Target Motion Characteristics Research Articles

Occluded vehicle tracking by integrating full-scale features and trajectory correction

In order to improve the tracking drift and identity switching (IDs) problems caused by occlusion in vehicle tracking, a method for tracking occluded vehicles that integrates full-scale features and trajectory correction is proposed based on the Deep SORT algorithm. Firstly, a full-scale feature extraction network is introduced to extract the features of different scales of the target and realize adaptive dynamic fusion to enhance the target appearance characteristics. Then, a trajectory correction algorithm is proposed to repair the tracking trajectory during occlusion, and the Kalman filter parameters are updated to reduce the accumulated linear error during occlusion and optimize the target motion characteristics. Finally, the occluded vehicle is tracked by combining appearance features and motion features. The feasibility of the proposed method is verified by ablation experiments and visualization analysis. The experimental results on the KITTI dataset show that compared with the existing typical methods, the proposed method achieves the highest comprehensive score of 78.13 and the lowest number of IDs of 192, which effectively improves the IDs problem in occluded vehicle tracking and improves the robustness of vehicle tracking.

Design of Adaptive Target Tracking Algorithm for Robots Based on Visual Attention Mechanism

Adaptive target tracking with visual attention represents a sophisticated approach to object detection and localization in dynamic environments. With principles inspired by human visual perception, this methodology employs mechanisms of selective attention to prioritize relevant visual information for tracking moving targets. By dynamically adjusting attentional focus based on salient visual cues and target motion characteristics, adaptive target tracking enhances the efficiency and accuracy of object localization in cluttered scenes. This research presents a novel adaptive target tracking algorithm designed for robotic systems, integrating a visual attention mechanism with the Fuzzy Clustering Multi-Point Tracking utilizing the Green Channel (FC-MPT-GC) approach. The proposed FC-MPT-GC model comprises of Fuzzy Clustering for the extraction of features in the robots-based environment. The FC-MPT-GC model uses the estimation of green channels in the classification environment. With the estimation of features in the environment with Fuzzy C-means clustering green channels are deployed in the deep learning, The proposed algorithm aims to enhance the adaptability and precision of target tracking in dynamic environments. By incorporating a visual attention mechanism, the algorithm dynamically allocates attentional focus to salient regions of the visual input, optimizing the tracking process for moving targets. The FC-MPT-GC methodology further refines target localization by utilizing fuzzy clustering and multi-point tracking strategies, particularly leveraging information from the Green Channel to improve robustness in various lighting conditions. Simulation analysis demonstrated that the proposed FC-MPT-GC model tracking accuracy is achieved at 95.1% with the minimal computation time of 15.2 ms.

HELOP: Multi-target tracking based on heuristic empirical learning algorithm and occlusion processing

Multi-target tracking is one of the important fields in computer vision, which aims to solve the problem of matching and correlating targets between adjacent frames. In this paper, we propose a fine-grained track recoverable (FGTR) matching strategy and a heuristic empirical learning (HEL) algorithm. The FGTR matching strategy divides the detected targets into two different sets according to the distance between them, and adopts different matching strategies respectively, in order to reduce false matching, we evaluated the trust degree of the target’s appearance feature information and location feature information, adjusted the proportion of the two reasonably, and improved the accuracy of target matching. In order to solve the problem of trajectory drift caused by the cumulative increase of Kalman filter error during the occlusion process, the HEL algorithm predicts the position information of the target in the next few frames based on the effective information of other previous target trajectories and the motion characteristics of related targets. Make the predicted trajectory closer to the real trajectory. Our proposed method is tested on MOT16 and MOT17, and the experimental results verify the effectiveness of each module, which can effectively solve the occlusion problem and make the tracking more accurate and stable.

Improved ship imaging algorithm for geosynchronous synthetic aperture radar (SAR) featured by subaperture time window selection optimization and inverse SAR translational compensation

An inverse synthetic aperture radar processing-based algorithm is improved for ship imaging in geosynchronous synthetic aperture radars (GEO SARs) to solve the technical challenges including long accumulation time and complex target motion characteristics. The improvement on the existing algorithm mainly involves four aspects. First, the formula for calculating the appropriate time length of subaperture is derived, according to the principle that the accumulated power of the ship’s echo corresponding to the subaperture duration ensures that the signal-to-noise ratio satisfies the requirement of GEO SARs. The derived formula solves the problem that how long the subaperture length should be taken with regards to a long accumulation time of GEO SAR when imaging ships. Second, we propose an optimization criterion for determining the appropriate subaperture time window by ensuring that the maximum number of ship pixels is detected in the subaperture image. The proposed criterion is used to optimize the location of the subaperture time window in the long accumulation time. Third, the SPECAN algorithm used in the scene imaging process is replaced by an improved nonlinear chirp scaling (NLCS) algorithm to improve the imaging accuracy of the scene. Our improvement on the traditional NLCS algorithm focuses on improving precisions of some of the formulas in the traditional algorithm. Furthermore, a translational compensation algorithm is proposed by considering the influence of spatial variance of slant range history of GEO SARs. The proposed translational compensation algorithm improves the compensation accuracy since the spatial variance of different scatterers in the scene is considered. A series of simulation experiments related with ship imaging in GEO SARs were carried out to verify the correctness of the proposed algorithm and the improvement on the imaging quality compared with that of the existing algorithm.

A Phase-Derived Velocity Measurement Method Based on the Generalized Radon–Fourier Transform With a Low SNR

The phase-derived velocity measurement (PDVM) technique can achieve a high measurement accuracy at the phase level and thus has great application prospects in the field of micromotion feature extraction and target recognition. To achieve a PDVM with a low signal-to-noise ratio (SNR), a PDVM method based on the generalized Radon–Fourier transform (GRFT) is proposed in this article. The main challenges that we overcome are phase extraction and phase ambiguity resolving under the condition of a low SNR. By utilizing the GRFT to estimate the target motion parameters, the echo peak position can be reconstructed, and then the peak phase value can be extracted. In the meantime, the phase ambiguity integer can be resolved based on the rough velocity estimation results obtained by the GRFT, and the phase ambiguity resolving can be realized at a low SNR. In addition, to suppress the influence of noise on the extracted phase, a filter design method based on the target motion characteristics is proposed to further improve the accuracy of the PDVM. In the simulation, the performance of the proposed method under different motion models and different SNR conditions is analyzed, and the effectiveness of the proposed method under low-SNR conditions is verified. Compared with directly using the GRFT, the proposed method has the advantages of strong applicability to different motion models and low computational load.

Fly-Around Control of Space Tumbling Target Under Multiple Constraints

Aiming at the practical problems of external unknown disturbance and internal modeling uncertainty, when space tumbling target flies around in close range, a sliding mode controller (SMC) based on active disturbance rejection control (ADRC) technology is proposed to realize real-time estimation and compensation of “total disturbance.” Firstly, according to the motion characteristics of space tumbling target, the relative motion equation in rotating line of sight (RLOS) coordinate system is established; Secondly, the compound controller is designed, and the convergence of the nonlinear state expansion observer and the stability of the closed-loop system are analyzed based on the root locus method and Lyapunov function method respectively; Finally, the simulation results show that the SMC based on ADRC technology can effectively suppress the disturbance and overcome the chattering problem of traditional sliding mode controller. It has a good control quality, and strong robustness is an easy method for engineering practice.

Small Target Tracking in Satellite Videos Using Background Compensation

Through the use of video technology, satellites can detect dynamic targets and analyze their motion characteristics. Target tracking can extract dynamic information about key ground targets for target monitoring and trajectory prediction by satellite video. Tracking algorithms are affected by target motion characteristics, such as velocity and direction, as well as background characteristics, such as illumination changes, occlusion, and background similarities with the target. However, these problems are seldom studied with satellite video cameras. Current algorithms are unsuitable for satellite video because of the poor texture and color features of the target in satellite video. Therefore, in this article, we enhance target tracking for satellite video technology using two aspects: 1) sample training strategy and 2) sample characterization. We establish a filter training mechanism for the target and background to improve the discrimination ability of the tracking algorithm. We then build a target feature model using a Gabor filter to enhance the contrast between the target and background. Moreover, we propose a tracking state evaluation index to avoid tracking drift. Tracking experiments using nine sets of Jilin-1 satellite videos show that the proposed approach can accurately locate a target under weak feature attributes. Therefore, this article contributes to more robust tracking using satellite video technology.

Space Target Detection in Complicated Situations for Wide-Field Surveillance

Long exposure time and wide field can effectively improve the ability of a space surveillance telescope to detect faint space targets. However, complicated situations pose challenges for space target detection. Background star images usually manifest a rotated streak, and target trajectories can be crossed, discontinuous, or nonlinear. This paper presents an accurate and robust space target detection method, namely, spatiotemporal pipeline multistage hypothesis testing (SPMHT), to overcome the issues. Specifically, the method includes the following two stages: First, in the spatiotemporal pipeline filtering step, Spatiotemporal-related Intersection over Union (SrIoU) is used to calculate the IoU score instead of the traditional method. Benefiting from the differences between motion characteristics of targets and stars and the insensitivity of the SrIoU score to the noise, the spatiotemporal pipeline filtering can effectively remove the streak images of background stars and obtain candidate points. Second, a series of candidate points is further organized into a tree structure. We pruned in the tree structure combined with these candidate trajectories by using velocity and direction feature of moving objects. Furthermore, in the search step, fast adaptive sequence region search is used to reduce the computational cost. The experimental results for two datasets, simulated image datasets and real captured image datasets, demonstrate the effectiveness in addressing the difficulties of space target detection in complicated situations.

Review on interferometric ISAR 3D imaging: Concept, technology and experiment

Abstract This paper presents an overview of Interferometric inverse synthetic aperture radar (InISAR) three-dimensional (3D) imaging according to the presented research chain. InISAR applies interferometric technology of multi antenna to inverse synthetic aperture radar (ISAR). It can realize 3D imaging of long distance non-cooperative targets, under all weather conditions. Hence it has wide range of applications, especially in military and civilian applications. In InISAR imaging, interferometric processing is applied to the two obtained ISAR images of each baseline, achieving the projection coordinates of the scattering centers along the baseline direction, which are denoted as x and y coordinates in this paper. Then, the 3D imaging result is established combined with the radial information, which is denoted as z coordinate in this paper. The InISAR 3D image is directly consistent with the physical size of the target. It can reflect the constant feature that is not affected by the target motion characteristics, effectively improving the target recognition ability. On the basis of the basic principle of InlSAR 3D imaging, this paper reviews and summarizes the development and research status of InISAR 3D imaging, focuses on the technical difficulties and main solutions, and points out the direction of further research in view of the existing problems.

Performance assessment of a programmable five degrees-of-freedom motion platform for quality assurance of motion management techniques in radiotherapy.

Inter-fraction and intra-fraction motion management methods are increasingly applied clinically and require the development of advanced motion platforms to facilitate testing and quality assurance program development. The aim of this study was to assess the performance of a 5 degrees-of-freedom (DoF) programmable motion platform HexaMotion (ScandiDos, Uppsala, Sweden) towards clinically observed tumor motion range, velocity, acceleration and the accuracy requirements of SABR prescribed in AAPM Task Group 142. Performance specifications for the motion platform were derived from literature regarding the motion characteristics of prostate and lung tumor targets required for real time motion management. The performance of the programmable motion platform was evaluated against (1) maximum range, velocity and acceleration (5 DoF), (2) static position accuracy (5 DoF) and (3) dynamic position accuracy using patient-derived prostate and lung tumor motion traces (3 DoF). Translational motion accuracy was compared against electromagnetic transponder measurements. Rotation was benchmarked with a digital inclinometer. The static accuracy and reproducibility for translation and rotation was <0.1mm or <0.1°, respectively. The accuracy of reproducing dynamic patient motion was <0.3mm. The motion platform's range met the need to reproduce clinically relevant translation and rotation ranges and its accuracy met the TG 142 requirements for SABR. The range, velocity and acceleration of the motion platform are sufficient to reproduce lung and prostate tumor motion for motion management. Programmable motion platforms are valuable tools in the investigation, quality assurance and commissioning of motion management systems in radiation oncology.

Effects of visual motion consistent or inconsistent with gravity on postural sway.

Vision plays an important role in postural control, and visual perception of the gravity-defined vertical helps maintaining upright stance. In addition, the influence of the gravity field on objects' motion is known to provide a reference for motor and non-motor behavior. However, the role of dynamic visual cues related to gravity in the control of postural balance has been little investigated. In order to understand whether visual cues about gravitational acceleration are relevant for postural control, we assessed the relation between postural sway and visual motion congruent or incongruent with gravity acceleration. Postural sway of 44 healthy volunteers was recorded by means of force platforms while they watched virtual targets moving in different directions and with different accelerations. Small but significant differences emerged in sway parameters with respect to the characteristics of target motion. Namely, for vertically accelerated targets, gravitational motion (GM) was associated with smaller oscillations of the center of pressure than anti-GM. The present findings support the hypothesis that not only static, but also dynamic visual cues about direction and magnitude of the gravitational field are relevant for balance control during upright stance.

Infrared Dim and Small Targets Detection Method Based on Local Energy Center of Sequential Image

In order to detect infrared (IR) dim and small targets in a strong clutter background, a method based on local energy center of sequential image is proposed. This paper began by using improved anisotropy for background prediction (IABP), followed by target enhancement by improved high‐order cumulates (HOC). Finally, on the basis of image preprocessing, the paper constructs a sequential image energy center detection algorithm that integrates the neighborhood, continuity, area, and energy and other motion characteristics of the target. Experiments showed that the improved anisotropic background predication could be loyal to the true background of the original image to the maximum extent, presenting a superior overall performance to other background prediction methods; the improved HOC significantly increased the signal‐noise ratio of images; when the signal‐noise ratio (SNR) is lower than 2.5 dB, the proposed method could effectively eliminate noise and detect targets.

Fractional-order information in the visual control of lateral locomotor interception.

Previous work on locomotor interception of a target moving in the transverse plane has suggested that interception is achieved by maintaining the target's bearing angle (often inadvertently confused and/or confounded with the target heading angle) at a constant value. However, dynamics-based model simulations testing the veracity of the underlying control strategy of nulling the rate of change in the bearing angle have been restricted to limited conditions of target motion, and only a few alternatives have been considered. Exploring a wide range of target motion characteristics with straight and curving ball trajectories in a virtual reality setting, we examined how soccer goalkeepers moved along the goal line to intercept long-range shots on goal, a situation in which interception is naturally constrained to movement along a single dimension. Analyses of the movement patterns suggested reliance on combinations of optical position and velocity for straight trajectories and optical velocity and acceleration for curving trajectories. As an alternative to combining such standard integer-order derivatives, we demonstrate with a simple dynamical model that nulling a single informational variable of a self-tuned fractional (rather than integer) order efficiently captures the timing and patterning of the observed interception behaviors. This new perspective could fundamentally change the conception of what perceptual systems may actually provide, both in humans and in other animals. (PsycINFO Database Record

Particle-filter Multi-target Tracking Algorithm Based on Dynamic Salient Features

In order to address the problem of tracking different moving targets in image sequence against a complicated background, this paper presents a particle-filter multi-target tracking algorithm based on their dynamic salient features. By making use of the research findings on visual attention, the algorithm adopts the robust dynamic salient features as a result of combining the gray-scale and details with the motion characteristics of such targets as the state vector of particle filter. The algorithm is highly robust as it contains salient features originating from the low-level features of the targets. Meanwhile the particle filter allows optimized estimation of non-linear and non-Gaussian models. As a consequence, the algorithm is capable of managing traces in tracking different targets and dealing with their appearance, disappearance, mergence, splitting and sheltering by obstacles. Experiments show that this new algorithm enables tracking of multiple targets in complicated image sequence.

Air Ship Missile Carrier Aircraft Attack Array Model Research

This paper studies air defense system kill zone, air ship missile firing mode and air ship missile launch area that impact air ship missile load matrix determine key problems, establishes air ship missile carrier array’s mathematics model, makes simulation calculation of the model, and analyzes simulation results, and gets the influence law of vehicle speed, height, target motion characteristics, deviation angle and pitching angle on attack matrix, and the results can provide reference for determining air ship missile carrier aircraft attack array.

Joint parameter estimation and target localization for bistatic MIMO radar system in impulsive noise

This paper presents a novel method for joint parameter estimation of multiple targets for bistatic multiple-input multiple-output radar system in impulsive noise. Firstly, a new signal array model is constructed according to a 3-D motion characteristics of target in impulsive noise environment. Secondly, Doppler frequency parameters are jointly estimated by peak searching of a fractional lower-order ambiguity function based on fractional Fourier transform (FLOS-FAF) and a dechirping method. Furthermore, two modified algorithms are proposed for the estimation of direction-of-departure and direction-of-arrival, including the MUSIC algorithm based on the fractional lower-order ambiguity function (FF-MUSIC) and the ESPRIT algorithm based on the fractional lower-order ambiguity function (FF-ESPRIT) in the fractional Fourier transform domain. Simulation results are presented to verify the effectiveness of the proposed method.

Smear signature morphology of surface targets with arbitrary motion in spotlight synthetic aperture radar imagery

This study develops a methodology for analytically predicting the detailed smear signature morphology of surface targets with arbitrary motion in spotlight synthetic aperture radar (SAR) imagery. The radar sensor is assumed to move with constant speed and heading on a level flight path with broadside imaging geometry. Cases of uniform target motion exhibit morphology of simply curved smear shapes, as has been reported previously. However, the present analysis shows that non-uniform target motion can cause complicated smear shapes in spotlight SAR imagery. Specifically, the method of stationary phase is applied to local subaperture SAR images to yield analytic non-parametric expressions for the morphology of smear signatures within spotlight SAR imagery for surface targets with arbitrary motion trajectories. This predictive capability offers the potential of extracting some characteristics of true target motion based upon the induced SAR smear signatures alone.

Fire flame detection based on GICA and target tracking

To improve the video fire detection rate, a robust fire detection algorithm based on the color, motion and pattern characteristics of fire targets was proposed, which proved a satisfactory fire detection rate for different fire scenes. In this fire detection algorithm: (a) a rule-based generic color model was developed based on analysis on a large quantity of flame pixels; (b) from the traditional GICA (Geometrical Independent Component Analysis) model, a Cumulative Geometrical Independent Component Analysis (C-GICA) model was developed for motion detection without static background and (c) a BP neural network fire recognition model based on multi-features of the fire pattern was developed. Fire detection tests on benchmark fire video clips of different scenes have shown the robustness, accuracy and fast-response of the algorithm.

TH‐C‐BRC‐09: Tracking Moving Tumors with a Scanned Carbon Beam: Robustness to Changing Target Motion Characteristics and Tracking Uncertainties

Purpose: Beam tracking of moving tumors with scanned carbon ion radiotherapy achieves uniform target dose by adapting carbon pencil beams to follow moving tissues using real‐time lateral magnetic deflection and range modulation. The purpose of the current study was to establish the robustness of target dose coverage that can be achieved in the presence of changing target motion characteristics and uncertainties in the motion tracking system. Methods: A 4D research treatment planning system code for ion therapy was extended to compute the resulting target dose coverage when various random and systematic uncertainties were introduced to ideal tracking dose simulations. Treatment plans were prepared using the in‐ house code for a hypothetical sphere target volume with sinusoidal motion in a water phantom. Uncertainties were simulated by shifting pencil beam coordinates and modifying target motion signals. To simulate imperfect tracking, we considered variations to the following: (1) respiratory period, (2) starting respiratory phase, (3) target position drift, (4) random tracking offsets, (5) range tracking wedge acceleration, and (6) external respiratory motion signal phase delay. We also considered imperfect tracking with combinations of varying target motion characteristics and tracking uncertainties. A total of 13713 dose distributions were computed. For all cases, mean, standard deviation, minimum, and maximum of target dose were analyzed. Results: For perfect tracking of the moving sphere, mean and standard deviation of target dose were 98.2 and 0.8 % of the prescription dose, respectively. Of the variations considered, systematic drifts of target position of up to 5 mm had the most significant effect on target coverage ((6.6 ± 20.3) % decrease), and range tracking wedge accelerations above 1.0 g had the least effect on target dose ((0.0 ±1.1) %). Conclusions: The results provide insight into beam tracking precision requirements and the need to monitor and mitigate changes in patient motion.Work supported by research grant from the Deutscher Akademischer Austausch Dienst, Bonn, Germany.

TU‐G‐BRC‐03: A Novel Markerless Technique for Real‐Time Lung Tumor Tracking

Purpose: A critical component missing in the radiotherapy process for lung cancer is real‐time monitoring of lung tumors. The large planning margin that is typically required to account for rumor motion compromises sparing of critical structures and normal tissues. A novel markerless method has been developed to monitor lung tumor motion during treatment.Methods: The method automatically identifies lung tumors directly on continuous MV treatment beam images. Prior information from planning CT and contours is used to generate three sets of DRRs for lung tumors, other moving organs, and static organs, respectively. Composite DRRs are produced by combining three sets of DRRs, with weighting factors assigned to compensate for different attenuation coefficients between MV and kV x‐ray energies. Individual locations of three set of DRR are optimized by comparing with acquired MV treatment beam images using a normalized cross correlation algorithm. The method was evaluated by monitoring a target moving inside a chest cavity phantom. It was also applied retrospectively to online MV images from lung cancer patients. Results: In phantom studies, the target and target motion characteristics were successfully located within seven different treatment fields; the detected target locations have a maximum deviation of 0.8 mm from expected locations. Lung tumors in lung patients were successfully tracked. Although the planning 4D CT indicated that the lung tumor motion range was within 0 mm, real‐time tracking suggested that: a) the lung tumor motion pattern is not regular; and b) the average tumor locations, the respiratory frequencies, and the tumor motion amplitude (range) vary considerably during treatment. Conclusions: This tracking method, which requires no implanted markers and can be performed with no additional imaging dose, will enable clinicians to reduce margins and safely deliver higher doses by accounting for the effects of displacement and motion during treatment.