Automatic Object Tracking Research Articles

Automatic Detection and Tracking of Objects of Interest in Video Data with Global Motion

Introduction. At present, automatic capture and tracking of moving objects in video data obtained by a video camera mounted on a mobile carrier represents a relevant research task. Its successful solution is challenged by such factors, as a non-uniform background, object overlapping between one another and the background, significant and rapid changes in the size of the object of interest, abrupt changes in the movement trajectory of the mobile carrier.Aim. To develop an automatic method for detecting moving objects followed by their tracking in video data obtained under difficult observation conditions. An additional requirement imposed on the tracking stage consists in the restriction of computing resources.Materials and methods. The method is based on a convolutional neural network with a YOLO architecture. Due to the restriction of computing resources, object tracking is implemented without neural network solutions. In order to ensure stable tracking, several detectors are used simultaneously with the subsequent analysis of the data obtained. The tracking stage involves a detector based on histograms of oriented gradients (HOG), supplemented by a detector based on correlation filtering and motion trajectory prediction based on the Kalman filter.Results. At the automatic detection stage, the TPR, averaged over all video files participating in the experiments, was equal to 0.81, with the FPR corresponding to 0.10. At the tracking stage, the failure rate (tracking failures) was 6·10 –5.Conclusion. The proposed method can be successfully used to detect and track objects at a distance of 1500 m with an object projection size on the frame of 5 × 5 pixels under the conditions of global motion, a non-uniform background, and significant changes in the properties of the object of interest.

Design of Automatic Underwater Object Detection and Tracking System for Penta Tubular ROV in Blue Economy Environment

The development of underwater vehicles has opportunity to be further improved. One of the most challenging for developing underwater vehicle is developing automatic tracking system for Autonomous Underwater Vehicle (AUV). This research develops AUV with automatic object detection and tracking systems. These systems are developed by using vision-based method. The AUV is designed with Penta Tubular type whereas the automatic object detection system is started from captured image taken by underwater camera in the form of Blue-Green-Red (BGR) color space, which then will go through a masking process. The object resulted by object detection system is then to be used as followed object for automatic tracking system. Meanwhile, Automatic object tracking system is established by using two vertical lines that are used as motion parameters of the AUV. By considering the object position compared with the two vertical lines, the automatic tracking guidance system can be generated. The developed automatic object detection and tracking systems can be used well for AUV, in which verage travel time for forward and backward motion are 14.20 seconds in 0.14 m/s, and 19.82 seconds in 0.1 m/s while average travel time is 15.36 seconds and 13.58 seconds for right and left motion test respectively.

Automatic tracking of objects using improvised Yolov3 algorithm and alarm human activities in case of anomalies

Automatic tracking of objects using improvised Yolov3 algorithm and alarm human activities in case of anomalies

КОРРЕЛЯЦИОННЫЙ АЛГОРИТМ АВТОМАТИЧЕСКОГО ОБНАРУЖЕНИЯ И СОПРОВОЖДЕНИЯ С ОБУЧЕНИЕМ В РЕАЛЬНОМ ВРЕМЕНИ

In order to develop a stable algorithm for automatic detection and tracking of nondeterministicobjects with real-time learning for embedded computing systems with optoelectronicdevices, within the framework of this work, a study and analysis of the existing world scientific andtechnical experience in the field of automatic tracking algorithms for general purposes was carriedout. The article shows that the most stable modern automatic tracking algorithms are a systemthat makes a decision about the current position, size and other parameters of the trackedimage based on the model being trained. The authors of the study identified the most effective ofthe applied basic algorithms suitable for use in embedded computing systems of robotic complexes,and developed a new algorithm for automatic detection and maintenance of non-deterministicobjects. A semi-natural testing of the developed algorithm was carried out and its effectivenesswas evaluated in solving problems not only of automatic tracking of objects, but also problems ofautomatic detection of objects using several reference images. In conclusion, proposals are presentedfor further improving the accuracy of the developed algorithm and for its optimization andimplementation in the special software of on-board computer systems of aircraft.

Automatic maneuver detection and tracking of space objects in optical survey scenarios based on stochastic hybrid systems formulation

The state space representation of active resident space objects can be posed in the form of a stochastic hybrid system. Satellite maneuvers may be accounted for according to control cost or heuristical considerations, yet it is possible to jointly consider a combination of both. In this work, Sequential Monte Carlo filtering techniques are applied to the maneuvering target tracking problem in an optical survey scenario, where the maneuver control inputs are characterized in a Bayesian inference process. Due to the scarcity of data inherent to space surveillance and tracking, model switching probabilities are not estimated but derived from the ability of the state representation to fit incoming measurements. A Markov Chain Monte Carlo sampling scheme is used to explore the region assumed accessible to the object in terms of the hypothesized post-maneuver observation and a novel and efficient control distance metric. Results are obtained for a simulated optical survey scenario, and comparisons are drawn with respect to a moving horizon least-squares estimator. The proposed framework is proved to allow for a capable implementation of an automated online maneuver detection algorithm, thus contributing to the reduction of uncertainty in the state of active space objects.

Comprehensive observations on pendulum oscillation using stereo vision

We have developed a stereo vision system for observing real objects for experimental physics demonstration. The stereo vision observes objects from two different points of view. An observer can identify the distance and velocity of an object based on at least two different points of view. This paper demonstrates our comprehensive observation in a pendulum oscillation experiment to test the stereo vision system where a stereo camera is in a forward view position. The stereo camera produces stereo images. We use automatic object tracking techniques to obtain the 3D position of the pendulum in real space based on the sequential stereo images of the pendulum movement. The experimental results in a 3D graph that shows the trajectory of the pendulum oscillation. We confirm the common understanding of the oscillation movement by obtaining the graph of the pendulum oscillation as a function of time. We can obtain the oscillation motion parameters such as amplitude, angular frequency, frequency, period, and initial phase. Also, we can show the damping oscillation phenomena and the damping factor.

Automatic Tracking of Objects Using Improvised Yolov3 Algorithm and Alarm Human Activities in Case of Anomalies

Automatic Tracking of Objects Using Improvised Yolov3 Algorithm and Alarm Human Activities in Case of Anomalies

Tracking of objects in a multi-sensor fusion system for border surveillance

In this work, we present a Fusion and Tracking system developed within the EU project FOLDOUT aimed to facilitate border guards work by fusing separate sensor information and presenting automatic tracking of objects detected in the surveillance area. The focus of FOLDOUT is on through-foliage detection in the inner and outermost regions of the EU. Fusing several sensor signals increases the effectiveness of detection, particularly in forested and other areas hidden by foliage. We use weighted maps (also called Heatmaps) to combine multi-sensor information; tracking is performed on the resulting fused objects; a track is created or updated based on cost calculation of associating fused detections temporally. We compare tracking results from individual sensors and from fused objects from data collected in a simulated border that is representative of actual EU borders in Bulgaria. The results show how tracking is enhanced if performed on fused data rather than from individual sensor information.

Quality of Radar Direction-Finding via Conical Scanning

Introduction. Conical scanning is applied for optimizing hardware resources in new devices, as well as when upgrading existing systems. All this explains the relevance of studying this type of direction finding systems.Aim. To adjust and complement the known calculation relations for the variance of direction finding results – an indicator of the quality (accuracy) of direction finding, as well as to determine the possibilities of optimizing direction finding and automatic object tracking processes.Materials and methods. Factors limiting the accuracy of direction finding via conical scanning were analyzed using spectral analysis. Mathematical modeling followed by statistical processing of quantitative results makes it possible to determine the conditions under which the influence of certain factors is predominant, as well as the conditions under which adjustment (completion) of the known calculation relations is required. The specified conditions are the errors at which the objects of direction finding are tracked. New calculation relations for the mentioned adjustment were determined by the methods of statistical radio engineering.Results. The validity of the calculation relations found is confirmed by mathematical modeling. Calculations and modeling lead to the need to optimize parameters for automatic object tracking systems.Conclusion. The study shows that, when choosing parameters for auto-tracking systems with conical scanning, it is important to implement object tracking not with minimal, but rather with optimized tracking errors in angular coordinates, which are to be estimated during direction finding. Moreover, the optimized errors (the values of static errors and the most probable values of the dynamic tracking errors) will require adjustment of the known analytical estimates for the variance of the direction finding results – the qualitative indicator of the direction finder (accuracy indicator). The determined analytical relationships allow such an adjustment to be performed, leading to an increased variance estimate by 10 dB.

Analysis of the stability of automatic tracking of super maneuvering air objects by radio technical tracking systems of the multichannel radar

Multichannel tracking radars with phased antenna arrays are widely used to track air targets. The use of a phased array in combination with digital computing technology allows to control the radar radiation pattern and track several targets in the time distribution mode. Air target tracking in a multichannel radar is provided by subsystems for measuring range, radial velocity and angular coordinates, in most cases, without adaptation to the external influence characteristics. When tracking supermaneuverable air targets, such as 5th and so-called 4++ generation fighters, there is a decrease in the accuracy and stability of tracking relative to the area without maneuver. If the tracking system algorithms are tuned to a low intensity of maneuvering or its absence, a significant increase in the error of tracking the aircraft in the maneuvering section will lead to disruption of auto tracking due to a significant dynamic component of the error. The stability of auto-tracking of maneuvering targets by subsystems of range, radial velocity, and angular coordinates with fixed parameters for the case when the setting of the parameters of the tracking system algorithms coincide with the characteristics of the external influence is analyzes in the paper. The influence of the observation model parameters, the stochastic model of the target movement with exponentially correlated values of the target acceleration, and the measurement period of the target coordinates on the potential tracking accuracy by radio technical tracking systems of the multichannel radar is investigated. To assess the stability of auto-tracking, it is proposed to use the equivalent aperture size of the discriminating characteristic. The influence of the parameters of the target movement stochastic model, the observation model, and the measurement period of the target coordinates on the stability of auto-tracking in terms of range, radial velocity, and angular coordinates is estimated. It is shown that the "weak link" is the radial velocity tracking system. As a result of the research carried out, it becomes possible to further assess the feasibility of adapting the auto-tracking systems to the target maneuvering characteristics and to develop recommendations for choosing the measurement period of the target coordinates.

Automatic Object Tracking and Segmentation Using Unsupervised SiamMask

In this paper we address the basic limitation of SiamMask - the state of the art single object tracking and segmentation algorithm. SiamMask requires semi-supervision in that it needs a bounding box to be drawn manually around the object that has to be tracked. This is however not always possible or feasible, and slows down the pipeline even in the best case. We overcome this limitation by using state-of-the-art object detection algorithms: Detectron2 and YOLO to automatically detect the object and then track using SiamMask. The main purpose of this study is to devise an efficient technique for an end-to-end object detection and tracking, which can then be used in other applications like self-driving cars, etc. We compared different approaches using current state-of-the-art tools for time and detection efficiency. One of the secondary aim was to test how the two approaches perform on different types of datasets. We note that YOLO gives better and more meaningful detection of objects in the scene. However, Detectron2 gives a higher detection speed than YOLO, making the overall detection and tracking process faster.

Optimization of Automatic Support Systems of Air Objects Based on Local Quadratic-biquadratic Functionals. I. Synthesis of Optimum Control

The features of the functioning of the systems of automatic tracking of air objects, optimized according to the classical quadratic functionals, are considered, and their shortcomings are noted when tracking objects moving along complex trajectories. A variant of the solution of the problem of the local optimization of automatic tracking systems based on minimizing quadratic-biquadratic quality functionals is proposed. An example of the synthesis of the tracking system of a radar goniometer is given and analyzed. In the second part of the article, the results of modeling the created system will be analyzed.

Intelligent Automatic Object Tracking Method by Integration of Laser Scanner System and INS

In the last years Autonomous vehicle navigation has had considerable improvements; however, this technology has not yet been able to effectively be integrated with mobile systems that work inside buildings, houses or without the influence of absolute references such as GPS. It is for this reason that this paper provides a methodological proposal that allows the integration of the use of object detection methods such as YOLO (You Only Look Once) or R-CNN (Region Convolutional Neural Networks), with laser scanning systems for the measurement of spatial coordinates and an INS system to obtain spatial positioning through relative references. This method integrates the principle of dynamic triangulation and the methodological proposal IKZ (Inertial Navigation with Kalman and Zero Update) that allow to obtain acceptable mobile autonomy times in which the sector of interest is re-scanned.

МНОГОАГЕНТНЫЙ АЛГОРИТМ АВТОМАТИЧЕСКОГО ОБНАРУЖЕНИЯ И СОПРОВОЖДЕНИЯ НЕДЕТЕРМИНИРОВАННЫХ ОБЪЕКТОВ

In order to develop a robust algorithm for the automatic detection and tracking of non-deterministic objects for embedded computing systems, in this work, a study and analysis in the field of state-of-the-art general-purpose automatic tracking algorithms is performed. The most successful of those algorithms suitable for long-term stable automatic tracking of objects (without a priori knowledge of the type of object being tracked) have already gone beyond solving exclu-sively tracking problems, and include a synergistic combination of several heterogeneous tracking algorithms, as well as at least one automatic detection and / or classification algorithm. Thus, the authors of the article conclude that the most stable modern automatic tracking algorithms are a multi-agent system that makes a decision about the current position, size and other parameters of the tracked object image based on intelligent voting of system’s submodules that independently monitor the object and form its model. Individual models of each of the submodules are updated based on the results of a collective decision. The authors of the study identified the most effective of the applied basic algorithms suitable for use in embedded computing systems of robotic systems, and developed a new multi-agent algorithm for the automatic detection and tracking of non-deterministic objects. The presented multi-agent algorithm includes a submodule for extracting and matching key points in images, a clustering and filtering submodule for key points using the DBSCAN algorithm, a tracking submodule based on the optical flow calculation algorithm, and a key point classification submodule. A semi-natural testing of the developed algorithm was carried out and its effectiveness in solving tasks not only of automatic tracking of objects, but also in tasks of automatic objects detection using several reference images were evaluated. In conclusion, the authors present steps for further improving the accuracy and performance of the developed algo-rithm for its forthcoming implementation for on-board computing systems of aerial vehicles.

COMPLEXING IMAGES IN A TWO-CHANNEL OPTICAL-ELECTRONIC SYSTEM AS A METHOD OF REDUCING THE PROBABILITY OF DISAPPOINTING AN AUTOMATIC SURVEILLANCE OF LOW CONTRAST OBJECTS

The paper discusses the work of the correlation algorithm for automatically tracking objects of interest in a two-channel optical-electronic system in a complex background-target environment and the presence of intentional interference. Conditions in which the contrast of the desired object in both channels is negligible but not equal to zero are considered difficult in the paper. Intentional interference refers to masking interference of natural and artificial origin, which helps to reduce the contrast of the object in both channels. By two-channel means an optical-electronic system, which includes channels of the visible and infrared ranges. It is believed that the multi-spectral images of both channels are reduced to a single time and scale, which allows them to be integrated using various methods. The purpose of this paper is to prove that the likelihood of disruption of automatic tracking is reduced when complexing the source images of the visible and infrared ranges, when the contrast of the desired object in both channels is negligible. For research the mathematical apparatus of the theory of random functions and simulation with subsequent statistical data processing are used. It is shown that the probability of disruption, characterized by the ratio of the correlation coefficients of two fragments of images, one of which contains the desired object, and the second not, depends on both the correlation coefficients and the values of their mean square deviations. The simulation shows that the breakdown of tracking occurs both when the mean square deviations of these correlation coefficients are equal, and at their different values, moreover, an increase in their difference increases the probability of a breakdown. The article shows that the likelihood of a breakdown in a two-channel optoelectronic system will decrease when using two channels, compared to working only in the visible or infrared channel. The obtained results substantiate the promise of using image complexing in multichannel systems of automatic tracking of objects in a complex background-target environment and the presence of deliberate interference.

Comparison of Tracking Techniques on 360-Degree Videos

With the availability of 360-degree cameras, 360-degree videos have become popular recently. To attach a virtual tag on a physical object in 360-degree videos for augmented reality applications, automatic object tracking is required so the virtual tag can follow its corresponding physical object in 360-degree videos. Relative to ordinary videos, 360-degree videos in an equirectangular format have special characteristics such as viewpoint change, occlusion, deformation, lighting change, scale change, and camera shakiness. Tracking algorithms designed for ordinary videos may not work well on 360-degree videos. Therefore, we thoroughly evaluate the performance of eight modern trackers in terms of accuracy and speed on 360-degree videos. The pros and cons of these trackers on 360-degree videos are discussed. Possible improvements to adapt these trackers to 360-degree videos are also suggested. Finally, we provide a dataset containing nine 360-degree videos with ground truth of target positions as a benchmark for future research.

Evaluation of detection and tracking performance in imaging systems

Imaging systems can be used to obtain situational awareness in maritime situations. Important tools for these systems are automatic detection and tracking of objects in the acquired imagery, in which numerous methods are being developed. When designing a detection or tracking algorithm, its quality should be ensured by a comparison with existing algorithms and/or with a ground truth. Detection and tracking methods are often designed for a specific task, so evaluation with respect to this task is crucial, which demands for different evaluation measures for different tasks. We, therefore, propose a variety of quantitative measures for the performance evaluation of detectors and trackers for a variety of tasks. The proposed measures are a rich set from which an algorithm designer can choose in order to optimally design and assess a detection or tracking algorithm for a specific task. We compare these different evaluation measures by using them to assess detection and tracking quality in different maritime detection and tracking situations, obtained from three real-life infrared video data sets. With the proposed set of evaluation measures, a user is able to quantitatively assess the performance of a detector or tracker, which enables an optimal design for his approach.

Control of unmanned aerial vehicles based on the detection algorithm

The work analyzes the current tasks of unmanned aircraft, which showed the need to develop algorithms and systems for automatic tracking of objects operating in real-time mode, in the tasks of conducting reconnaissance flights and targeting precision weapons. An algorithm for detecting objects in the image by correlation filtration based on a two-dimensional discrete Fourier transform for control problems of unmanned aerial vehicles is developed. A software implementation of the detection algorithm using the OpenCV / C ++ computer vision library (EmguCV / C #) is obtained. A number of methods of computer libraries used for constructing program code are presented. Time schedules and recommendations for using the algorithm in different shooting conditions are given.

A Hybrid Approach for Real-Time Object Detection and Tracking to Cover Background Turbulence Problem

Objectives: A number of techniques available in literature do not discuss the problem of incorrect object detection in the scenes having unstable or moving background. A novel hybrid technique is proposed in the paper to cover this problem. Methods/Statistical Analysis: The new approach proposed is the hybrid of two well known techniques for object detection. One is frame-differencing approach and other is skin colour modelling. This newer technique exploits the fact that the demerit of one technique works as merit for other and hence hybrid technique resolves the problem of moving background or turbulence in background. A real time video having turbulence in background is used for testing the efficiency of the approach. Findings: With accuracy of 0.97, the proposed approach outperforms the individual approaches i.e. frame-differencing and skin colour modelling. A very low value of False Positive Rate (FPR) for proposed approach compared to other approaches confirms the least incorrect detections. High value of True Positive Rate (TPR) conveys that fall out of correct object is least in the video by proposed approach. Results show that the proposed approach better applicable for background with turbulence. Application/Improvements: Automatic object detection and tracking in applications as surveillance are tractable with the proposed hybrid approach.

Collaborative Annotation of Videos Relying on Weak Consistency

This work discusses a distributed interactive video system that supports video annotation using simultaneous hyperlinking by multiple users. The users mark and annotate objects within the video with links to other media such as text, images, websites, or other videos. Annotations are visualized on the client user interface as an overlay close to the objects. Our system is intuitive to use; for example, it contains automatic object-tracking functionality that correctly positions the annotations, even when the form or location of an object changes. Thus, our first contribution discusses the adaptive object-tracking algorithm used for this repositioning. It shows improved precision and reliability in comparison to nonadaptive algorithms. A second key issue is to keep the system responsive when the number of concurrent annotators increases. Thus, we rely on the concept of eventual consistency between different network entities. While this weak form of consistency allows temporary inconsistencies, it ensures that a consistent state can be reached. Thus, the second contribution is the design and evaluation of our distributed interactive video system, which relies on the weak consistency paradigm.