Key Postures Research Articles

A High‐Precision Dynamic Movement Recognition Algorithm Using Multimodal Biological Signals for Human–Machine Interaction

Accurate recognition of human dynamic movement is essential for seamless human–machine interaction (HMI) across various domains. However, most of the existing methods are single‐modal movement recognition, which has inherent limitations, such as limited feature representation and instability to noise, which will affect its practical performance. To address these limitations, this article proposes a novel fusion approach that can integrate two biological signals, including electromyography (EMG) and bioelectrical impedance (BI). The fusion method combines EMG for capturing dynamic movement features and BI for discerning key postures representing discrete points within dynamic movements. In this method, the identification of key postures and their temporal sequences provide a guiding framework for the selection and weighted correction of probability prediction matrices in EMG‐based dynamic recognition. To verify the effectiveness of the method, six dynamic upper limb movements and nine key postures are defined, and a Universal Robot that can follow movements is employed for experimental validation. Experimental results demonstrate that the recognition accuracy of the dynamic movement reaches 96.2%, representing an improvement of nearly 10% compared with single‐modal signal. This study illustrates the potential of multimodal fusion of EMG and BI in movement recognition, with broad prospects for application in HMI fields.

Simulation of entertainment interactive robots based on intelligent algorithms in personalized fitness training systems

In the current context of rapid development of information technology and AI, entertainment interactive robots can play an auxiliary role in fitness training. Intelligent technical services can provide athletes with timely training course information and gamified sports training. Various data will also be uploaded to technical platforms for analysis. This article studies the design of a personalized fitness training system for entertainment interactive robots based on intelligent algorithms. Firstly, this article analyzes the main technologies of photon integrated biosensors, establishes a linear photon integrated biosensor imaging model. Choose to use intelligent algorithms to optimize coverage for wireless sensor networks. This problem is transformed into a mathematical function model, and then the optimal solution is sought through algorithms, effectively reducing the energy consumption of the network and extending its lifespan. The system improves interaction with humans by collecting training action videos, automatically recognizes the types of human training actions, and automatically extracts key postures from training actions for comparison and analysis with standard postures. The practice and research of human–computer interaction have shown that in addition to the actual performance of social robots, the emotional interaction of social robots can also affect people's trust in their abilities.

Research on Positioning Tracking Mode of Sports Rehabilitation Training Based on Self-Powered Sensor Based on Particle Swarm Optimization Algorithm

The theme of today’s world is peace and development. A stable external environment has made people’s average life expectancy gradually increased, and the world is rapidly aging. Aging has brought many problems, such as the increase in the number of patients with limb dysfunction due to various diseases, which has gradually increased the demand for rehabilitation training. With traditional rehabilitation training methods, the training scenes are single and boring, and patients are prone to resisting. This paper designs and implements a real-time rehabilitation training guidance system based on self-powered sensors for the rehabilitation training needs of stroke patients. The system uses self-powered sensors to collect human motion information in real time, and compares it with the key posture sequences in the standard motion library to obtain corresponding matching results and guide patients to perform correct rehabilitation training. Using the rotation quaternion of 25 bone points in the patient’s rehabilitation exercise to calculate and update the rotation quaternion of the corresponding bone point of the character model, the function of the character model to follow the patient’s mirror motion is realized. This allows patients to control the completion of their rehabilitation movements without the need for medical staff to accompany them. And the stability of the system is optimized based on the particle swarm optimization algorithm. After traversal optimization, the current sensitivity coefficient of the model is reduced by about 75% compared with that before the correction, indicating that the current stability of the model obtained at this time has been improved to a certain extent. However, in the regression model of the self-powered sensor established by the particle swarm optimization algorithm, its parameters are reduced by about 82% compared with those before the correction, which shows that the current stability of the model has been greatly improved at this time, and the operating current of the receiving loop has been greatly improved.

Motion Recognition Based on Deep Learning and Human Joint Points

In order to solve the problem that the traditional feature extraction methods rely on manual design, the research method is changed from the traditional method to the deep learning method based on convolutional neural networks. The experimental results show that the larger average DTW occurs near the 55th calculation, that is, about the 275th frame of the video. In the 55th calculation, the joint angle with the largest DTW distance is the right knee joint. A multiscene action similarity analysis algorithm based on human joint points has been realized. In the fitness scene, by analyzing the joint angle through cosine similarity, the time of fitness key posture in the action sequence can be recognized. In the sports scene, through the similarity analysis of joint angle sequences by the DTW algorithm, we can get the similarity between people's actions in the sports video and the joint positions with large differences in some time intervals, and the real validity of the experiment is verified. The accuracy of motion recognition before and after the improvement is 95.2% and 97.1%, which is 0.19% higher than that before the improvement. The methods and results are widely used in the fields of sports recognition, movement specification, sports training, health management, and so on.

Crowd Stability Analysis Based on Pedestrian Abnormal Postures

Abnormal behaviors of pedestrians in crowd gathering public places are important factors affecting the stability of crowd flow. Pedestrian abnormal postures are important manifestation of abnormal behaviors, which often leads to local turbulence, disturbance and density-speed fluctuations. It is urgent to discover the disturbance mechanism of abnormal pedestrian posture on the stability of crowd flow. This study intends to establish machine vision, kinematics, dynamic models and crowd confluence dynamic models for typical abnormal pedestrian postures in public places. We mainly use computer vision related technology based to recognize abnormal postures of pedestrians in videos, constructs a network matrix of key posture nodes, and studys the kinematics characteristics of abnormal posture nodes. Considering the number of pedestrians and the characteristics of the architectural scenes, we design a workflow to select the appropriate macro or micro dynamic model to build the crowd flow model. To validate the propuesd model, case in Shanghai Hongqiao railway station is studied.

Judgment of Athlete Action Safety in Sports Competition Based on LSTM Recurrent Neural Network Algorithm

Athlete injury has always been an important factor that plagues sports. In order to reduce the probability of athletes’ sports injury and improve the judgment of athletes’ action safety, the inherent laws of sports actions are fully excavated, the development of action safety is promoted, and learners and instructors are caused to fully understand the safety of actions. This study uses the LSTM (long short-term memory) cyclic neural network algorithm to judge the safety of athletes in sports competitions. The experiment verifies the effectiveness of the LSTM cyclic neural network algorithm in basketball segmentation and recognition. Sports injury is one of the important factors affecting the performance of all sports, and the problem of athletes’ injury is worrying, so it is very necessary to effectively prevent potential sports injuries. Through the investigation of different professional athletes, the LSTM cyclic neural network algorithm is used for the whole process of extracting an independent motion action including continuous actions. It is used to distinguish key postures and nonkey postures in an action, and to judge the correctness of the action. Basketball skills here are mainly the movements of basic skills such as moving, passing the ball, dribbling, shooting, breaking with the ball, personal defense, grabbing the ball, stealing the ball, and grabbing the ball. The research results prove that the LSTM recurrent neural network algorithm has a good effect on the safety of athletes. For athletes, 41.9% of people can improve the safety of their movements by strengthening strength training.

Clothing Insulation Rate and Metabolic Rate Estimation for Individual Thermal Comfort Assessment in Real Life.

Satisfactory indoor thermal environments can improve working efficiencies of office staff. To build such satisfactory indoor microclimates, individual thermal comfort assessment is important, for which personal clothing insulation rate () and metabolic rate (M) need to be estimated dynamically. Therefore, this paper proposes a vision-based method. Specifically, a human tracking-by-detection framework is implemented to acquire each person’s clothing status (short-sleeved, long-sleeved), key posture (sitting, standing), and bounding box information simultaneously. The clothing status together with a key body points detector locate the person’s skin region and clothes region, allowing the measurement of skin temperature () and clothes temperature (), and realizing the calculation of from and . The key posture and the bounding box change across time can category the person’s activity intensity into a corresponding level, from which the M value is estimated. Moreover, we have collected a multi-person thermal dataset to evaluate the method. The tracking-by-detection framework achieves a mAP50 (Mean Average Precision) rate of 89.1% and a MOTA (Multiple Object Tracking Accuracy) rate of 99.5%. The estimation module gets an accuracy of 96.2% in locating skin and clothes. The M estimation module obtains a classification rate of 95.6% in categorizing activity level. All of these prove the usefulness of the proposed method in a multi-person scenario of real-life applications.

An Efficient Two-Stream Network for Isolated Sign Language Recognition Using Accumulative Video Motion

Sign language is the primary communication medium for persons with hearing impairments. This language depends mainly on hand articulations accompanied by nonmanual gestures. Recently, there has been a growing interest in sign language recognition. In this paper, we propose a trainable deep learning network for isolated sign language recognition, which can effectively capture the spatiotemporal information using a small number of signs’ frames. We propose a hierarchical sign learning module that comprises three networks: dynamic motion network (DMN), accumulative motion network (AMN), and sign recognition network (SRN). Additionally, we propose a technique to extract key postures for handling the variations in the sign samples performed by different signers. The DMN stream uses these key postures to learn the spatiotemporal information pertaining to the signs. We also propose a novel technique to represent the statical and dynamic information of sign gestures into a single frame. This approach preserves the spatial and temporal information of the sign by fusing the sign’s key postures in the forward and backward directions to generate an accumulative video motion frame. This frame was used as an input to the AMN stream, and the extracted features were fused with the DMN features to be fed into the SRN for the learning and classification of signs. The proposed approach is efficient for isolated sign language recognition, especially for recognizing static signs. We evaluated this approach on the KArSL-190 and KArSL-502 Arabic sign language datasets, and the obtained results on KArSL-190 outperformed other techniques by 15% in the signer-independent mode. Additionally, the proposed approach outperformed the state-of-the-art techniques on the Argentinian sign language dataset LSA64. The code is available at https://github.com/Hamzah-Luqman/SLR_AMN.

Sports Competition Assistant System Based on Fuzzy Big Data and Health Exercise Recognition Algorithm

When material desires are satisfied, people begin to pursue more and more spiritual levels. Health exercises have an excellent auxiliary effect on people’s flexibility and physical fitness, so more and more people choose health exercises. However, the movement of health exercises returns to Chengdu and affects the efficiency of physical training. Therefore, we have designed a sports competition assistance system based on vague big data and a health exercise recognition algorithm. First of all, in this article, the standard score comparison database is created by extending the standard action data. In addition, the system architecture is further given, and the key 3D data-based acquisition module design is given. In addition, the system architecture is further given, and the basic 3D data acquisition unit design is given. In this document, the depth characteristics filtered by the Fourier Pyramid are fused to the bone characteristics, and the merged data is sorted based on the support engine, thus designing the action recognition unit. A hidden Markov model (HMM) human action recognition algorithm based on pose selection is proposed. This method uses two affine propagation (AP) clustering algorithms to cluster the features, automatically select the key posture of each action, and correspond to the hidden state of the HMM. These hidden state labels are used to initialize the parameters of the HMM to train the model, and the trained model is used to implement action classification. The result shows that the design in the article has a more accurate recognition result, which provides a powerful tool for the referee to score. Using the Fourier Pyramid filtering method, through a large number of health exercises for comparison, the ability to judge the degree of standard health exercises is significantly improved, the efficiency is increased by 25%, and the accuracy rate is increased by 15%.

Human Pose Estimation Based on Multi-resolution Feature Parallel Network for Public Security

Smoking detection is an essential part of safety production management. With the wide application of artificial intelligence technology in all kinds of behavior monitoring applications, the technology of real-time monitoring smoking behavior in production areas based on video is essential. In order to carry out smoking detection, it is necessary to analyze the position of key points and posture of the human body in the input image. Due to the diversity of human pose and the complex background in general scene, the accuracy of human pose estimation is not high. To predict accurate human posture information in complex backgrounds, a deep learning network is needed to obtain the feature information of different scales in the input image. The human pose estimation method based on multi-resolution feature parallel network has two parts. The first is to reduce the loss of semantic information by hole convolution and deconvolution in the part of multi-scale feature fusion. The second is to connect different resolution feature maps in the output part to generate the high-quality heat map. To solve the problem of feature loss of previous serial models, more accurate human pose estimation data can be obtained. Experiments show that the accuracy of the proposed method on the coco test set is significantly higher than that of other advanced methods. Accurate human posture estimation results can be better applied to the field of smoking detection, and the smoking behavior can be detected by artificial intelligence, and the alarm will be automatically triggered when the smoking behavior is found.

Research on Balance Control of Freestyle Skiing Aerial Skills Based on Ant Colony Algorithm

Aiming at solving the problem of poor performance of airborne freestyle skiing balance, this paper presents the research of airborne freestyle skiing balance control based on ant colony algorithm. On the basis of defining the trajectory division of the airborne balance of freestyle skiing and the track of the center of gravity of the human body, a sensor is used to collect the data of the airborne balance of freestyle skiing, and the moving average, denoising, and normalizing processes are done. The training label of the ant colony algorithm is made by the analog signal matrix, the implementation foreground of the key posture frame of freestyle skiing is extracted, the disturbed area in the key posture frame is removed by the clustering algorithm, and the key posture area of freestyle skiing is obtained. The incremental clustering of the data of the key posture area of freestyle skiing is conducted, the incremental posture data mining model of freestyle skiing is established, the mining parameters are input into the mining model, and the incremental data mining is realized by the ant colony algorithm to complete the research on the control of the airborne balance of freestyle skiing. The results show that the proposed method has good reliability, good convergence, and strong response ability.

Human Posture Estimation and Sustainable Events Classification via Pseudo-2D Stick Model and K-ary Tree Hashing

This paper suggests that human pose estimation (HPE) and sustainable event classification (SEC) require an advanced human skeleton and context-aware features extraction approach along with machine learning classification methods to recognize daily events precisely. Over the last few decades, researchers have found new mechanisms to make HPE and SEC applicable in daily human life-log events such as sports, surveillance systems, human monitoring systems, and in the education sector. In this research article, we propose a novel HPE and SEC system for which we designed a pseudo-2D stick model. To extract full-body human silhouette features, we proposed various features such as energy, sine, distinct body parts movements, and a 3D Cartesian view of smoothing gradients features. Features extracted to represent human key posture points include rich 2D appearance, angular point, and multi-point autocorrelation. After the extraction of key points, we applied a hierarchical classification and optimization model via ray optimization and a K-ary tree hashing algorithm over a UCF50 dataset, an hmdb51 dataset, and an Olympic sports dataset. Human body key points detection accuracy for the UCF50 dataset was 80.9%, for the hmdb51 dataset it was 82.1%, and for the Olympic sports dataset it was 81.7%. Event classification for the UCF50 dataset was 90.48%, for the hmdb51 dataset it was 89.21%, and for the Olympic sports dataset it was 90.83%. These results indicate better performance for our approach compared to other state-of-the-art methods.

Causal actions enhance perception of continuous body movements

Our experience of motion depends not only on spatiotemporal features of stimuli, but also on our recognition of seemingly higher-level properties, as when we see an actor's body movements as goal-directed. Here, we examined how the perception of social causation in human actions guides the perceptual interpolation of motion in the observation of body movements. Natural human-object interactions were recorded for videos in which a person prepared to catch a ball thrown by another person. We manipulated the number of image frames between key postures to yield a short clip with different frame rates, and asked participants to judge whether the catcher's action showed smooth movements or sudden changes. In the causal condition, the catcher faced toward the ball and the thrower to preserve an intention-based causal relation between the ball's movement and the catcher's action in which the former causes the catcher's intention to act. In the non-causal condition, the catcher performed the same movements to raise their hands to catch a ball, except that they faced away from the ball, creating the impression of either a psychic reaction or coincidental non-goal-directed behavior, which makes movements of the ball appear to be an implausible cause of the catcher's intention to act. Across four experiments, we found that humans were more likely to judge the catcher's body movements to be continuous in the causal condition than in the non-causal condition. The effect was maintained as long as the intention-based causal relation was present, even when only part of the chain of causal events was observed. These findings indicate that intention-based cause-effect relations in human actions guide perceptual interpolation of body movements.

Pathological kinematic patterns of the tarsal complex in stage II adult-acquired flatfoot deformity.

The in vivo kinematic characteristics of the tarsal joints during gait stance phase were still unclear in adult-acquired foot deformity (AAFD). This study included seven healthy subjects (14 feet) and 12 stage II AAFD patients (14 feet). The 3D models of tarsal bones were reconstructed based on CT scan. Each subject took standard gait on the single fluoroscopy system. Continuous lateral fluoroscopic images were collected. The key postures during the stance phase were selected. The 2D-3D registration technique was applied to explore the spatial motions of the tarsal joints in 6 degrees of freedom (DOF). During the whole stance phase, the AAFD talo-navicular joint (TNJ) exhibited ROM of 13 ± 6° in the sagittal plane while the normal subjects showed ROM of 7 ± 3° (p = 0.004). In AAFD, the subtalar joint (STJ) demonstrated 19 ± 8° and 7 ± 3° of motion in coronal and horizontal plane, respectively, while the normal subjects showed 14 ± 4°(p = 0.031) and 11 ± 3° (p = 0.014) of motion, respectively. Additionally STJ of AAFD patients showed significantly less dorsiflexion during the weight acceptance and showed significantly less external rotation both during the weight acceptance and single limb support of stance phase. In conclusion, for stage II AAFD patients, the talonavicular joint and the subtalar joint showed hypermobility in dorsi/planterflexion and inversion/eversion, respectively, during the gait stance phase while the internal/external rotation of the subtalar joint was reduced. The current study improves our understanding of the pathological kinematics of the tarsal complex in AAFD patients. Notice should be taken about these tarsal joints mobility in AAFD during clinical practice. © 2017 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 37:477-482, 2019.

Development and evaluation of two posture-tracking user interfaces for occupational health care

The spinal column requires special care through exercises focused on muscle strengthening, flexibility, and mobility to minimize the risk of developing musculoskeletal disorders that may affect the quality of life. Guidelines for spinal column exercises are commonly presented through printed and multimedia guides accompanied with demonstrations performed by a physiotherapist, occupational health expert, or physical fitness trainer. However, existing guides lack interaction and oral explanations may not always be clear to the user, leading to decreased engagement and motivation to start, continue, or complete an exercise program. In this article, we present two interactive and engaging posture-tracking user interfaces intended to promote proper spinal column exercise form. One user interface employs a wooden manikin with an integrated inertial measurement unit to provide a tangible user interaction. The other user interface presents a mobile application that provides instructions and explanations about the exercises. Both user interfaces allow recording key postures during the exercise for reference and feedback. We compared the usability of the interfaces through a series of flexion and extension exercises, monitored with an inertial measuring unit worn around the torso, and a Microsoft Kinect V2 vision-based sensor. Although no significant differences between the manikin user interface and the mobile application were found in terms of usability, the inertial measurement unit provided more accurate and reliable data in comparison to the Microsoft Kinect V2 as a result of body occlusions in front of the sensor caused during the torso flexion. Although both user interfaces provide different experiences and performed well, we believe that a combination of both will improve user engagement and motivation, while providing a more accurate motion profile.

A Control Scheme That Uses Dynamic Postural Synergies to Coordinate a Hybrid Walking Neuroprosthesis: Theory and Experiments.

A hybrid walking neuroprosthesis that combines functional electrical stimulation (FES) with a powered lower limb exoskeleton can be used to restore walking in persons with paraplegia. It provides therapeutic benefits of FES and torque reliability of the powered exoskeleton. Moreover, by harnessing metabolic power of muscles via FES, the hybrid combination has a potential to lower power consumption and reduce actuator size in the powered exoskeleton. Its control design, however, must overcome the challenges of actuator redundancy due to the combined use of FES and electric motor. Further, dynamic disturbances such as electromechanical delay (EMD) and muscle fatigue must be considered during the control design process. This ensures stability and control performance despite disparate dynamics of FES and electric motor. In this paper, a general framework to coordinate FES of multiple gait-governing muscles with electric motors is presented. A muscle synergy-inspired control framework is used to derive the controller and is motivated mainly to address the actuator redundancy issue. Dynamic postural synergies between FES of the muscles and the electric motors were artificially generated through optimizations and result in key dynamic postures when activated. These synergies were used in the feedforward path of the control system. A dynamic surface control technique, modified with a delay compensation term, is used as the feedback controller to address model uncertainty, the cascaded muscle activation dynamics, and EMD. To address muscle fatigue, the stimulation levels in the feedforward path were gradually increased based on a model-based fatigue estimate. A Lyapunov-based stability approach was used to derive the controller and guarantee its stability. The synergy-based controller was demonstrated experimentally on an able-bodied subject and person with an incomplete spinal cord injury.

A framework for interpreting, modeling and recognizing human body gestures through 3D eigenpostures

In this article we propose a novel system for recognizing human gestures through acquisition and processing of volumetric data sequences. Volumetric sequences are acquired with two different approaches, a multi-camera set-up and a multi-Kinect $$^\mathrm{TM}$$ set-up. The recognition based on volumetric representation does not require any skeleton fitting or limb tracking and the system relies on the extraction of robust features directly from the available 3D data. Volumetric shape descriptors are, in fact, invariant with respect to viewpoint and body size; they are designed to provide us with a unique signature for each posture. Hidden Markov Models (HMMs), trained on different gestures, are then used for identifying a set of key postures and classifying their sequences over a set of possible actions. The paper also presents a method for identifying the number of hidden states of the HMMs that describe gestures. Despite its implementation and conceptual simplicity, the number of states that we estimate with this method turns out to match that of other classical approaches such as the Akaike Information Criterion (AIC) or the Bayesian Information Criterion (BIC). The same approach is also applied in the definition of the Gaussian Mixture for the Hidden states Observations providing us with good results. Extensive tests were performed on a database that we acquired, which is made of ten different actions, each performed by five different actors and in five different ways (different speed and orientation) and on another public database, achieving a 96% correct recognition rate.

Motion Segmentation and Balancing for a Biped Robot's Imitation Learning

Techniques for transferring human behaviors to robots through learning by imitation/demonstration have been the subject of much study. However, direct transfer of human motion trajectories to humanoid robots does not result in dynamically stable robot movements because of the differences in human and humanoid robot kinematics and dynamics. An imitating algorithm called posture-based imitation with balance learning (Post-BL) is proposed in this paper. This Post-BL algorithm consists of three parts: a key posture identification method is used to capture key postures as knots to reconstruct the motion imitated; a clustering method classifies key postures with high similarity; and a learning method enhances the static stability of balance during imitation. In motion reproduction, the proposed system smoothly transits between key poses and the robot learns to maintain balance by slightly adjusting the leg joints. The developed balance controller uses a reinforcement learning mechanism, which is sufficient to stabilize the robot during online imitation. The experimental results for simulation and a real humanoid robot show that the Post-BL algorithm allows demonstrated motions to be imitated balance to be preserved.

병원 의료종사자의 생활습관자세와 업무특성이 근골격계 자각증상에 미치는 영향

Purpose: The purpose of this study was to identify the influences of hospital employees' daily life posture habits and work-related factors upon musculoskeletal subjective symptoms. Methods: This study was a descriptive survey study. Data were collected using structured a self-report questionnaire between April 1 and May 31, 2015. One hundred and ninety two employees were recruited in three hospitals. The collected data were analyzed using descriptive statistics, <TEX>${\chi}^2$</TEX> test, t-test, and binomial logistic regression. Results: The habit of leaning on one side and the habit of bending the back in an improper posture are key postures based on lifestyle affecting musculoskeletal subjective symptoms in neck, shoulders, arms, waist, and legs. Labours accompanying repeated arm movements for a long time are key work-related risk factors affecting musculoskeletal subjective symptoms in arms. Conclusion: The results of this study confirmed that, to prevent musculoskeletal diseases, it is necessary to identify and mediate personal factors like daily life posture habits as well as work-related risk factors. They may be utilized as basic materials for education of musculoskeletal health promotion and development of life guidance programs.

Humanoid Robot's Visual Imitation of 3-D Motion of a Human Subject Using Neural-Network-Based Inverse Kinematics

A sequence of 3-D motion images of a human facing an imitator [i.e., a humanoid robot (HR)] is captured using a stereo-vision system installed on the HR. After acquiring enough motion sequences via image ratio and background registration, the estimation of seven feature points (i.e., the head, two arm tips, two leg tips, and two elbows) of the human are obtained. Using five feature points (i.e., the head, two arm tips, and two leg tips), the extraction of key posture frames from the recorded videos is achieved. To ensure a stable imitation motion of the HR, the human motions of the lower body (LB) and the upper body (UB) are imitated separately. All 11 stable motions for the LB are classified by the motion direction, the motion state, and the support phase, whereas only the two arm tips are employed to imitate the motions of the UB. Due to the disadvantages of traditional inverse kinematics (IK), neural-network-based IK is developed to reduce the computation time and improve the accuracy of the time response of the system. Finally, the combined motions of the LB and the UB with suitable interpolation and time intervals are applied to imitate the 3-D motion of a human. The corresponding experimental results confirm the efficacy of the proposed technique.