Motion Pattern Recognition Research Articles

Preliminary Study on Continuous Recognition of Elbow Flexion/Extension Using sEMG Signals for Bilateral Rehabilitation.

Surface electromyography (sEMG) signals are closely related to the activation of human muscles and the motion of the human body, which can be used to estimate the dynamics of human limbs in the rehabilitation field. They also have the potential to be used in the application of bilateral rehabilitation, where hemiplegic patients can train their affected limbs following the motion of unaffected limbs via some rehabilitation devices. Traditional methods to process the sEMG focused on motion pattern recognition, namely, discrete patterns, which are not satisfactory for use in bilateral rehabilitation. In order to overcome this problem, in this paper, we built a relationship between sEMG signals and human motion in elbow flexion and extension on the sagittal plane. During the conducted experiments, four participants were required to perform elbow flexion and extension on the sagittal plane smoothly with only an inertia sensor in their hands, where forearm dynamics were not considered. In these circumstances, sEMG signals were weak compared to those with heavy loads or high acceleration. The contrastive experimental results show that continuous motion can also be obtained within an acceptable precision range.

Bispectrum-based sEMG multi-domain joint feature extraction for upper limb motion classification

sEMG based motion pattern recognition is the focus in the rehabilitation medical engineering area. In order to get more information to characterize the sEMG signal of different upper limb motions, the signal data acquisition program is designed and the non-Gaussian characteristic of the sEMG signal is analyzed, the result shows that the sEMG signal collected is non-Gaussian signal. Bispectrum as a third-order statistics contains non-Gaussian information and the integral of bispectrum slice is extracted as feature. After that, the PCA method is adopted to reduce the bispectrum feature dimension. Then, the integral of bispectrum slice after PCA and the integral value of sEMG are combined as a multi-domain joint feature called BisIE. Finally, the experiment is executed to validate the effectiveness of the feature extraction method proposed by the SVM classifier compared with power spectrum-based multi-domain joint feature MMIE. The average classification accuracy of BisIE is about 97% and that of MMIE is about 93%. Besides, for the same subject, the classification accuracy of BisIE is higher than that of MMIE. The result shows that the proposed feature BisIE is effective in promoting sEMG-based upper limb motion recognition accuracy.

On the role of spatial phase and phase correlation in vision, illusion, and cognition.

Numerous findings indicate that spatial phase bears an important cognitive information. Distortion of phase affects topology of edge structures and makes images unrecognizable. In turn, appropriately phase-structured patterns give rise to various illusions of virtual image content and apparent motion. Despite a large body of phenomenological evidence not much is known yet about the role of phase information in neural mechanisms of visual perception and cognition. Here, we are concerned with analysis of the role of spatial phase in computational and biological vision, emergence of visual illusions and pattern recognition. We hypothesize that fundamental importance of phase information for invariant retrieval of structural image features and motion detection promoted development of phase-based mechanisms of neural image processing in course of evolution of biological vision. Using an extension of Fourier phase correlation technique, we show that the core functions of visual system such as motion detection and pattern recognition can be facilitated by the same basic mechanism. Our analysis suggests that emergence of visual illusions can be attributed to presence of coherently phase-shifted repetitive patterns as well as the effects of acuity compensation by saccadic eye movements. We speculate that biological vision relies on perceptual mechanisms effectively similar to phase correlation, and predict neural features of visual pattern (dis)similarity that can be used for experimental validation of our hypothesis of “cognition by phase correlation.”

離床行動予測を目的としたベッド上での動作パターン識別

離床行動予測を目的としたベッド上での動作パターン識別

Development of electroencephalographic pattern classifiers for real and imaginary thumb and index finger movements of one hand.

This study aimed to find effective approaches to electroencephalographic (EEG) signal analysis and resolve problems of real and imaginary finger movement pattern recognition and categorization for one hand. Eight right-handed subjects (mean age 32.8 [SD=3.3] years) participated in the study, and activity from sensorimotor zones (central and contralateral to the movements/imagery) was recorded for EEG data analysis. In our study, we explored the decoding accuracy of EEG signals using real and imagined finger (thumb/index of one hand) movements using artificial neural network (ANN) and support vector machine (SVM) algorithms for future brain-computer interface (BCI) applications. The decoding accuracy of the SVM based on a Gaussian radial basis function linearly increased with each trial accumulation (mean: 45%, max: 62% with 20 trial summarizations), and the decoding accuracy of the ANN was higher when single-trial discrimination was applied (mean: 38%, max: 42%). The chosen approaches of EEG signal discrimination demonstrated differential sensitivity to data accumulation. Additionally, the time responses varied across subjects and inside sessions but did not influence the discrimination accuracy of the algorithms. This work supports the feasibility of the approach, which is presumed suitable for one-hand finger movement (real and imaginary) decoding. These results could be applied in the elaboration of multiclass BCI systems.

A novel STFT-ranking feature of multi-channel EMG for motion pattern recognition

Electromyography (EMG) is widely applied for neural engineering. For motion pattern recognition, many features of multi-channel EMG signals were investigated, but the relationships between muscles were not considered. In this study, a novel STFT-ranking feature based on short-time Fourier transform (STFT) is proposed. The novelty of STFT-ranking features is considering and covering the relationship information between EMG signals and multiple muscles in a motion pattern. With an exoskeleton robot arm, two series of motion patterns corresponding to the shoulder and elbow in the sagittal plane were investigated. EMG signals from six muscles were acquired in arm motion patterns when participants worn the robot arm. Four types of feature combinations, including seven conventional features, were compared with the STFT-ranking feature. The principal component analysis (PCA) and support vector machine (SVM) were used to build the motion recognition model. With the STFT-ranking feature, the recognition performance (93.9%) is superior to the conventional features (33.3–90.8%). The recognition variation is smaller (SD=4.3%) than the other features tested (SD=5.9–13.8%). These achievements will contribute to the advancement of control method of exoskeleton robots or power orthoses based on multi-channel EMG signals in the future. Based on the principle of STFT-ranking feature, the method also has potential for other multi-channel signal applications, such as electroencephalography (EEG) signal processing, speech recognition, and acoustic analysis.

Full-body movement pattern recognition in climbing*

The aim of this study was to propose a method for full-body movement pattern recognition in climbing, by computing the 3D unitary vector of the four limbs and pelvis during performance. One climber with an intermediate skill level traversed two easy routes of similar rates of difficulty (5c difficulty on French scale), 10m in height under top-rope conditions. The first route was simply designed to allow horizontal edge-hold grasping, while the second route was designed with more complexity to allow both horizontal and vertical edge-hold grasping. Five inertial measurement units (IMUs) were attached to the pelvis, both feet and forearms to analyse the 3D unitary vector of each limb and pelvis. Cluster analysis was performed to detect the number of clusters that emerged from coordination of the four limbs and pelvis during climbing performance. Analysis revealed 22 clusters with 11 clusters unique across the two routes. Six clusters were unique to the simple hold design route and five clusters emerged only in the complex hold design route. We conclude that clustering supported identification of full-body orientations during traversal, representing a level of analysis that can provide useful information for performance monitoring in climbing.

High Dynamic-Range Motion Imaging Based on Linearized Doppler Radar Sensor

Miniaturized Doppler radar sensor (DRS) for noncontact motion detection is a hot topic in the microwave community. Previously, small-scale physiological signals such as human respiration and heartbeat rates are the primary interest of study. In this paper, we propose a comprehensive approach that can be used to improve the demodulation linearity of microwave DRSs, such that detailed time-domain motion information ranging from micro-scale to large scale can be accurately reconstructed. Experiments show that based on a digital-IF receiver architecture, dynamic dc offset tracking, and the extended differentiate and cross-multiply arctangent algorithm, the displacement and velocity of both micrometer-scale vibration of a tuning fork and meter-scale human walking can be accurately recovered. Our work confirms that substantial time-domain motion information is carried by the signals backscattered from moving objects. Retrieval of such information using DRSs can be potentially used in a wide range of healthcare and biomedical applications, such as motion pattern recognition and bio-signal measurements.

Pattern Recognition of Movements in Wushu Based on Image Processing Technology

In this paper, the genetic algorithm for the key point of image capture is improved, and the joint programming of VC and Matlab is used to achieve data transmission, the FAB body identification system signal that collected with VC was processed by Matlab software, eventually Tai Chi chuan movement design computer system has got. In order to verify the validity and reliability of the platform, taking the movements development of Wu style Tai Chi chuan and 24 style Tai Chi chuan for example, this article has performed three-dimensional simulation design on Tai Chi movements. The time curve of the knee joint displacement and the average angle table of knee movement have been got by the computer design platform of Tai Chi movements, which provides digital technology support for the study of tai chi movements design.

Real-Time Physical Activity Recognition Using a Single Triaxial Accelerometer Based on HMM

This study focuses on physical activity recognition of elder people using a single triaxial accelerometer attached on waist. With accelerations data acquired continuously by a wearable wireless device, a motion pattern recognition method based on acceleration time series analysis of human activity states is proposed. We use Hidden Markov Model (HMM) to identify physical activity series. Experimental result shows that the activity recognition rate is 95.7% using this method, which can play an important role in the adjuvant therapy.

A comparison of upper-limb motion pattern recognition using EMG signals during dynamic and isometric muscle contractions

Multichannel electromyography (EMG) signals are one of the common methods used in human motion pattern recognition. In exoskeleton robot control, EMG signals are measured during dynamic or isometric muscle contractions. Various types of contraction can cause EMG signals to vary, affecting recognition performance. A motion pattern recognition model using EMG signals from either dynamic or isometric muscle contractions has not yet been fully investigated. In this study, a novel feature extraction method, using the short-time Fourier transform ranking (STFT-ranking) feature, was employed to determine multichannel EMG signals. The performance of the novel feature and conventional features for motion pattern recognition using EMG signals, which included time-domain and frequency-domain features, was compared during dynamic and isometric muscle contractions. Experiments were conducted using an exoskeleton robotic arm to aid users in generating EMG signals of designated motion patterns. Among the features tested, the STFT-ranking feature yielded an accuracy rate exceeding 90% when the EMG signals used in the training and validation feature data sets were of the same type of muscle contraction. After examining the STFT-ranking feature projected onto the PCA space, the STFT-ranking feature was determined to offer more satisfactory performance than the other features tested for motion pattern recognition, because the feature data it collected from various motion patterns were more separable. The experimental results also revealed that it is preferable that EMG signals from the same type of muscle contraction, whether dynamic or isometric, are consistently used in both the training and validation (control) phases. Inconsistent EMG signals in the training and validation phases yielded a negative effect on motion pattern recognition performance. The methodology developed in this study has potential applications in exoskeleton robot control and rehabilitation.

Video Stabilization, Camera Motion Pattern Recognition and Motion Tracking Using Spatiotemporal Regularity Flow

In this paper we propose a different approach based on a spatio-temporal feature called the Spatio Temporal Regularity Flow (SPREF) to stabilize unwanted camera motions in a video, recognize the camera motion patterns between consecutive frames and Group of Frames(GOF) and track the motion of an object in a video with the background subtracted. The method for stabilization based on Camera Motion uses the Translational Regularity flow vectors (TSPREF). In this method we fit the TSPREF vectors into parametric model to calculate the unstabilized global motion. An adaptive Gaussian smoothing method is used to smoothen the global motion followed by motion compensation to produce a stabilized sequence. Experimental results are provided and the stabilization achieved is validated using the qualitative measure Interframe Transform Fidelity (ITF).In camera motion pattern recognition we make use of TSPREF vectors to recognize the cognizant camera motion patterns. This is done for consecutive frames as well as Group of Frames(GOF)of different video sequences. In motion tracking we use the TSPREF vectors to track the moving object present in a video. The test videos taken have a still background with one or two moving objects. In all the cases we have the background subtracted from the moving object. 

Extracting Postural Synergies for Robotic Grasping

We address the problem of representing and encoding human hand motion data using nonlinear dimensionality reduction methods. We build our work on the notion of postural synergies being typically based on a linear embedding of the data. In addition to addressing the encoding of postural synergies using nonlinear methods, we relate our work to control strategies of combined reaching and grasping movements. We show the drawbacks of the (commonly made) causality assumption and propose methods that model the data as being generated from an inferred latent manifold to cope with the problem. Another important contribution is a thorough analysis of the parameters used in the employed dimensionality reduction techniques. Finally, we provide an experimental evaluation that shows how the proposed methods outperform the standard techniques, both in terms of recognition and generation of motion patterns.

Movement Pattern Recognition Ability of Malaysian Rhythmic Gymnastics Judges

Movement Pattern Recognition Ability of Malaysian Rhythmic Gymnastics Judges

Movement Pattern Recognition Ability Of Malaysian Rhythmic Gymnastics Judges

This study examined the significance of movement pattern recognition ability in the judging performance of 30 rhythmic gymnastics judges who were classified correspondingly as expert, non-expert and novice. Results indicated that the expert and non-expert participants were superior to the novice participants in their movement pattern recognition ability and their judging performance but there were no differences between the expert and non-expert participants. When the participants were regrouped according to their movement pattern recognition ability; low, average and high, the results showed that there were very significant differences in the judging performances of the three groups classified according to their movement pattern recognition ability. This suggests that movement pattern recognition ability influences the judging performance of a rhythmic gymnastics judge.

AR and WNN Hybrid Method for Pattern Recognition of SME Signals

Wavelet neural networks (WNN) combine the functions of time–frequency localization from the wavelet transform and of self-studying from the neural network, which make them particularly suitable for the classification of complex patterns. Based on auto-regressive (AR) model and WNN, pattern recognition of prosthesis movements was studied in this paper. Firstly, we present an Auto-regressive (AR) Model analysis of surface myoelectric (SME) signals recorded on the ulnar flexor carpi and extensor carpi region of the right hand in resting position. Then we chose BP neural network classifier, RBF neural network classifier and Wavelet Neural Network (WNN) classifier to research the correlation between SME signals and wristwork. Four types of prosthesis movements are recognized by extracting four-order AR coefficient and construct them as eigenvector into different Neural Network. The experiment results show that the proposed WNN method achieved a mean classification accuracy of 93.75% of four movements (hand open (HO), hand close (HC), forearm intorsion (FI) and forearm extorsion (FE)), which outperformed the method using BP Neural Network (with mean classification accuracy 86.25%) and the RBF Neural Network method (87.5%). It indicated that the AI and WNN hybrid method was feasible for pattern recognition of SME signals.

Movement Pattern Recognition Assisted Map Matching for Pedestrian/Wheelchair Navigation

Today's mobile technology features several sensors that when integrated can provide ubiquitous navigation assistance to pedestrians including wheelchair users. Common sensors found in most smartphones are Global Positioning System (GPS), accelerometer, and compass. In this paper, a user's movement pattern recognition algorithm to improve map matching efficiency and accuracy in pedestrian/wheelchair navigation systems/services is discussed. The algorithm integrates GPS positions, orientation data from compass, and movement states recognized from accelerometer data in a client/server architecture. The algorithm is tested in an Android mobile phone, and the results show that the proposed map matching algorithm is efficient and provides good accuracy.

Movement pattern recognition in basketball free-throw shooting

The purpose of the present study was to analyze the movement patterns of free-throw shooters in basketball at different skill levels. There were two points of interest. First, to explore what information can be drawn from the movement pattern and second, to examine the methodological possibilities of pattern analysis. To this end, several qualitative and quantitative methods were employed. The resulting data were converged in a triangulation. Using a special kind of ANN named Dynamically Controlled Networks (DyCoN), a ‘complex feature’ consisting of several isolated features (angle displacements and velocities of the articulations of the kinematic chain) was calculated. This ‘complex feature’ was displayed by a trajectory combining several neurons of the network, reflecting the devolution of the twelve angle measures over the time course of each shooting action. In further network analyses individual characteristics were detected, as well as movement phases. Throwing patterns were successfully classified and the stability and variability of the realized pattern were established. The movement patterns found were clearly individually shaped as well as formed by the skill level. The triangulation confirmed the individual movement organizations. Finally, a high stability of the network methods was documented.

Conceptual Imitation Learning in a Human-Robot Interaction Paradigm

In general, imitation is imprecisely used to address different levels of social learning from high-level knowledge transfer to low-level regeneration of motor commands. However, true imitation is based on abstraction and conceptualization. This article presents a model for conceptual imitation through interaction with the teacher to abstract spatio-temporal demonstrations based on their functional meaning. Abstraction, concept acquisition, and self-organization of proto-symbols are performed through an incremental and gradual learning algorithm. In this algorithm, Hidden Markov Models (HMMs) are used to abstract perceptually similar demonstrations. However, abstract (relational) concepts emerge as a collection of HMMs irregularly scattered in the perceptual space but showing the same functionality. Performance of the proposed algorithm is evaluated in two experimental scenarios. The first one is a human-robot interaction task of imitating signs produced by hand movements. The second one is a simulated interactive task of imitating whole body motion patterns of a humanoid model. Experimental results show efficiency of our model for concept extraction, proto-symbol emergence, motion pattern recognition, prediction, and generation.

Methods for Recognition and Classification of Human Motion Patterns – A Prerequisite for Intelligent Devices Assisting in Sports Activities

AbstractSensor and computing technologies provide people with information on their performance and load when doing sports. In order to automatically give advices on how to continue exercising and/or to adjust the sports equipment during the physical activity, intelligent devices are required. These devices rely on models for recognition and classification of patterns in the motion currently performed. Different methods and models, such as Neural Networks, Hidden Markov models or Support Vector Machines have proven to be applicable for this purpose. Pros and cons of the different approaches are discussed. Practical applications are presented and experiences reported.