Trajectory Extraction Research Articles

Augmented Multiple Vehicles’ Trajectories Extraction Under Occlusions With Roadside LiDAR Data

Object occlusion is a common issue in Light Detection and Ranging (LiDAR)-based vehicle tracking technology. The occlusions can cause variance in vehicle location and speed calculation. How to link the vehicle trajectories caused by occlusion issues is a challenge for traffic engineers and researchers. This paper developed an augmented vehicle tracking method under occlusions with the roadside LiDAR data. The proposed method can be divided into two parts. The first part based on the corner point is used to choose a representative vehicle tracking point. And the second part based on the GNN algorithm is employed to link the vehicles’ trajectories under two occlusion situations. The performance of the proposed method has been evaluated using roadside lidar data collected from four different scenarios. The test results showed that more than 89% of disconnected trajectories can be fixed with the proposed method, which is superior compared to the state-of-the-art method. The proposed method can benefit a lot of transportation areas, such as traffic volume count, vehicle speed tracking, and traffic safety analysis.

Deep Learning Techniques for Vehicle Trajectory Extraction in Mixed Traffic

Deep Learning Techniques for Vehicle Trajectory Extraction in Mixed Traffic

Research on Real-Time Anomaly Detection of Fishing Vessels in a Marine Edge Computing Environment

Fishing vessel monitoring systems (VMSs) play an important role in ensuring the safety of fishing vessel operations. Traditional VMSs use a cloud centralized computing model, and the storage, processing, and visualization of all fishing vessel data are completed in the monitoring center. Due to the limitation of maritime communications, the data generated by fishing vessels cannot be fully utilized, and communication delays lead to inadequate warnings in cases of fishing vessel abnormalities. In this paper, we present a real-time anomaly detection model (RADM) for fishing vessels based on edge computing. The model runs in the edge layer, making full use of the information of moving edge nodes and nearby nodes, and combines a historical trajectory extraction detection model with an online anomaly detection model to detect anomalies. The detection model of historical trajectory extraction mines frequent patterns in historical trajectories through multifeature clustering and identifies trajectories that are different from the frequent patterns as anomalies. Online anomaly detection algorithms detect anomalous behavior in specific scenarios based on the spatiotemporal neighborhood similarity and reduce the impact of anomaly evolution. Experiments show that RADM was more effective than traditional methods in real-time anomaly detection of fishing vessels, which provides a new method for upgrading the technology of traditional VMS.

Trajectories through temporal networks

What do football passes and financial transactions have in common? Both are networked walk processes that we can observe, where records take the form of timestamped events that move something tangible from one node to another. Here we propose an approach to analyze this type of data that extracts the actual trajectories taken by the tangible items involved. The main advantage of analyzing the resulting trajectories compared to using, e.g., existing temporal network analysis techniques, is that sequential, temporal, and domain-specific aspects of the process are respected and retained. As a result, the approach lets us produce contextually-relevant insights. Demonstrating the usefulness of this technique, we consider passing play within association football matches (an unweighted process) and e-money transacted within a mobile money system (a weighted process). Proponents and providers of mobile money care to know how these systems are used—using trajectory extraction we find that 73% of e-money was used for stand-alone tasks and only 21.7% of account holders built up substantial savings at some point during a 6-month period. Coaches of football teams and sports analysts are interested in strategies of play that are advantageous. Trajectory extraction allows us to replicate classic results from sports science on data from the 2018 FIFA World Cup. Moreover, we are able to distinguish teams that consistently exhibited complex, multi-player dynamics of play during the 2017–2018 club season using ball passing trajectories, coincidentally identifying the winners of the five most competitive first-tier domestic leagues in Europe.

Vehicle Trajectory Extraction in Traffic Scene Based on Monocular Vision

Accurately extracting the trajectory information of multiple vehicles in urban traffic videos is of great significance in applications. It can help analyze and predict drivers’ behavior, count road flow, and restore accident scenarios. However, due to the complexity of the interaction between vehicles and the environment, it is still a challenging problem to accurately extract the trajectory of multi-target vehicles in urban traffic scenes, which often requires the assistance of a variety of different devices, which increases the practicality and difficulty of application. A method is proposed based on the combination of detection and tracking. This method only needs to use a video stream captured by a fixed panoramic camera and the coordinate information of several feature points in the scene to extract the speed and trajectory information of the vehicle in the scene. Different optimization methods are used in each link to reduce errors, and build a simulation environment for the test scenario, reconstruct and verify the extracted trajectory information in the simulation environment. Experiments have proved that the effect of the method is basically satisfactory. It has the advantages of fewer constraints, convenient deployment, and accurate restoration while ensuring a certain real-time performance, which has practical engineering significance.

Optical analysis of facial nerve degeneration in Bell’s palsy

This paper is an initial proof of concept for an optical speckled-based method for the evaluation of facial nerve paralysis. Differences between the affected and the healthy sides of the face were measured in patients with Bell's palsy—a peripheral facial nerve paralysis. The patients’ faces were illuminated with two symmetrical spots on their nasolabial folds and the reflected speckle patterns were analyzed. Muscle activity was evaluated by muscle tone contraction-release motion inducing associated skin tilting movements. The skin movements were imaged with a defocused lens, which enables extraction of the speckle pattern's time changing trajectory. We found an asymmetry ratio expression that may be the key for the estimation of the degeneration level in Bell's palsy.

Ship behavior prediction via trajectory extraction-based clustering for maritime situation awareness

This study presents a method in which historical AIS data are used to predict the future trajectory of a selected vessel. This is facilitated via a system intelligence-based approach that can be subsequently utilized to provide enhanced situation awareness to navigators and future autonomous ships, aiding proactive collision avoidance. By evaluating the historical ship behavior in a given geographical region, the method applies machine learning techniques to extrapolate commonalities in relevant trajectory segments. These commonalities represent historical behavior modes that correspond to the possible future behavior of the selected vessel. Subsequently, the selected vessel is classified to a behavior mode, and a trajectory with respect to this mode is predicted. This is achieved via an initial clustering technique and subsequent trajectory extraction. The extracted trajectories are then compressed using the Karhunen–Loéve transform, and clustered using a Gaussian Mixture Model. The approach in this study differs from others in that trajectories are not clustered for an entire region, but rather for relevant trajectory segments. As such, the extracted trajectories provide a much better basis for clustering relevant historical ship behavior modes. A selected vessel is then classified to one of these modes using its observed behavior. Trajectory predictions are facilitated using an enhanced subset of data that likely correspond to the future behavior of the selected vessel. The method yields promising results, with high classification accuracy and low prediction error. However, vessels with abnormal behavior degrade the results in some situations, and have also been discussed in this study.

Precise seam tracking in robotic welding by an improved image processing approach

The intelligent welding robots have been widely developed and applied into the industrial production lines to improve the manufacturing efficiency. The main and important factor to improve the robotic welding performance, quality, and intelligence is the seam recognition accuracy. Higher industrial production and demands of higher precision in gas metal arc welding (GMAW) process have increased robot use. Vision feedback in robotic welding provides flexibility, productivity, and precision and promotes vaster range of applicability. In this case, machine vision techniques that are widely used to identify and locate the welding seam allow planning an industrial robot path. Previous investigations have proposed various approaches to address the precise tracking of welding seam curve. In majority of cases, the simultaneous uses of laser-based and optical lighting approaches to determine the location of the seam were suggested. However, complex seams remain a big challenge in robotic welding, and no definite way has been proposed to produce a precise robot path. This paper aims to develop a comprehensive robot trajectory extraction method for complex welding seam through processing of noisy images and laser data. Since the structured light on laser basis and stereo vision propose more possibility in specifying the place of the seam, this study considers a two-source processing for GMAW. A new improved edge detection algorithm is also proposed to enclose the weld zone and determine the seam and the pool parameters. To evaluate the performance of this method, experimenting different types of work pieces under random welding conditions are carried out, and the results are verified. A considerable improvement in speed and detection of error between the measured and real seam coordinates is reported.

Video-based eye tracking performance for computer-assisted diagnostic support of diabetic neuropathy

Diabetes is currently one of the major public health threats. The essential components for effective treatment of diabetes include early diagnosis and regular monitoring. However, health-care providers are often short of human resources to closely monitor populations at risk. In this work, a video-based eye-tracking method is proposed as a low-cost alternative for detection of diabetic neuropathy. The method is based on the tracking of the eye-trajectories recorded on videos while the subject follows a target on a screen, forcing saccadic movements. Upon extraction of the eye trajectories, representation of the obtained time-series is made with the help of heteroscedastic ARX (H-ARX) models, which capture the dynamics and latency on the subject's response, while features based on the H-ARX model's predictive ability are subsequently used for classification. The methodology is evaluated on a population constituted by 11 control and 20 insulin-treated diabetic individuals suffering from diverse diabetic complications including neuropathy and retinopathy. Results show significant differences on latency and eye movement precision between the populations of control subjects and diabetics, while simultaneously demonstrating that both groups can be classified with an accuracy of 95%. Although this study is limited by the small sample size, the results align with other findings in the literature and encourage further research.

Research on Ship Trajectory Extraction Based on Multi-Attribute DBSCAN Optimisation Algorithm

Abstract With the vigorous development of maritime traffic, the importance of maritime navigation safety is increasing day by day. Ship trajectory extraction and analysis play an important role in ensuring navigation safety. At present, the DBSCAN (density-based spatial clustering of applications with noise) algorithm is the most common method in the research of ship trajectory extraction, but it has shortcomings such as missing ship trajectories in the process of trajectory division. The improved multi-attribute DBSCAN algorithm avoids trajectory division and greatly reduces the probability of missing sub-trajectories. By introducing the position, speed and heading of the ship track point, dividing the complex water area and vectorising the ship track, the function of guaranteeing the track integrity can be achieved and the ship clustering effect can be better realised. The result shows that the cluster fitting effect reaches up to 99.83%, which proves that the multi-attribute DBSCAN algorithm and cluster analysis algorithm have higher reliability and provide better theoretical guidance for the analysis of ship abnormal behaviour.

Automatic velocity picking from semblances with a new deep-learning regression strategy: Comparison with a classification approach

The physical basis, parameterization, and assumptions involved in root-mean-square (rms) velocity estimation have not significantly changed since they were first developed. However, these three aspects are all good targets for novel application of the recent emergence of machine learning (ML). Therefore, it is useful at this time to provide a tutorial overview of two state-of-the-art ML implementations; we have designed and evaluated classification and regression neural networks for the extraction of apparent rms velocity trajectories from semblance data. Both networks share a similar end-to-end trainable structure, except for the final layer. In the classification network, the velocity picking is performed by finding the largest amplitude trajectory through all velocity bins. The regression network, on the other hand, applies a differentiable soft-argmax function that converts the feature maps directly to apparent rms velocity values as functions of traveltime. Relative confidence maps can also be estimated from both neural networks. A large number of synthetic models with horizontal layers are created, and common-midpoint gathers are simulated from those models as training samples. Transfer learning is applied to fine-tune the networks with a small number of samples for testing with synthetic and field data from more complicated (2D) models. Tests using synthetic data show that the regression and classification networks can give reasonable velocity predictions from semblances, but the regression network gives higher accuracy.

Self-consistent theory of mobility edges in quasiperiodic chains

We introduce a self-consistent theory of mobility edges in nearest-neighbour tight-binding chains with quasiperiodic potentials. Demarcating boundaries between localised and extended states in the space of system parameters and energy, mobility edges are generic in quasiperiodic systems which lack the energy-independent self-duality of the commonly studied Aubry-Andr\'e-Harper model. The potentials in such systems are strongly and infinite-range correlated, reflecting their deterministic nature and rendering the problem distinct from that of disordered systems. Importantly, the underlying theoretical framework introduced is model-independent, thus allowing analytical extraction of mobility edge trajectories for arbitrary quasiperiodic systems. We exemplify the theory using two families of models, and show the results to be in very good agreement with the exactly known mobility edges as well numerical results obtained from exact diagonalisation.

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.

Video-SAR Imaging of Dynamic Scenes Using Low-Rank and Sparse Decomposition

With the goal of persistent surveillance over a scene of interest, this paper proposes an approach for joint imaging and trajectory extraction of multiple moving objects, using circular video-SAR (ViSAR) mode. Unlike stationary SAR imaging where the phase perturbation of the received signal arises from the platform motion, the moving targets impose additional phase errors, contributing to defocusing of the moving targets’ images. Here, for proper imaging of the moving targets, specifically in low-signal-to-clutter-ratio (SCR) scenarios, their signatures are decomposed from the stationary background clutter by low rank and sparse decomposition (LRSD), before refocusing to their original positions. However, the rotation of the scene, which emerges as a consequence of circular SAR geometry, leads to undesirable decomposition results via LRSD. To facilitate the applicability of LRSD for moving target extraction in our setting, considering this rotation in our model, the decomposition is accomplished while this undesirable effect is automatically compensated for. In addition, to further guarantee that LRSD yields satisfactory decomposition results, phase errors due to platform motion are also jointly corrected along with the clutter separation process. Then, having decomposed the sequential sparse images of moving targets’ signatures, the additional phase errors caused by moving targets are compensated to refocus them to their original positions. After that, these focused sparse images of moving targets are combined to construct a single image where the trajectories of moving targets can be observed. Using this image, a composite image is also constructed, including both the moving objects’ trajectories and the stationary background, which can be used for target tracking applications. Through extensive experimental results we show the effectiveness of the proposed method on both synthetic and real SAR images.

A Vehicle Trajectory Extraction Method for Traffic Simulating Modeling

Vehicle driving behavior in urban traffic environment is interactive and complex. In order to simulate the driving behavior of real vehicles, it needs to rely on a large number of real vehicle trajectory data. However, making the simulated model more consistent with the real situation, a large vehicle trajectory dataset is needed. Therefore, this paper proposes a method to automatically extract accurate vehicle trajectory from video captured by UAV camera. The proposed computer vision method obtains the trajectory by detecting the vehicle and then tracking the trajectory, and calculates the relevant parameters of the vehicle, such as position, speed and direction. Therefore, the extracted data can be used as real driving behavior data for simulation modeling. The experimental results show that the proposed method is effective and reliable.

A Stacked Denoising Autoencoder and Long Short-Term Memory Approach With Rule-Based Refinement to Extract Valid Semantic Trajectories

Indoor location-based services have been widely investigated to take advantage of semantic trajectories for providing user oriented services in indoor environments. Although indoor semantic trajectories can provide seamless understanding to users regarding the provided location-based services, studies on the application of deep learning approaches for robust and valid semantic indoor localization are lacking. In this study, we combined a stacked denoising autoencoder and long short term memory technique with a rule-based refinement method applying a rule-based hidden Markov model (HMM) to perform robust and valid semantic trajectory extraction. In particular, our rule-based HMM approach incorporates a direct set of rules into HMM to resolve invalid movements of the extracted semantic trajectories and is extensible to various deep learning techniques. We compared the performance of our proposed approach with that of other cutting-edge deep learning approaches on two different real-world data sets. The experimental results demonstrate the feasibility of our proposed approach to produce more robust and valid semantic trajectories.

MIFF

Human Mobility Extraction with cellular Signaling Data (SD) is essential for human mobility understanding, epidemic control, and wireless network planning. SD log the detailed interactions between cellphones and cellular towers, but suffer from a spatio-temporal uncertainty problem due to cellular network tower-level load rebalancing (switching users between towers) and cellphone usage activities. To date, most models focus on utilizing better data like RSSI or GPS, do not directly address uncertainty. To address the SD uncertainty issue, we utilize two insights based on (i) individuals' regular mobility patterns and (ii) common co-movement mobility patterns between cellphone users as suggested by fundamental human mobility nature. Accordingly, we design a Multi-Information Fusion Framework (MIFF) to assist in extracting road-level human mobility based on cell-tower level traces. To evaluate the effectiveness of MIFF, we conduct experiments on one-month SD obtained from a cellular service operator, and SD manually collected by handheld mobile devices in two cities in China. Four transportation modes, namely railways, cars, buses, and bikes are evaluated. Experimental results show that with MIFF, our road-level trajectory extraction accuracy can be improved by 5.0% on Point correct matching index and 68.5% on Geographic Error on average.

Estimating Micro-Level On-Road Vehicle Emissions Using the K-Means Clustering Method with GPS Big Data

Due to the advanced spatial data collection technologies, the locations of vehicles on roads are now being collected nationwide, so there is a demand for applying a micro-level emission calculation methods to estimate regional and national emissions. However, it is difficult to apply this method due to the low data collection rate and the complicated calculation procedure. To solve these problems, this study proposes a vehicle trajectory extraction method for estimating micro-level vehicle emissions using massive GPS data. We extracted vehicle trajectories from the GPS data to estimate the emission factors for each link at a specific time period. Vehicle trajectory data was divided into several groups through a k-means clustering method, in which the ratios of each operating mode were used as variables for clustering similar vehicle trajectories. The results showed that the proposed method has an acceptable accuracy in estimating emissions. Furthermore, it was also confirmed that the estimated emission factors appropriately reflected the driving characteristics of links. If the proposed method were utilized to update the link-based micro-level emission factors using continuously accumulated trajectory data for the road network, it would be possible to efficiently calculate the regional- or national-level emissions only using traffic volume.

An Improved Near-Field Computer Vision for Jet Trajectory Falling Position Prediction of Intelligent Fire Robot.

An improved Near-Field Computer Vision (NFCV) system for intelligent fire robot was proposed that was based on our previous works in this paper, whose aims are to realize falling position prediction of jet trajectory in fire extinguishing. Firstly, previous studies respecting the NFCV system were briefly reviewed and several issues during application testing were analyzed and summarized. The improved work mainly focuses on the segmentation and discrimination of jet trajectory adapted to complex lighting environment and interference scenes. It mainly includes parameters adjustment on the variance threshold and background update rate of the mixed Gaussian background method, jet trajectory discrimination based on length and area proportion parameters, parameterization, and feature extraction of jet trajectory based on superimposed radial centroid method. When compared with previous works, the proposed method reduces the average error of prediction results from 1.36 m to 0.1 m, and the error variance from 1.58 m to 0.13 m. The experimental results suggest that every part plays an important role in improving the functionality and reliability of the NFCV system, especially the background subtraction and radial centroid methods. In general, the improved NFCV system for jet trajectory falling position prediction has great potential for intelligent fire extinguishing by fire-fighting robots.

Longitudinal-Scanline-Based Arterial Traffic Video Analytics with Coordinate Transformation Assisted by 3D Infrastructure Data

High-resolution vehicle trajectory data can be used to generate a wide range of performance measures and facilitate many smart mobility applications for traffic operations and management. In this paper, a Longitudinal Scanline LiDAR-Camera model is explored for trajectory extraction at urban arterial intersections. The proposed model can efficiently detect vehicle trajectories under the complex, noisy conditions (e.g., hanging cables, lane markings, crossing traffic) typical of an arterial intersection environment. Traces within video footage are then converted into trajectories in world coordinates by matching a video image with a 3D LiDAR (Light Detection and Ranging) model through key infrastructure points. Using 3D LiDAR data will significantly improve the camera calibration process for real-world trajectory extraction. The pan-tilt-zoom effects of the traffic camera can be handled automatically by a proposed motion estimation algorithm. The results demonstrate the potential of integrating longitudinal-scanline-based vehicle trajectory detection and the 3D LiDAR point cloud to provide lane-by-lane high-resolution trajectory data. The resulting system has the potential to become a low-cost but reliable measure for future smart mobility systems.