Trajectory Recognition Research Articles

Volleyball track acquisition and intelligent analysis technology

Volleyball has been developing rapidly since 1996. It has been widely used both in competitive events and among ordinary people. However, with the continuous improvement of the sports level, the traditional manual training methods cannot meet the existing technical requirements. It is a mainstream method to analyze the track of volleyball by the computer in volleyball training. However, there are still some technical problems such as low precision and incomplete analysis. Therefore, this paper puts forward the research of volleyball track acquisition and intelligent analysis technology. In this paper, the shortcomings of the existing technology are systematically analyzed, and on this basis, the optimization and improvement scheme is proposed. The core technology of this project is to improve the original image preprocessing technology and strengthen the system’s feature extraction ability. Finally, combined with the CAMSHAFT moving object tracking algorithm, the technical scheme of this paper is formed. Through a series of technical improvements, the system effectively improves the ability to extract and analyzing the track of volleyball. In order to further verify the practical effectiveness of this scheme, a number of comparative experiments including algorithm accuracy comparison experiment, trajectory recognition detection, and algorithm signal-to-noise ratio verification are carried out. The object of comparison is the current mainstream common filtering algorithm. Through the analysis of experimental data, this method is more accurate than the common filtering algorithm in the extraction of volleyball trajectory, which effectively improves the comprehensive performance and robustness of the traditional method.

Intelligent recognition of dance training movements based on machine learning and embedded system

Recognizing human movement is an important research topic in the field of human-computer interaction, and people expect it to be used in smart homes, virtual reality, and electronic games. Based on the interaction between humans and computers, more and more attention has been paid, especially in the field of smart home action recognition. Through observation, people can understand the intention of intelligent interaction is included in the main part. However, the current recognition algorithms still cannot meet the actual requirements of the accuracy, real-time and robustness of human motion recognition. Especially in order to recognize complex human movements in real time, it is imperative to solve several problems in motion capture and recognition. Establishing the feature parameter angle of the feature vector space of motion data, using the pre-recognition algorithm is based on multi-class support vector machines. The motion recognition algorithm takes advantage of the accurate and fast classification function of svm. Based on the structural differences of the motion data, most of the data can be correctly identified. The optimal motion recognition algorithm uses hmm to correct the svm error recognition result through the random constraint relationship between the error recognition data and the actual label. Based on data simulation and analysis, each variable determined by the grid search algorithm has the highest accuracy in the optimization of each variable of the support vector machine. Finally, a smart home simulation experiment interactive system was built, and a local database was created, including 1,300 processes. The real-time algorithm uses the data in the local database for training and testing. Experimental results show that the motion recognition algorithm in this paper improves the accuracy and robustness of complex motion recognition. While meeting the real-time recognition conditions, the correct answer rate of the final operation can reach 9.6%. The human motion trajectory recognition system uses the three-dimensional trajectory of gestures to recognize motion. The information in the three-dimensional space is more comprehensive, and the orbit recognition is more robust.

A hybrid Hausdorff distance track correlation algorithm based on time sliding window

In multi-sensor target tracking, track correlation is the key to the unification of global situation. Hausdorff distance has been applied to power fault elimination, point cloud data, medical measurement, image segmentation, vehicle trajectory recognition and other directions. To solve the problem of track correlation, a hybrid Hausdorff distance track correlation algorithm based on time sliding window is proposed. The hybrid Hausdorff distance based on position, speed and azimuth is defined, and the time sliding window is added on this basis. Simulation results show that the proposed algorithm can maintain a high correct correlation rate under the conditions of target maneuver, time asynchronism, and inconsistent radar sampling frequency, the algorithm is superior to traditional track correlation algorithm.

Bioinspired transparent and antibacterial electronic skin for sensitive tactile sensing

Abstract Triboelectric electronic skin has attracted increasing attention in recent years by virtue of the self-powered feature. Complex and harsh technologies are extensively used to create high specific surface area to improve the sensitivity of tactile sensing. The fabrication of hierarchical micro or nano structures with efficient and facile strategy remains a tough task. Besides, rarely any progress has been reported to consider the problem of bacterial growth for long time contact of the electronic skin with the human body. Here we report a flexible, transparent, and antibacterial triboelectric electronic skin for sensitive tactile sensing. Inspired by the microstructure of rose petal, a special “replication” method is developed to prepare the sensing layer with uniform and hierarchical nano-in-micro structure, avoiding the diversity and heterogeneity of the rose petal structure. When combined with a fiber-based conductive network and an antibacterial substrate, this highly shape adaptive electronic skin can be attached conformably on any curved surface and allow the functions of distance resolution, multi-tactile sensing, and trajectory recognition in real-time. The work presents a significant advance towards multifunctional electronic skin with excellent practicability.

Movement Trajectory Recognition of Sign Language Based on Optimized Dynamic Time Warping

Movement trajectory recognition is the key link of sign language (SL) translation research, which directly affects the accuracy of SL translation results. A new method is proposed for the accurate recognition of movement trajectory. First, the gesture motion information collected should be converted into a fixed coordinate system by the coordinate transformation. The SL movement trajectory is reconstructed using the adaptive Simpson algorithm to maintain the originality and integrity of the trajectory. The algorithm is then extended to multidimensional time series by using Mahalanobis distance (MD). The activation function of generalized linear regression (GLR) is modified to optimize the dynamic time warping (DTW) algorithm, which ensures that the local shape characteristics are considered for the global amplitude characteristics and avoids the problem of abnormal matching in the process of trajectory recognition. Finally, the similarity measure method is used to calculate the distance between two warped trajectories, to judge whether they are classified to the same category. Experimental results show that this method is effective for the recognition of SL movement trajectory, and the accuracy of trajectory recognition is 86.25%. The difference ratio between the inter-class features and intra-class features of the movement trajectory is 20, and the generalization ability of the algorithm can be effectively improved.

Welding Seam Trajectory Recognition for Automated Skip Welding Guidance of a Spatially Intermittent Welding Seam Based on Laser Vision Sensor.

To solve the problems of low teaching programming efficiency and poor flexibility in robot welding of complex box girder structures, a method of seam trajectory recognition based on laser scanning displacement sensing was proposed for automated guidance of a welding torch in the skip welding of a spatially intermittent welding seam. Firstly, a laser scanning displacement sensing system for measuring angles adaptively is developed to detect corner features of complex structures. Secondly, a weld trajectory recognition algorithm based on Euclidean distance discrimination is proposed. The algorithm extracts the shape features by constructing the characteristic triangle of the weld trajectory, and then processes the set of shape features by discrete Fourier analysis to solve the feature vector used to describe the shape. Finally, based on the Euclidean distance between the feature vector of the test sample and the class matching library, the class to which the sample belongs is identified to distinguish the weld trajectory. The experimental results show that the classification accuracy rate of four typical spatial discontinuous welds in complex box girder structure is 100%. The overall processing time for weld trajectory detection and classification does not exceed 65 ms. Based on this method, the field test was completed in the folding special container production line. The results show that the system proposed in this paper can accurately identify discontinuous welds during high-speed metal active gas arc welding (MAG) welding with a welding speed of 1.2 m/min, and guide the welding torch to automatically complete the skip welding, which greatly improves the welding manufacturing efficiency and quality stability in the processing of complex box girder components. This method does not require a time-consuming pre-welding teaching programming and visual inspection system calibration, and provides a new technical approach for highly efficient and flexible welding manufacturing of discontinuous welding seams of complex structures, which is expected to be applied to the welding manufacturing of core components in heavy and large industries such as port cranes, large logistics transportation equipment, and rail transit.

App trajectory recognition over encrypted internet traffic based on deep neural network

Despite the increasing popularity of mobile applications and the widespread adoption of encryption techniques, mobile devices are still susceptible to security and privacy risks. In this paper, we propose ActiveTracker, a new type of sniffing attack that can reveal the fine-grained trajectory of userâs mobile app usage from a sniffed encrypted Internet traffic stream. It firstly adopts a sliding window based approach to divide the encrypted traffic stream into a sequence of segments corresponding to different app activities. Then each traffic segment is represented by a normalized temporal-spacial traffic matrix and a traffic spectrum vector. Based on the normalized representation, a deep neural network (DNN) model which consists of an app filter and an activity classifier is developed to extract comprehensive features from the input and uncover the crucial app usage trajectory conducted by the user. By extensive experiments on real-world app usage traffic collected from volunteers and on our synthetic traffic data, we show that the proposed approach achieves up to 79.65% accuracy in recognizing app trajectory over encrypted traffic streams.

Taking a History in Neurocritically Ill Patients.

A clinical history leads to an examination, tests and a diagnosis. This time-honored sequence in medicine remains valid in critical illness, but in the heat of the moment there is a quickly appearing inevitable sketchiness. Intensivists should never be too unquestioning, too comfortable with incomplete information, or too unwilling to start over if information is muddled or contradictory. No scale in neurology looks at history. There is no tool or requirement to provide a standard system of communication. I review the essentials of history taking in a neurocritically ill patient. Examples of the value of a good medical history are shown but also the familiar biases when asking questions. There are obstacles, errors of commission and omission, and the importance of recognition of a clinical trajectory.

ImCAPTCHA: Imperceptible CAPTCHA Based on Cursor Trajectories

In the process of network interaction, CAPTCHA (completely automated public Turing test to tell computers and humans apart) is usually required. At present, the main method for differentiating between humans and machines is to use CAPTCHA, which includes images and text. Users need to interface with CAPTCHA for verification, which prolongs the interaction step. The interactive experience is poor, and the process of verification is not concealed. To simplify the verification process, optimize the interaction behavior and increase the security of verification, this paper proposes a CAPTCHA method based on a user's mouse operation behavior, and carries out a study on cursor trajectory recognition. This method can reduce the manual procedure and improve the user interactive experience. In this article, we propose an ensemble learning algorithm model based on sliding sampling to classify and recognize cursor trajectory data. Experiments show that the classification performance of the model is reliable for recognition based on human–machine cursor trajectories, which is a new concept for CAPTCHA.

Control Your Home With a Smartwatch

With the wide applications of smart devices and mobile computing, smart home becomes a hot issue in the household appliance industry. The controlling and interaction approach plays a key role in users' experience and turns into one of the most important selling points for profit growth. Considering the robustness and privacy protection, wearable devices equipped with MEMS, e.g., smartphones, smartwatches, or smart wristbands, are thought of one of the most feasible commercial solutions for interaction. However, the low-cost built-in MEMS sensors do not perform well in capturing finely grained human activity directly. In this paper, we propose a method that leverages the arm constraint and historical information recorded by MEMS sensors to estimate the maximum likelihood action in a two-phases model. First, in the arm posture estimation phase, we leverage the kinematics model to analyze the maximum likelihood position of users' arms. Second, in the trajectory recognition phase, we leverage the gesture estimation model to identify the key actions and output the instructions to devices by SVM. Our substantial experiments show that the proposed solution can recognize eight kinds of postures defined for man-machine interaction in the smart home application scene, and the solution implements efficient and effective interaction using low-cost smartwatches, and the interaction accuracy is >87%. The experiments also show that the algorithm proposed in this paper can be well applied to the perceptual control of smart household appliances, and has high practical value for the application design of the perceptual interaction function of household appliances.

A room-level tag trajectory recognition system based on multi-antenna RFID reader

Trajectory of objects is a typical data source of indoor location aware applications and smart systems. In some object tracking applications, the continuity of moving trajectories recognition is more important than its positioning accuracy. To solve the problem of signal coverage blind areas and multi-tag identification reliability, this paper proposes a tag trajectory index method that solves the problem of localized tag trajectories between readers, and establishes a tag identification model based on RFID reader with multiple antennas, which is more economic and efficient in deployment. Furthermore, it proposes a multi-antenna optimization control scheme, which automatically adapts to the actual demands to control the number of working antennas, so as to improve the recognition efficiency. A counterpart indoor tag trajectory reasoning data structure OPTR-tree is proposed, which provides an approach for multi-tag track fast recognition in indoor environment at room-level accuracy. Both theoretical analysis and experimental results show that the proposed multi-antenna model can improve the reliability of multiple tags recognition inside buildings Hopefully, the indoor trajectories data can offer a new way for more effective disaster management.

LSTM‐based dynamic probability continuous hand gesture trajectory recognition

In the field of continuous hand-gesture trajectory recognition, aiming at the problems of existing a lot of noise for handwriting trajectories, and difficult to segment multiple continuous hand gestures accurately, a long short-term memory-based dynamic probability (DP-LSTM) method is proposed. Firstly, obtain the classification result for each sub-period in the whole time period by using LSTM; secondly, cluster the classification results by non-maximum suppression for trajectory algorithm to eliminate interference of invalid subsets; Finally, the end point of the valid trajectory is obtained according to the characteristics of the probability change, thus realising dynamic trajectory segmentation and recognition. In order to evaluate the performance of the DP-LSTM, this method is evaluated by using an Arabic numerals gesture database. The experiments show that the DP-LSTM has a high recognition rate for continuous hand gestures and can recognise its in real time.

Mobile terminal trajectory recognition based on improved LSTM model

Mobile terminal hand trajectory recognition is challenged by spatio-temporal variations, complex background, light variations as well as the limited computing resources for mobile devices. The authors propose a hand trajectory recognition method based on improved Long-Short Term Memory network (LSTM) model. First, for the sake of eliminating the effects of complex background and light, the SSD hand detector and multi-color space are combined to segment gesture. To eliminate the influence of trajectory points' overlap, the sparseness of the trajectory points is used to determine the start and end of the trajectory; then, to maintain the feature difference while decreasing the feature extraction time, they extract the trajectory feature sequence based on the central point direction angle; Finally, a new loss function (M-Softmax loss) is proposed in LSTM training process to maximise the inter-class variations and minimise intra-class variations. At the same time, the adaptive training iteration number is proposed to ensure the convergence of the model and reduce the training time cost. The experimental results illustrate that the improved model learned features are more compact, and the model is easier to deal with external data changes. Authors’ proposed method has great significance in practical applications.

Phase-Space-Based Pilot Main Protection for a Transmission Line Immune to Timing Attack and Controllable Shunt Reactors

Pilot current differential protection for transmission lines is an important part of ensuring the safety of the power grid. The loss of data synchronization due to cyberattacks and the uncertain compensation degree of controllable shunt reactors will evidently deteriorate the performance of pilot current differential protection and may lead to the maloperation of the protection if no effective countermeasure is implemented. To address the above-mentioned problems, a novel line pilot protection criterion based on the recognition of the phase space trajectory is proposed. First, the basic principle of phase space and its parameter selection scheme are analyzed. Then, the fault component instantaneous power difference (FCIPD) is selected as the time series for phase space reconstruction, and the distribution characteristics of the phase space trajectory of the FCIPD under different conditions are analyzed. Subsequently, a novel pilot main protection criterion for a transmission line is proposed. PSCAD/EMTDC-based simulation results verify that the proposed criterion is immune to timing attack and controllable shunt reactors and has the advantages of free setting, high speed, and high-fault-resistance tolerance capability.

Synergistic sensing of stratified structures enhancing touch recognition for multifunctional interactive electronics

Efficient touch feedback, capable of monitoring the magnitude of force and identifying active location, is significant to artificial intelligence and interactive robotics. It generally needs the integration of multitudinous sensing elements and intricate manufacturing procedures. Here we propose a multifunctional paper-based touch sensor to realize touch trajectory recognition as well as achieve pressure information. The asymmetric and symmetric structures are designed to skillfully construct localization layer and pressure sensing layer. These functional layers effectively assemble a scalable touch sensor and, thus, greatly simplify the device's architecture with the competitive advantages of easiness in fabrication, cost-effectiveness, self-switching characteristic, and programmability in interactive function. Through coding and using the electrical signals, human-computer interaction, human-machine interaction, and force-enhanced cryptographic matrix are explored and demonstrate the feasibility of the proposed touch sensor. This work provides a novel mechanosensational sensing paradigm to leverage the complex physics of a feasible strategy for advancing human-related interactive electronics.

Patterns of Gestational Weight Gain in Women with Overweight or Obesity and Risk of Large for Gestational Age

Objective: Timing of gestational weight gain (GWG) may influence perinatal outcomes differently. This study aimed to find associations of latent GWG patterns with risk of large for gestational age (LGA) in women with overweight or obesity. Methods: A total of 4,438 women with overweight or obesity were included in the analyses. Latent trajectories of GWG associated with LGA were identified by trajectory analysis. GWG, risk of LGA and early pregnancy factors were compared between these identified groups. Results: This study identified four distinct GWG trajectory groups associated with LGA, each group including, respectively, 78.6, 19.0, 1.4, and 0.9% of the participants. Group 1 presented a typical curve with lower GWG in early pregnancy and relatively higher GWG in mid- and late pregnancy. Women in group 2 showed sustained high GWG and high LGA prevalence than women in group 1 (48.24 vs. 21.56%, p < 0.0001). A catch-up in GWG after low weight gain did not result in significantly higher risk of LGA in group 3 compared to group 1. On the other hand, a rapid GWG in the first two-thirds of pregnancy followed by a strict weight control was associated with elevated risk of LGA in group 4 than group 1 (62.50 vs. 21.56%, p < 0.0001). Conclusions: Women affected by overweight or obesity combined with high GWG in early mid-pregnancy were at elevated risk of LGA. Early clinical recognition of a poor GWG trajectory will enable early intervention in high-risk groups.

Automatic trajectory recognition in Active Target Time Projection Chambers data by means of hierarchical clustering

The automatic reconstruction of three-dimensional particle tracks from Active Target Time Projection Chambers data can be a challenging task, especially in the presence of noise. In this article, we propose a non-parametric algorithm that is based on the idea of clustering point triplets instead of the original points. We define an appropriate distance measure on point triplets and then apply a single-link hierarchical clustering on the triplets. Compared to parametric approaches like RANSAC or the Hough transform, the new algorithm has the advantage of potentially finding trajectories even of shapes that are not known beforehand. This feature is particularly important in low-energy nuclear physics experiments with Active Targets operating inside a magnetic field. The algorithm has been validated using data from experiments performed with the Active Target Time Projection Chamber developed at the National Superconducting Cyclotron Laboratory (NSCL). The results demonstrate the capability of the algorithm to identify and isolate particle tracks that describe non-analytical trajectories. For curved tracks, the vertex detection recall was 86% and the precision 94%. For straight tracks, the vertex detection recall was 96% and the precision 98%. In the case of a test set containing only straight linear tracks, the algorithm performed better than an iterative Hough transform.

Perception of Surrounding Sound Source Trajectories in the Horizontal Plane: A Comparison of VBAP and Basic-Decoded HOA

Despite the fundamental role played by sound in multiple virtual reality contexts, few studies have explored the perception of virtual sound source motion in the acoustic space. The goal of this study was to compare the localization of virtual moving sound sources rendered with two different spatialization techniques: Vector BaseAmplitude Panning (VBAP) and fifth-order Ambisonics (HOA), both implemented in a soundproofed room and in their most basic form (basic decoding of HOA, VBAP without spread parameter). The perception of virtual sound trajectories surrounding untrained subjects (n=23) was evaluated using a new method based on a drawing-augmented multiple-choice questionnaire. In the spherical loudspeaker array used in this study, VBAP proved to be a robust spatialization technique for sound trajectory rendering in terms of trajectory recognition and height perception. In basic-decoded HOA, subjects exhibited far more disparate trajectory recognition and height perception performances but performed better in perceiving sound source movement homogeneity.

Structured dynamic time warping for continuous hand trajectory gesture recognition

Continuous hand gesture recognition is an important area of HCI and challenged by various writing habits and unconstrained hand movement. In this paper, we propose a Structured Dynamic Time Warping (SDTW) approach for continuous hand trajectory recognition. We first propose an automatic continuous trajectory segmentation approach which combines templates and velocity information to spot the beginning and ending points in hand gesture trajectories. Then we assign different weights to feature sequences based on the structured information, from the positions of corner points in the arbitrary trajectories. Finally, we evaluate the SDTW on the Continuous Letter Trajectory (CLT) database. Experimental results show that the proposed approach is robust to the diversity of same handwritten letter, and significantly outperforms state-of-the-art approaches.

Quantitative recognition method for borehole trajectory models

Abstract In order to accurately calculate drilled trajectories, the method of quantitatively recognizing borehole trajectory models was provided, and a case analysis was conducted. Because the measurement-while-drilling data provide with measured values of tool-face angle besides inclination angle and azimuth angle, this paper presents the technological approach of recognizing borehole trajectory models based on tool-face angle. A universal tool-face angle equation was established based on the directional deflection mechanism of steerable drilling tools, and it can calculate the tool-face angles with characteristic parameters of various borehole trajectory models. Then, by evaluating the error between the theoretical values and the measured values of tool-face angle, the trajectory model most consistent with the actual well trajectory can be selected. The model recognition of borehole trajectory provides with the quantitative evaluation index and selection basis of survey calculation methods, which can avoid subjectively and randomly selecting the survey calculation method, and consequently improve the monitoring accuracy and reliability of borehole trajectory.