Hybrid Tracking Research Articles

A Robotic Flexible Drill and Its Navigation System for Total Hip Arthroplasty

This paper presents a robotic flexible drill and its navigation system for total hip arthroplasty (THA). The new robotic system provides an unprecedented and unique capability to perform curved femoral milling under the guidance of a multimodality navigation system. The robotic system consists of three components. Firstly, a flexible drill manipulator comprises multiple rigid segments that act as a sheath to a flexible shaft with a drill/burr attached to the end. The second part of the robotic system is a hybrid tracking system that consists of an optical tracking system and a position tracking system. Optical tracking units are used to track the surgical objects and tools outside the drilling area, while a rotary encoder placed at each joint of the sheath is synchronized to provide the position information for the flexible manipulator with its virtual object. Finally, the flexible drill is integrated into a computer-aided navigation system. The navigation system provides real time guidance to a surgeon during the procedure. The flexible drill system is then able to implement THA by bone milling. The final section of this paper is an evaluation of the flexible and steerable drill and its navigation system for femoral bone milling in sawbones.

이동형 위성안테나 시스템의 위성추적 성능개선을 위한 스텝추적과 모노펄스추적의 혼합추적 방법

이동형 위성안테나 시스템의 위성추적 성능개선을 위한 스텝추적과 모노펄스추적의 혼합추적 방법

On the Relation Between the Minimum Principle and Dynamic Programming for Classical and Hybrid Control Systems

Hybrid optimal control problems are studied for a general class of hybrid systems where autonomous and controlled state jumps are allowed at the switching instants and in addition to terminal and running costs switching between discrete states incurs costs. The statements of the Hybrid Minimum Principle and Hybrid Dynamic Programming are presented in this framework and it is shown that under certain assumptions the adjoint process in the Hybrid Minimum Principle and the gradient of the value function in Hybrid Dynamic Programming are governed by the same set of differential equations and have the same boundary conditions and hence are almost everywhere identical to each other along optimal trajectories. Analytic examples are provided to illustrate the results and, in particular, a Riccati formalism for linear quadratic hybrid tracking problems is presented.

Hybrid tracking model and GSLM based neural network for crowd behavior recognition

Crowd behaviors analysis is the ‘state of art’ research topic in the field of computer vision which provides applications in video surveillance to crowd safety, event detection, security, etc. Literature presents some of the works related to crowd behavior detection and analysis. In crowd behavior detection, varying density of crowds and motion patterns appears to be complex occlusions for the researchers. This work presents a novel crowd behavior detection system to improve these restrictions. The proposed crowd behavior detection system is developed using hybrid tracking model and integrated features enabled neural network. The object movement and activity in the proposed crowded behavior detection system is assessed using proposed GSLM-based neural network. GSLM based neural network is developed by integrating the gravitational search algorithm with LM algorithm of the neural network to increase the learning process of the network. The performance of the proposed crowd behavior detection system is validated over five different videos and analyzed using accuracy. The experimentation results in the crowd behavior detection with a maximum accuracy of 93% which proves the efficacy of the proposed system in video surveillance with security concerns.

HeartBEATS: A hybrid energy approach for real-time B-spline explicit active tracking of surfaces

In this manuscript a novel method is presented for left ventricle (LV) tracking in three-dimensional ultrasound data using a hybrid approach combining segmentation and tracking-based clues. This is accomplished by coupling an affine motion model to an existing LV segmentation framework and introducing an energy term that penalizes the deviation to the affine motion estimated using a global Lucas–Kanade algorithm. The hybrid nature of the proposed solution can be seen as using the estimated affine motion to enhance the temporal coherence of the segmented surfaces, by enforcing the tracking of consistent patterns, while the underlying segmentation algorithm allows to locally refine the estimated global motion. The proposed method was tested on a dataset composed of 24 4D ultrasound sequences from both healthy volunteers and diseased patients. The proposed hybrid tracking platform offers a competitive solution for fast assessment of relevant LV volumetric indices, by combining the robustness of affine motion tracking with the low computational burden of the underlying segmentation algorithm.

Hybridizing and Relaxing Dependence Tracking for Efficient Parallel Runtime Support

It is notoriously challenging to develop parallel software systems that are both scalable and correct. Runtime support for parallelism—such as multithreaded record and replay, data race detectors, transactional memory, and enforcement of stronger memory models—helps achieve these goals, but existing commodity solutions slow programs substantially to track (i.e., detect or control) an execution’s cross-thread dependencies accurately. Prior work tracks cross-thread dependencies either “pessimistically,” slowing every program access, or “optimistically,” allowing for lightweight instrumentation of most accesses but dramatically slowing accesses that are conflicting (i.e., involved in cross-thread dependencies). This article presents two novel approaches that seek to improve the performance of dependence tracking. Hybrid tracking (HT) hybridizes pessimistic and optimistic tracking by overcoming a fundamental mismatch between these two kinds of tracking. HT uses an adaptive, profile-based policy to make runtime decisions about switching between pessimistic and optimistic tracking. Relaxed tracking (RT) attempts to reduce optimistic tracking’s overhead on conflicting accesses by tracking dependencies in a “relaxed” way—meaning that not all dependencies are tracked accurately—while still preserving both program semantics and runtime support’s correctness. To demonstrate the usefulness and potential of HT and RT, we build runtime support based on the two approaches. Our evaluation shows that both approaches offer performance advantages over existing approaches, but there exist challenges and opportunities for further improvement. HT and RT are distinct solutions to the same problem. It is easier to build runtime support based on HT than on RT, although RT does not incur the overhead of online profiling. This article presents the two approaches together to inform and inspire future designs for efficient parallel runtime support.

EnViSoRS: Enhanced Vision System for Robotic Surgery. A User-Defined Safety Volume Tracking to Minimize the Risk of Intraoperative Bleeding

In abdominal surgery, intra-operative bleeding is one of the major complications that affect the outcome of minimally invasive surgical procedures. One of the causes is attributed to accidental damages to arteries or veins, and one of the possible risk factors falls on the surgeon's skills. This paper presents the development and application of an Enhanced Vision System for Robotic Surgery (EnViSoRS), based on a user-defined Safety Volume (SV) tracking to minimise the risk of intra-operative bleeding. It aims at enhancing the surgeon's capabilities by providing Augmented Reality (AR) assistance towards the protection of vessels from injury during the execution of surgical procedures with a robot. The core of the framework consists in: (i) a hybrid tracking algorithm (LT-SAT tracker) that robustly follows a user-defined Safety Area (SA) in long term; (ii) a dense soft tissue 3D reconstruction algorithm, necessary for the computation of the SV; (iii) AR features for visualisation of the SV to be protected and of a graphical gauge indicating the current distance between the instruments and the reconstructed surface. EnViSoRS was integrated with a commercial robotic surgery system (the dVRK system) for testing and validation. The experiments aimed at demonstrating the accuracy, robustness, performance and usability of EnViSoRS during the execution of a simulated surgical task on a liver phantom. Results show an overall accuracy in accordance with surgical requirements (< 5mm), and high robustness in the computation of the SV in terms of precision and recall of its identification. The optimisation strategy implemented to speed up the computational time is also described and evaluated, providing AR features update rate up to 4 fps without impacting the real-time visualisation of the stereo endoscopic video. Finally, qualitative results regarding the system usability indicate that the proposed system integrates well with the commercial surgical robot and has indeed potential to offer useful assistance during real surgeries.

OC-0277: Assessment of the implant geometry in interstitial brachytherapy by a hybrid tracking system

OC-0277: Assessment of the implant geometry in interstitial brachytherapy by a hybrid tracking system

Robust and real-time pose tracking for augmented reality on mobile devices

This paper addresses robust and ultrafast pose tracking on mobile devices, such as smartphones and small drones. Existing methods, relying on either vision analysis or inertial sensing, are either too computational heavy to achieve real-time performance on a mobile platform, or not sufficiently robust to address unique challenges in mobile scenarios, including rapid camera motions, long exposure time of mobile cameras, etc. This paper presents a novel hybrid tracking system which utilizes on-device inertial sensors to greatly accelerate the visual feature tracking process and improve its robustness. In particular, our system adaptively resizes each video frame based on inertial sensor data and applies a highly efficient binary feature matching method to track the object pose in each resized frame with little accuracy degradation. This tracking result is revised periodically by a model-based feature tracking method (Hare et al. 2012) to reduce accumulated errors. Furthermore, an inertial tracking method and a solution of fusing its results with the feature tracking results are employed to further improve the robustness and efficiency. We first evaluate our hybrid system using a dataset consisting of 16 video clips with synchronized inertial sensing data and then assess its performance in a mobile augmented reality application. Experimental results demonstrated our method’s superior performance to a state-of-the-art feature tracking method (Hare et al. 2012), a direct tracking method (Engel et al. 2014) and the Vuforia SDK (Ibanez and Figueras 2013), and can run at more than 40 Hz on a standard smartphone. We will release the source code with the pubilication of this paper.

An experimentally validated couch and MLC tracking simulator used to investigate hybrid couch-MLC tracking.

Couch and MLC tracking are two novel techniques to mitigate intrafractional tumor motion on a conventional linear accelerator, but both techniques still have residual dosimetric errors. Here, we first propose and experimentally validate a software tool to simulate couch and MLC tracking, and then use the simulator to study hybrid couch-MLC tracking for improved tracking performance. The tracking simulator requires a treatment plan and a motion trajectory as input and simulates the delivered monitor units and motion of all accelerator parts as function of time. The simulator outputs accelerator log files synchronized with the target motion as well as the MLC exposure error, which is a simple dose error surrogate. A series of couch and MLC tracking experiments were used to determine appropriate parameters for the simulator dynamics and to validate the simulator by its ability to reproduce the experimental tracking accuracy. Three hybrid couch-MLC tracking strategies were investigated. All strategies divided the target motion in beam's eye view into motion perpendicular and parallel to the MLC leaves. In the hybrid strategies, couch tracking compensated for the following target motion components (in order of decreasing couch tracking contribution): (a) all perpendicular motion, (b) residual perpendicular motion less than half a leaf width, and (c) persistent residual perpendicular motion that was stable at a time scale of 1s. MLC tracking compensated for the remaining target motion. All tracking strategies were simulated with two prostate and two lung cancer single-arc VMAT plans using 695 prostate trajectories and 160 lung tumor trajectories. The tracking error was quantified as the MLC exposure error. The couch motion was quantified as the mean speed, acceleration, and jerk of the couch. The simulator reproduced the experimental gantry position with a mean (maximum) root-mean-square (rms) error of 0.07°(0.2°). The geometrical rms tracking error was reproduced with mean (maximum) absolute errors of 0.20mm(0.23mm) and 0.1mm(0.23mm) for MLC tracking parallel and perpendicular to the MLC leaves, and 0.40mm(0.46mm), 0.09mm(0.25mm), and 0.20mm(0.46mm) for couch tracking in the left-right, anterior-posterior, and cranio-caudal directions. The MLC exposure error of VMAT MLC tracking was reproduced with a mean absolute error of 5.6%. All hybrid tracking strategies reduced the couch motion relative to pure couch tracking and improved the tracking accuracy compared with pure MLC tracking. The mean MLC exposure error reduction relative to no tracking was 66.6% (couch tracking), 72.9% (hybrid (1)), 70.2% (2), 59.1% (3), and 55.6% (MLC tracking) for lung tumor motion and 76.5% (couch tracking), 76.1% (1), 74.3% (2), 72.3% (3), and 35.9% (MLC tracking) for prostate motion. For prostate motion, pure MLC tracking resulted in rather large MLC exposure errors that were more than halved with all hybrid tracking strategies. A couch and MLC tracking simulator was developed and experimentally validated against a series of tracking experiments. All hybrid couch-MLC tracking strategies improved MLC tracking. Two strategies also improved couch tracking of lung tumors. In particular, MLC tracking of prostate may be greatly improved by a modest degree of couch motion.

Martial Arts, Dancing and Sports dataset: A challenging stereo and multi-view dataset for 3D human pose estimation

Human pose estimation is one of the most popular research topics in the past two decades, especially with the introduction of human pose datasets for benchmark evaluation. These datasets usually capture simple daily life actions. Here, we introduce a new dataset, the Martial Arts, Dancing and Sports (MADS), which consists of challenging martial arts actions (Tai-chi and Karate), dancing actions (hip-hop and jazz), and sports actions (basketball, volleyball, football, rugby, tennis and badminton). Two martial art masters, two dancers and an athlete performed these actions while being recorded with either multiple cameras or a stereo depth camera. In the multi-view or single-view setting, we provide three color views for 2D image-based human pose estimation algorithms. For depth-based human pose estimation, we provide stereo-based depth images from a single view. All videos have corresponding synchronized and calibrated ground-truth poses, which were captured using a Motion Capture system. We provide initial baseline results on our dataset using a variety of tracking frameworks, including a generative tracker based on the annealing particle filter and robust likelihood function, a discriminative tracker using twin Gaussian processes [1], and hybrid trackers, such as Personalized Depth Tracker [2]. The results of our evaluation suggest that discriminative approaches perform better than generative approaches when there are enough representative training samples, and that the generative methods are more robust to diversity of poses, but can fail to track when the motion is too quick for the effective search range of the particle filter. The data and the accompanying code will be made available to the research community.

Zero-stopping constraint-based hybrid tracking model for dynamic and high-dense crowd videos

ABSTRACTOwing to the importance of video surveillance in the public area, tracking finds significant applications using computer vision algorithms to observe the activity of human. In tracking, multi-object tracking is an active research to analyse and detect the activity of anomalies in the crowded scenes. Accordingly, different multi-object tracking algorithms are proposed in the literature to track the human behaviour of the crowded scenes. In this paper, we have presented a zero-stopping criteria-based hybrid tracking algorithm for high-dense crowd videos. Here, head objects are detected using the proposed objective function which considers both colour and texture property of videos. Then, tracking based on motion is performed using the proposed HSIM measure which includes structural similarity (SSIM) and the proposed similarity function. Along with, the data prediction model, exponential weighted moving average (EWMA), is also utilised to track the spatial location of human objects. These two tracking models are then hybridised to obtain the final tracked output. The experimentation is performed with three marathon sequences and the performance is evaluated with particle filtering-based algorithm using tracking number, tracking distance and optimal subpattern assignment metric (OSPA).

Self-corrective knowledge-based hybrid tracking system using BIM and multimodal sensors

Researchers have recently devoted considerable attention to acquiring location awareness of assets. They have explored various technologies, such as video cameras, radio signal strength indicator-based sensors, and motion sensors, in the development of tracking systems. However, each system presents unique drawbacks especially when applied in complex indoor construction environments; this paper classifies them into two categories: absolute tracking and relative tracking. By understanding the nature of problems in each tracking category, this research develops a novel tracking methodology that uses knowledge of the strengths and weaknesses of various components used in the proposed tracking system. This paper presents the development of a hybrid-tracking system that integrates Bluetooth Low Energy (BLE) technology, motion sensors, and Building Information Model (BIM). The hypothesis tested through this integration was whether such knowledge-based integration could provide a method that can correct errors found in each of the used sensing technologies and thereby improve the reliability of the tracking system. Field experimental trials were conducted in a full-scale indoor construction site to assess the performance of individual components and the integrated system. The results indicated that the addition of map knowledge from a BIM model showed the capability of correcting improbable movements. Furthermore, the knowledge-based decision making process demonstrated its capability to make positive interaction by reducing the positioning errors by 42% on average. In sum, the proposed hybrid-tracking system presented a novel method to compensate for the weakness of each system component and thus achieve a more accurate and precise tracking in dynamic and complex indoor construction sites.

A hybrid tracking system for image-guided spine surgery using a tracked mobile C-arm: a phantom study

In recent years, an increasing number of various surgeries are observed utilizing fluoroscopy. The radiation exposure received by patients and medical staff and the surgical guidance in multiple planes frequently necessitate the positioning of a mobile C-arm. Operative navigation enables a mobile C-arm to provide multiplanar surgical guidance and decreases the radiation dose to the patient and operating room personnel. In this study, we propose a video-based tracked mobile C-arm (referred to as a "tracked C-arm system") to position the system. This system defines a reference framework to maintain the video-optical tracker data and computed tomography (CT) or cone-beam CT images' alignment as fine as possible despite patient or tracker movement. By employing our uniquely designed "six-facet" reference marker attached to the spine phantom, registration between the video-optical tracker and the spine phantom is maintained at arbitrary angles of the mobile C-arm. The tracked C-arm system provides a statistically significant improvement (P < 0.001) in target registration error in comparison with the conventional system: 0.80 $\pm $ 0.34 mm versus 1.60 $\pm $ 0.43 mm, respectively. The tracked C-arm system is designed to generate digitally reconstructed radiograph images from the mobile C-arm perspective, with projection error on the order of 0.74 $\pm $ 0.13 mm. Integration of the hybrid tracking system with mobile C-arm guidance has the capability to provide registration, reduce radiation exposure, and improve target registration accuracy.

Hybrid tracking detector based on semiconductor and gas technologies

New Silicon-Gas Pixel Detector (SiGPD) is the result of the further development already proposed Gas-Pixel Detector (GPD). The only disadvantage of GPD was the uncertainty of the particle arriving time. The problem can be solved by additional silicon pixels implementation inside existed electronic chip epitaxial layer during regular chip production. The cost of new Si-Gas Pixel Detector remains practically the same. The new detectors have self-triggering properties and can be used for a first level trigger generation in the particle physics experiments and for many other applications.

Crowd Behavior Recognition Using Hybrid Tracking Model and Genetic algorithm Enabled Neural Network

In the current era, crowd behavior analysis is important topic due to the significance of video surveillance in the public area. Literature presents a handful of works for crowd behavior detection and analysis. Even though, the complicated challenges such as, low quality video, wide variation in the density of crowds and difficult motion patterns pose a complicated challenges for the researchers in crowd behavior detection. In order to alleviate these issues, we develop a crowd behavior detection system using hybrid tracking model and integrated features enabled neural network. The proposed crowd behavior detection system estimate the direction of movement of objects as well their activity using proposed GLM-based neural network. The proposed GLM-based neural network integrates the LM algorithm with genetic algorithm to improve the learning process of neural network. The performance of the proposed crowd behavior detection algorithm is validated with five different video and the performance is extensively analyzed using accuracy. From research outcome, we proved that the proposed system obtained the maximum accuracy of 95% which is higher than the existing methods taken for comparison.

Kalman Filter for Hybrid Tracking Technique in Augmented Reality

Augmented reality started to emerge as a promising visualization technique that tracks real objects and adds virtual content into real world context using camera view. Many augmented reality solutions are based on computer vision techniques to identify and track objects. Problems that must be solved are image transformations, chaotic environment, lighting condition and occlusion from users’ or objects in the environment, which causes virtual content to disappear. This has a negative impact for augmented reality usability, therefore, object recognition and tracking in real-time becomes difficult and sometimes an impossible task. In this research orientation-position information acquisition using computer vision and sensor fusion techniques are analysed. Experiments are accomplished with predefined assumptions and simulated orientation-position information. Conditions for optimal orientation-position estimates are introduced. Research results are compared and supplementary properties are presented of a proposed hybrid tracking technique using Kalman filter.DOI: http://dx.doi.org/10.5755/j01.eie.22.6.17228

Mining video data

Learning spaces influence how we act, however there is a lack of systemic research addressing the impact of environments on teaching and learning. In this paper, we introduce a hybrid tracking technique in which a colour model is combined with algorithms to identify human positions, and applied to video data. The aim of identifying patterns of movement that could be used to indicate successful collaboration in open plan learning spaces. We apply the method to a previously analyzed dataset, to demonstrate how multiple analytic techniques can be used to build a complex understanding of learner movement in relation to collaboration and learning. We conclude with suggestions of the ways in which the results could be used by instructors to inform orchestration of complex learning environments, as well as directions for future research.

A Global Tracking Controller for Underactuated Aerial Vehicles: Design, Analysis, and Experimental Tests on Quadrotor

This paper investigates the global tracking control of underactuated aerial vehicles. In particular, the globally exponentially stable attitude tracking controller serving as the inner loop of the overall controller is investigated. It can avoid the common problems that may be accompanied by other attitude controllers, such as singularity and unwinding. In order to overcome the topological obstacles of global control on $SO(3)$ , rotational motion of a rigid body is expressed in the exponential coordinate restricted in a compact domain. We then construct a hybrid tracking error dynamics whose states are all represented in Euclidean space. The geometric properties of the state space are mathematically analyzed. The tracking controller of underactuated aerial vehicles is, thus, designed based on the hybrid tracking error dynamics. The global exponential stability of the closed-loop attitude subsystem, as well as the global asymptotical stability of the closed-loop overall system under assumptions, is analyzed using Lyapunov's method strictly. These properties are demonstrated by simulation results. The proposed controller is also implemented and tested on our self-developed quadrotor, showing the feasibility of the controller in realtime applications.

A hybrid approach to estimate the complex motions of clouds in sky images

Tracking the motion of clouds is essential to forecasting the weather and to predicting the short-term solar energy generation. Existing techniques mainly fall into two categories: variational optical flow, and block matching. In this paper, we summarize recent advances in estimating cloud motion using ground-based sky imagers and quantitatively evaluate state-of-the-art approaches. Then we propose a hybrid tracking framework to incorporate the strength of both block matching and optical flow models. To validate the accuracy of the proposed approach, we introduce a series of synthetic images to simulate the cloud movement and deformation, and thereafter comprehensively compare our hybrid approach with several representative tracking algorithms over both simulated and real images collected from various sites/imagers. The results show that our hybrid approach outperforms state-of-the-art models by reducing at least 30% motion estimation errors compared with the ground-truth motions in most of simulated image sequences. Moreover, our hybrid model demonstrates its superior efficiency in several real cloud image datasets by lowering at least 15% Mean Absolute Error (MAE) between predicted images and ground-truth images.