Trajectory Characteristics Research Articles

Small Aerial Target Detection Using Trajectory Hypothesis and Verification

Due to cluttered backgrounds, long-range imaging, rapid relative motion, etc., the detection of small aerial targets remains a challenge. Existing works have illustrated the importance of temporal cues for robust small aerial target detection. However, the strong assumptions about the temporal cues give such methods high computational complexity and weak adaptability. To address this problem, we treat trajectory verification as a classification problem using convolutional neural networks. A novel small aerial target detection method that uses trajectory hypothesis and verification is proposed in this study. First, the IMU assisted inter-frame difference is used to extract target candidates from each single image. Then, based on the continuous and smoothness characteristics of the target trajectory, hypotheses of the required length are generated by linking target candidates in adjacent frames. Finally, the trajectory hypotheses are sent to a trained classification convolution neural network for verification. The detected targets are traced back from the verified trajectory. Furthermore, trajectory merge is conducted to link adjacent trajectory segments. The proposed method can achieve real-time processing. Experimental results on public datasets show that the proposed method performs better than existing methods.

A Study of Hypersonic Vehicle-Borne SAR Imaging Under Plasma Sheath

Hypersonic vehicle-borne synthetic aperture radar (SAR) imaging has promising application prospects for remote sensing owing to the characteristics of fast response and flexible trajectory. However, a newly observed phenomenon that occurs during hypersonic flight, called the plasma sheath, seriously affects hypersonic vehicle-borne SAR imaging. Therefore, in this study, the effect of plasma sheath on hypersonic vehicle-borne SAR imaging is investigated systematically. First, a hypersonic vehicle-borne SAR signal model under plasma sheath is developed. Then, the phase shift and amplitude attenuation which are critical in the SAR signal model under plasma sheath are deduced. Moreover, the analytical formulas for the linear and quadratic phase errors are derived. Based on the developed signal model, point target response and an SAR image of an area under plasma sheath show that linear phase error will cause image shift, and quadratic phase error will lead to pulse broadening. In addition, the amplitude attenuation will make the point target response submerged in noise, and signal amplitude nonlinear distortion will result in an asymmetric distortion phenomenon. To further explore the effect of plasma sheath on hypersonic vehicle-borne SAR imaging, quantitative evaluation and analysis of SAR imaging degradation are conducted under different electron densities, carrier frequencies, and bandwidths. The evaluation results in this study will benefit the application of hypersonic vehicle-borne SAR imaging under plasma sheath, and the analytical formulas of linear and quadratic phase error provide the possibility to compensate for SAR imaging degradation under plasma sheath in the future.

Research on basketball dunk motion recognition method based on characteristic point trajectory

Research on basketball dunk motion recognition method based on characteristic point trajectory

Research on basketball dunk motion recognition method based on characteristic point trajectory

Research on basketball dunk motion recognition method based on characteristic point trajectory

TA-LSTM: A Time and Attribute Aware LSTM for Deep Flight Track Clustering

The rapid development of civil aviation imposes an urgent need for efficient air traffic management. Flight track clustering is the premise and foundation of air traffic control. Flight track points compose time-series data with non-identical point- to-point intervals, and the effective latent representation of track data is key to the flight trajectory clustering. Long Short-Term Memory (LSTM) is an important and popular method to extract the features of trajectory time-series data. However, LSTM ignores the irregular time intervals between track points and fails to effectively distinguish the importance of different attributes at each track point during feature extraction. In addition, the existing methods based on LSTM either ignore the distinct impacts of different attributes at each track point or simply assume that the influences of the previous track points on the subsequent ones decrease with time. In this paper, we propose a flight trajectory feature extraction unit called TA-LSTM (Time and Attribute Aware Long Short- Term Memory) and a flight track clustering model based on TA- LSTM. The flight track clustering model consists of a feature extraction layer and a clustering layer. The feature extraction layer adopts TA-LSTM to obtain the latent representation of flight tracks, which adds a time control gate and an attribute control gate to the standard LSTM unit. The time control gate manipulates the action of each track point on subsequent points to obtain the characteristics of irregular time intervals. The attribute control gat enables LSTM to capture the impact of different attributes on the flight track features at various points. The clustering layer takes the latent representation of flight tracks as the input to obtain the clustering results. The clustering model updates the parameters of TA-LSTM and the cluster centers by minimizing the Kullback-Leibler (K-L) divergence. We conduct experiments using Automatic Dependent Surveillance-Broadcast track data provided by VariFlight. In comparison with popular and state-of-the-art methods, the proposed method obtains superior pe formance in three widely used evaluation metrics, i.e., Silhouette Coefficient, Calinski-Harabaz index and Davies-Bouldin index.

Chinese Sign Language Recognition Based on Two-stream CNN and LSTM Network

Sign language recognition is the use of computer technology to convert sign language into text or speech to facilitate the communication between deaf-mute people and normal people. This paper takes Chinese sign language words as the research object, and proposes a new method of sign language recognition based on two-stream 3D-CNN and LSTM network. First, the key frame extraction algorithm is used to remove redundant data frames in the original data, and then a two-stream3D-CNN is used to learn local hand change features and global trajectory features at the same time, and aggregated as the feature input of the video clip to the LSTM codec network. In order to focus on the video frames that express the meaning of sign language, a time attention mechanism is introduced in the LSTM encoding and decoding network. On the DEVISIGN-D sign language data set, an experiment was compared with three sign language recognition algorithms, the experimental results show that the method can identify Chinese isolated words sign language very well, with an accuracy rate of 98.4%.

A Novel Model for Ship Trajectory Anomaly Detection Based on Gaussian Mixture Variational Autoencoder

The use of traditional methods in anomaly detection of multi-class ship trajectories showed some limitations in terms of robustness and learning ability of trajectory features. In view of this, an anomaly detection model for ship trajectory data based on Gaussian Mixture Variational Autoencoder (GMVAE) is proposed in this study using an unsupervised classification method. The proposed model modifies Variational Autoencoder (VAE) by changing the inferential distribution of prior distribution and approximate posterior to the Gaussian mixture model. A high-dimensional hidden space is constructed to learn the features of multi-class trajectory data, and the Dynamic Time Warping (DTW) method is applied to measure the error between the reconstructed trajectory and the original trajectory in order to judge whether the ship trajectory is abnormal. The Automatic Identification System (AIS) data from the US coastal areas are used to verify the proposed model, and the results are compared with other commonly used models in a manually labeled dataset. The research results indicate that the detection rate of the proposed model is 91.26%, and the false alarm rate is 0.68%, which performs the best. Using the Gaussian mixture model to describe the distribution of hidden space can improve the learning ability of multi-class trajectories of VAE, thus increasing the robustness of the model. This research can provide technical support for ship trajectory data analysis and risk management of maritime transportation.

Characteristic area-based method for continuous coverage analysis of satellites constellation

Coverage capacity analysis of satellites constellation is the basis of satellite system design and performance evaluation, which is essential in the application of satellite technology. A characteristic area based method for continuous coverage analysis is proposed in this paper. The trajectory characteristics of the Walker constellation are first analyzed. The Earth's surface is partitioned into a group of characteristic areas, and the characteristic distance theory is defined and proposed accordingly. On this basis, the formulations of the characteristic point of each characteristic area and the corresponding characteristic distance with all different situations can be analytically obtained. Additionally, sufficient and necessary conditions for continuous coverage are given accordingly. The complex continuous coverage problem can be converted to the analysis of the relations between the critical geocentric angle of the satellite coverage range and the characteristic distance. Computational results indicate the effectiveness of the proposed method and reveal the feasibility in instructing the constellation design.

Early developmental profiles of sensory features and links to school-age adaptive and maladaptive outcomes: A birth cohort investigation.

Sensory-based subtypes among autistic children have been well documented, but little is known about longitudinal sensory subtypes beyond autistic populations. This prospective study aimed to identify subtypes based on trajectories of parent-reported sensory features measured at 6-19 months, 3-4, and 6-7 years of age among a community-based birth cohort (N = 1,517), and to examine their associations with school-age clinical and adaptive/maladaptive outcomes on a subset sample (N = 389). Latent class growth analysis revealed five trajectory subtypes varying in intensity and change rates across three sensory domains. In contrast to an Adaptive-All Improving subtype (35%) with very low sensory features and overall better school-age outcomes, an Elevated-All Worsening subtype (3%), comprised of more boys and children of parents with less education, was associated with most elevated autistic traits and poorest adaptive/maladaptive outcomes. Three other subtypes (62% in total) were generally characterized by stable or improving patterns of sensory features at mild to moderate levels, and challenges in certain outcome domains. Our findings indicate that characterizing children based on early sensory trajectories may contribute to earlier detection of subgroups of children with sensory challenges who are more likely to experience developmental challenges by school age, followed by early targeted interventions for improved long-term outcomes.

Visual extensions and anomaly detection in the pNEUMA experiment with a swarm of drones

The usage of Unmanned Aerial Systems (UAS) in traffic monitoring has advantages such as broader vision, flexibility, privacy, and cost-efficiency compared to other traffic monitoring sensors like loop detectors or fixed surveillance cameras. UAS has been instrumental in the recent pNEUMA experiment, where a swarm of drones collected a large-scale urban traffic dataset containing vehicle trajectories. These trajectory data are subject to challenges such as noise due to visual restrictions, perspective distortions, and human-induced errors. While the pNEUMA dataset is missing its imagery part, we present an extended version of it named pNEUMA Vision, which incorporates imagery data and annotations of vehicles in the form of image coordinates along with newly added vehicle trajectory features like azimuth. Moreover, we demonstrate that visually restricted trajectories are also highly prone to become anomalies, and vice-versa, through novel anomaly detection proposed in this literature. Specifically, we distinguish between stationary and non-stationary errors and argue that the latter account for the largest part of the noise. Furthermore, we analyze the new visual dataset with two different computer vision methods for estimating the number of vehicles on the roads from the input images. Particularly, we show that using a density map to count vehicles from drone images achieves comparable results to conventional vehicle counting methods via detection. Results show that the time–space diagrams plotted from density map predictions could identify the congested urban roads’ queues better than a widely used vehicle detection method.

Fast Adiabatic Control of an Optomechanical Cavity

The development of quantum technologies present important challenges such as the need for fast and precise protocols for implementing quantum operations. Shortcuts to adiabaticity (STAs) are a powerful tool for achieving these goals, as they enable us to perform an exactly adiabatic evolution in finite time. In this paper, we present a shortcut to adiabaticity for the control of an optomechanical cavity with two moving mirrors. Given reference trajectories for the mirrors, we find analytical expressions that give us effective trajectories which implement an STA for the quantum field inside the cavity. We then solve these equations numerically for different reference protocols, such as expansions, contractions and rigid motions, thus confirming the successful implementation of the STA and finding some general features of these effective trajectories.

A Metallurgical Dynamics-Based Method for Production State Characterization and End-Point Time Prediction of Basic Oxygen Furnace Steelmaking

Basic Oxygen Furnace (BOF) steelmaking is an important way for steel production. Correctly recognizing different blowing periods and abnormal refining states is significant to ensure normal production process, while accurately predicting the end-point time helps to increase the first-time qualification rate of molten steel. Since the decarburization products CO and CO2 are the main compositions of off-gas, information of off-gas is explored for BOF steelmaking control. However, the problem is that most of the existing research directly gave the proportions of CO and CO2 as model input but barely considered the variation information of off-gas to describe the production state. At the same time, the off-gas information can be expected to recognize the last blowing period and predict the end-point time earlier than the existing methods that are based on sub-lance or furnace flame image, but little literature makes an attempt. Therefore, this work proposes a new method based on functional data analysis (FDA) and phase plane (PP), defined as FDA-PP, to describe and predict the BOF steelmaking process from the metallurgical dynamics viewpoint. This method extracts the total proportion of CO and CO2 and its first-order derivative as dynamics features of steelmaking process via FDA, which indicate the reaction velocity and acceleration of decarburization reaction, and describes the evolution of dynamics features via PP. Then, the FDA-PP method extracts the features of phase trajectories for production state recognition and end-point time prediction. Experiments on a real production dataset demonstrate that the FDA-PP method has higher production state recognition accuracy than the classical phase space, SVM, and BP methods, which is 87.78% for blowing periods of normal batches, 90.94% for splashing anomaly, and 81.29% for drying anomaly, respectively. At the same time, the FDA-PP method decreases the mean relative prediction error (MRE) of the end-point time prediction for abnormal batches by about 10% compared with the SVM and BP methods.

An Improved Social Force Model of Pedestrian Twice–Crossing Based on Spatial–Temporal Trajectory Characteristics

Pedestrian two-stage crossing, as one of the key elements of the urban roadway network, affects not only vehicle flow at signalized interactions, but also road capacities in the transport system. Therefore, it is vital to deeply understand the behavioral characteristics of pedestrian twice-crossing in order to improve the safety and efficiency of the road transport network. Based on our previous study, this study continues to improve the social force model by classifying the trajectory type of pedestrian twice crossing. In the interactive aggregation, the pedestrian trajectory line was divided into two types: straight path and curved path. The Work–Energy Principle and Impulse–Momentum Principle were used to identify the spatial and temporal characteristics of pedestrian twice-crossing behaviors. It was found that when pedestrians on the two sides are facing very close in a congested section, the maximum repulsive force appears to be a dramatic increase and remains for a period of time. This result provides us with direction for updating the social force model, focusing on the repulsive force generated by the opposite flow. The improved model can achieve high precision in predicting pedestrian twice-crossing behaviors. The findings of this study have great implications for designing pedestrian facilities and optimizing pedestrian signal timings, thus helping to increase the mobility and safety of pedestrian twice-crossing.

Study on the trajectory of tip leakage vortex and energy characteristics of mixed-flow pump under cavitation conditions

To investigate the mechanism of tip leakage vortex (TLV) and energy loss caused by pump cavitation. The SST k-ω turbulence model coupled with the Reynolds-averaged Navier-Stokes equations, combined with the homogeneous flow cavitation model is employed. And the accuracy of the numerical simulation is verified experimentally. The results show that the cavitation intensity seriously affects the load distribution on suction side (SS), and the variation pattern of the low-pressure area distribution on SS is related to the cavitation location. As cavitation deteriorates, the working capacity of the impeller is weakened and the guide vane is also affected, resulting in a sharp drop in the energy performance of the pump. The malignant cavitation becomes the main reason for the decreasing head and unstable operation of the mixed flow pump. Cavitation deterioration reduces the output power of the impeller, and the power distribution characteristics are affected by the degree of cavitation. Malignant cavitation leads to increased frictional and turbulent dissipation losses and increased total energy loss, resulting in a decrease in the effective energy obtained, leading to a decrease in the overall pump performance. The results can provide theoretical reference for improving the cavitation performance of the pumps.

Visual servoing path-planning for approaching cylindrical objects

Eye-in-hand systems have received significant attention for the task of approaching an object identified by its reference image shown in advance. Image based visual servoing (VS) methods demonstrate robustness to image noises, but encounter difficulties especially when camera displacement is large. More challenges exist when the object is cylindrical with less texture. This paper proposes new feature set and pertinent trajectory planning scheme to achieve a convergent path with tolerable violation of camera field of view (FOV) limits, allowing transient loss of part of the observed object. Specifically, new features and interaction matrices are developed to achieve global convergence. Feature trajectories are then planned with tolerable FOV violation through a constrained nonlinear minimization path-planning technique. They are later tracked via an adapted IBVS controller in a switching manner. Simulation with two views of a real drinking vessel validates the proposed method and demonstrates its adaption to partial FOV violation.

Equilibrium social activity during an epidemic

During an infectious-disease epidemic, people make choices that impact transmission, trading off the risk of infection with the social-economic benefits of activity. We investigate how the qualitative features of an epidemic's Nash-equilibrium trajectory depend on the nature of the economic benefits that people get from activity. If economic benefits do not depend on how many others are active, as usually modeled, then there is a unique equilibrium trajectory, the epidemic eventually reaches a steady state, and agents born into the steady state have zero expected lifetime welfare. On the other hand, if the benefit of activity increases as others are more active (“social benefits”) and the disease is sufficiently severe, then there are always multiple equilibrium trajectories, including some that never settle into a steady state and that welfare dominate any given steady-state equilibrium. Within this framework, we analyze the equilibrium impact of a policy that modestly reduces the transmission rate. Such a policy has no long-run effect on society-wide welfare absent social benefits, but can raise long-run welfare if there are social benefits and the epidemic never settles into a steady state.

Dysconnectivity in Schizophrenia Revisited: Abnormal Temporal Organization of Dynamic Functional Connectivity in Patients With a First Episode of Psychosis.

Abnormal functional connectivity between brain regions is a consistent finding in schizophrenia, including functional magnetic resonance imaging (fMRI) studies. Recent studies have highlighted that connectivity changes in time in healthy subjects. We here examined the temporal changes in functional connectivity in patients with a first episode of psychosis (FEP). Specifically, we analyzed the temporal order in which whole-brain organization states were visited. Two case-control studies, including in each sample a subgroup scanned a second time after treatment. Chilean sample included 79 patients with a FEP and 83 healthy controls. Mexican sample included 21 antipsychotic-naïve FEP patients and 15 healthy controls. Characteristics of the temporal trajectories between whole-brain functional connectivity meta-states were examined via resting-state functional MRI using elements of network science. We compared the cohorts of cases and controls and explored their differences as well as potential associations with symptoms, cognition, and antipsychotic medication doses. We found that the temporal sequence in which patients' brain dynamics visited the different states was more redundant and segregated. Patients were less flexible than controls in changing their network in time from different configurations, and explored the whole landscape of possible states in a less efficient way. These changes were related to the dose of antipsychotics the patients were receiving. We replicated the relationship with antipsychotic medication in the antipsychotic-naïve FEP sample scanned before and after treatment. We conclude that psychosis is related to a temporal disorganization of the brain's dynamic functional connectivity, and this is associated with antipsychotic medication use.

EyeT4Empathy: Dataset of foraging for visual information, gaze typing and empathy assessment

We present a dataset of eye-movement recordings collected from 60 participants, along with their empathy levels, towards people with movement impairments. During each round of gaze recording, participants were divided into two groups, each one completing one task. One group performed a task of free exploration of structureless images, and a second group performed a task consisting of gaze typing, i.e. writing sentences using eye-gaze movements on a card board. The eye-tracking data recorded from both tasks is stored in two datasets, which, besides gaze position, also include pupil diameter measurements. The empathy levels of participants towards non-verbal movement-impaired people were assessed twice through a questionnaire, before and after each task. The questionnaire is composed of forty questions, extending a established questionnaire of cognitive and affective empathy. Finally, our dataset presents an opportunity for analysing and evaluating, among other, the statistical features of eye-gaze trajectories in free-viewing as well as how empathy is reflected in eye features.

Modeling functional and structural brain‐predicted age in relation to Alzheimer disease and cognition

AbstractBackgroundEstimates of “brain‐predicted age” quantify apparent brain age compared to normative trajectories of neuroimaging features. Brain‐predicted age is elevated in symptomatic Alzheimer disease (AD), but is underexplored in preclinical AD. Prior studies have typically modeled brain‐predicted age with volumetric magnetic resonance imaging (MRI), but other modalities, including resting‐state functional connectivity (FC) and multimodal MRI, have more recently been explored. In the present study, we tested the sensitivity of brain‐age predictions derived from volumetric, FC, and multimodal MRI to group differences in AD status and individual differences in cognition.MethodWe trained three Gaussian process regression models to predict age from (1) FC, (2) regional brain volumes, and (3) multimodal MRI (FC+volumetric) in 493 control participants (cognitively normal, amyloid‐negative [A‐], 18‐89 years old). We applied the trained models to independent samples staged by the presence or absence of amyloid (A+/A‐) and tau (T+/T‐) using PET (Centiloid) and CSF (amyloid β‐42/40 ratio, phosphorylated tau 181): 122 A‐T‐, 62 A+T‐, 62 A+T+, as well as 154 symptomatic AD participants. Global cognition was assessed using an established composite of cognitive tests across multiple domains.ResultAll models successfully predicted age in the control training set, with the multimodal model (mean absolute error [MAE]=5.3 years) outperforming the unimodal volumetric (MAE=6.0 years) and FC models (MAE=8.3 years) (Figure 1). In all three models, the brain age gap (BAG) was significantly elevated in symptomatic AD compared to cognitively normal controls. In the FC and multimodal models, the BAG was significantly reduced in A+T+ participants compared to T‐ controls (Figure 2). In symptomatic AD, lower global cognitive composite scores were associated with elevated volumetric and multimodal BAGs (Figure 3).ConclusionBoth FC and volumetric estimates of brain age are sensitive to symptomatic AD, consistent with previous reports. Additionally, FC‐based brain age may exhibit a biphasic response to preclinical AD pathology while volumetric brain age may be sensitive to cognitive decline in the symptomatic stage. Multimodal estimates of brain age capture each of these modality‐specific relationships, and further, improve sensitivity to healthy age differences. Hence, multimodal neuroimaging models may be useful in staging early symptomatic and preclinical AD.

Anthropomorphic Reaching Movement Generating Method for Human-Like Upper Limb Robot.

How to generate anthropomorphic reaching movement remains a challenging problem in service robots and human motor function repair/reconstruction equipment. However, there is no universally accepted computational model in the literature for reproducing the motion of the human upper limb. In response to the problem, this article presents a computational framework for generating reaching movement endowed with human motion characteristics that imitated the mechanism in the control and realization of human upper limb motions. This article first establishes the experimental paradigm of human upper limb functional movements and proposes the characterization of human upper limb movement characteristics and feature movement clustering methods in the joint space. Then, according to the specific task requirements of the upper limb, combined with the human sensorimotor model, the estimation method of the human upper limb natural postures was established. Next, a continuous task parametric model matching the characteristic motion class is established by using the Gaussian mixture regression method. The anthropomorphic motion generation method with the characteristics of the smooth trajectory and the ability of natural obstacle avoidance is proposed. Finally, the anthropomorphic motion generation method proposed in this article is verified by a human-like robot. The measurement index of the human-likeness degree of the trajectory is given. The experimental results show that for all four tested tasks, the human-likeness degrees were greater than 90.8%, and the trajectories' jerk generated by this method is very similar to the trajectories' jerk of humans, which validates the proposed method.