improve the neural network performance which opens a scope for further development.

Gait Recognition Based on Silhouettes Sequences and Neural Networks for Human Identification

Human gait recognition is moving ahead by the need for automated person identification and verification at a distance in many applications. In this paper our system presented silhouette videos and neural networks based system for human identification. For each sequence of silhouette images, an automated Region of Interest (ROI) algorithm applied to reduce dimensionality, extract gait features has been attempted with 3level 2Dimension Discrete Wavelet Transform (3L-2D-DWT), edges detection and gait cycle were used to extract relevant feature. Back propagations neural networks used as pattern recognition. The MATLAB software graphics user interface is designed to display result and simplify the use of the system. The developed system has been evaluated using (TUM-IITKGP) which contains three different type of walking categories and the results demonstrate that the proposed system achieves 98.8%, 95.87% and 88% for normal walk, walking with backpack and walking with hand in pocket category respectively of correct recognition. We concluded that the different side of view movement increase reliability of the key extracted feature and improve the neural network performance which opens a scope for further development.

View-invariant gait recognition method by three-dimensional convolutional neural network

Gait as an important biometric feature can identify a human at a long distance. View change is one of the most challenging factors for gait recognition. To address the cross view issues in gait recognition, we propose a view-invariant gait recognition method by three-dimensional (3-D) convolutional neural network. First, 3-D convolutional neural network (3DCNN) is introduced to learn view-invariant feature, which can capture the spatial information and temporal information simultaneously on normalized silhouette sequences. Second, a network training method based on cross-domain transfer learning is proposed to solve the problem of the limited gait training samples. We choose the C3D as the basic model, which is pretrained on the Sports-1M and then fine-tune C3D model to adapt gait recognition. In the recognition stage, we use the fine-tuned model to extract gait features and use Euclidean distance to measure the similarity of gait sequences. Sufficient experiments are carried out on the CASIA-B dataset and the experimental results demonstrate that our method outperforms many other methods.

Recognition of human gait in oblique and frontal views using Kinect

This study describes the recognition of human gait in the oblique and frontal views using novel gait features derived from the skeleton joints provided by Kinect. In D-joint, the skeleton joints were extracted directly from the Kinect, which generates the gait feature. On the other hand, H-joint distance is a feature of distance between the hip joint with other skeleton joints. Prior to the gait feature extraction, the skeleton joints provided by Kinect were pre-processed in order to standardize the size of the skeleton image as well as to detect the gait feature within a full gait cycle. To classify gait patterns according to its own group, a multi-layer perceptron was employed in the pattern recognition stage. Results show that a perfect recognition of human gait (100%) was attained for the frontal view using the feature of H-joint distance at the optimal multi-layer perceptron (20 hidden units).

Fusion of Random Walk and Discrete Fourier Spectrum Methods for Gait Recognition

Gait-based person identification suffers from the problem of different covariate factors such as clothing and carrying objects, which drastically reduce the recognition rate. Most existing methods capture dynamic and static information and remove the covariate factors without any systematic study. However, it has been reported in the literature that the head is one of the important features and the removal of the head from static information decreases the recognition rate. In our preliminary study, we developed a novel random walk (RW)-based gait extraction method that retains the head portion and removes certain static body parts to reduce the effect of covariate factors. The RW-based method is a novel gait feature extraction method and should be exploited more for its discriminative power to separate different body parts efficiently. However, the dynamic part is also significant in gait information, which is not very effectively represented in the RW-based gait extraction method. Therefore, a discrete Fourier transform (DFT)-based frequency component of the gait is considered to represent the dynamic part of gait information. Furthermore, we propose a novel gait recognition algorithm that fuses dynamic and static information from DFTand RW-based representations. The proposed method systematically retains the discriminative static gait information along with the frequency attribute embedded as the dynamic gait information. Extensive experiments on the Chinese Academy of Sciences, Institute of Automation and the HumanID datasets have been carried out to demonstrate that the proposed fused gait features-based approach outperforms the existing methods, particularly when there are substantial appearance changes.

Walking Gait Step Length Asymmetry Induced by Handheld Device

The modeling and feature extraction of human gait motion are crucial in biomechanics studies, human localization, and robotics applications. Recent studies in pedestrian navigation aim at extracting gait features based on the data of low-cost sensors embedded in handheld devices, such as smartphones. The general assumption in pedestrian dead reckoning (PDR) strategy for navigation application is that the presence of a device in hand does not impact the gait symmetry and that all steps are identical. This hypothesis, which is used to estimate the traveled distance, is investigated in this paper with an experimental study. Ten healthy volunteers participated in motion lab tests with a 0.190 kg device in hand. Several walking trials with different device carrying modes and several gait speeds were performed. For a fixed walking speed, it is shown that the steps differ in their duration when holding a mass equivalent to a smartphone mass, which invalidates classical symmetry hypothesis in the PDR step length modeling. It is also shown that this hypothesis can lead to a 2.5% to 6.3% error on the PDR estimated traveled distance for the different walking trials.

Robust Gait Recognition by Integrating Inertial and RGBD Sensors.

Gait has been considered as a promising and unique biometric for person identification. Traditionally, gait data are collected using either color sensors, such as a CCD camera, depth sensors, such as a Microsoft Kinect, or inertial sensors, such as an accelerometer. However, a single type of sensors may only capture part of the dynamic gait features and make the gait recognition sensitive to complex covariate conditions, leading to fragile gait-based person identification systems. In this paper, we propose to combine all three types of sensors for gait data collection and gait recognition, which can be used for important identification applications, such as identity recognition to access a restricted building or area. We propose two new algorithms, namely EigenGait and TrajGait, to extract gait features from the inertial data and the RGBD (color and depth) data, respectively. Specifically, EigenGait extracts general gait dynamics from the accelerometer readings in the eigenspace and TrajGait extracts more detailed subdynamics by analyzing 3-D dense trajectories. Finally, both extracted features are fed into a supervised classifier for gait recognition and person identification. Experiments on 50 subjects, with comparisons to several other state-of-the-art gait-recognition approaches, show that the proposed approach can achieve higher recognition accuracy and robustness.

Multilinear Laplacian discriminant analysis for gait recognition

The extraction of discriminative features in presence of covariates for robust human gait recognition is a challenging task. The effect of covariate can be modelled as unknown feature contamination problem resulting in the conversion of useful or relevant feature into irrelevant one. This study presents new gait representation and recognition technique. The new technique represents gait features based on the Gabor function and discrete cosine transform of binary silhouettes. It is called as Gabor cosine feature, which represents binary gait video sequence as third order tensor. The discrimination capability of the extracted gait features has been enhanced using a new multilinear Laplacian discriminant analysis (MLDA). MLDA exploits benefit of Laplacian weighted scatter difference instead of simple scatter difference, generalised Rayleigh quotient as a class separability measure. The feasibility and performance of the proposed scheme has been evaluated using USF, CASIA, OU‐ISIR dataset. The experimental results show competitive performance in comparison with conventional gait recognition schemes.

Review on Vision-Based Gait Recognition: Representations, Classification Schemes and Datasets

Gait has unique advantage at a distance when other biometrics cannot be used since they are at too low resolution or obscured, as commonly observed in visual surveillance systems. This paper provides a survey of the technical advancements in vision-based gait recognition. A wide range of publications are discussed in this survey embracing different perspectives of the research in this area, including gait feature extraction, classification schemes and standard gait databases. There are two major groups of the state-of-the-art techniques in characterizing gait: Model-based and motion-free. The model-based approach obtains a set of body or motion parameters via human body or motion modeling. The model-free approach, on the other hand, derives a description of the motion without assuming any model. Each major category is further organized into several subcategories based on the nature of gait representation. In addition, some widely used classification schemes and benchmark databases for evaluating performance are also discussed.

Robust Arbitrary-View Gait Recognition Based on 3D Partial Similarity Matching.

Existing view-invariant gait recognition methods encounter difficulties due to limited number of available gait views and varying conditions during training. This paper proposes gait partial similarity matching that assumes a 3D object shares common view surfaces in significantly different views. Detecting such surfaces aids the extraction of gait features from multiple views; 3D parametric body models are morphed by pose and shape deformation from a template model using 2D gait silhouette as observation. The gait pose is estimated by a level set energy cost function from silhouettes including incomplete ones. Body shape deformation is achieved via Laplacian deformation energy function associated with inpainting gait silhouettes. Partial gait silhouettes in different views are extracted by gait partial region of interest elements selection and re-projected onto 2D space to construct partial gait energy images. A synthetic database with destination views and multi-linear subspace classifier fused with majority voting is used to achieve arbitrary view gait recognition that is robust to varying conditions. Experimental results on CMU, CASIA B, TUM-IITKGP, AVAMVG, and KY4D data sets show the efficacy of the propose method.

Multi-gait recognition using hypergraph partition

Gait recognition is a challenging problem in computer vision, especially when multi-persons walk together, called as multi-gait recognition. Multi-gait recognition includes two aspects: participant segmentation and participant recognition. In this paper, we propose to segment each participant by hypergraph partition and recognize each participant by multi-linear canonical correlation analysis algorithm (UMCCA). Firstly, raw pixel areas are obtained by grid, and each pixel area is taken as a hypergraph vertex. Then HOG-based detection and tracking technology is used to calculate the weight of each hyperedge. After segmentation, UMCCA is used to extract gait features. Finally, identity of multi-gait is recognized. The experimental results demonstrate that our proposed method achieves good performance on multi-gait dataset.