Conventional Tracking Methods Research Articles

Path-following control method for articulated vehicles based on dual heuristic programming

A path-following control method based on dual heuristic programming (DHP) is proposed to address the problem whereby unmanned articulated vehicles need to set different controller parameters to track complex routes and cannot adapt to complex routes. First, the path-following control system structure is designed based on the articulated vehicle experimental platform, and the error model is derived based on the kinematic model of the articulated vehicle. Second, the payoff function is designed considering the error and stability indices, and the actor and critic of the path-following control method based on the DHP algorithm are approximated using a multilayer feedforward neural network. Finally, the path-following quality of the method is verified using simulations and real vehicle tests and compared with the conventional pure tracking and linear-quadratic regulator methods. The results show that the DHP-based method is able to follow complex reference paths and obtain better control results than the traditional methods without iterative tuning of the controller parameters, which improves the adaptiveness of the articulated vehicle path-following control method to complex environments.

Deep learning-based markerless lung tumor tracking in stereotactic radiotherapy using Siamese networks.

Radiotherapy (RT) is involved in about 50% of all cancer patients, making it a very important treatment modality. The most common type of RT is external beam RT, which consists of delivering the radiation to the tumor from outside the body. One novel treatment delivery method is volumetric modulated arc therapy (VMAT), where the gantry continuously rotates around the patient during the radiation delivery. Accurate tumor position monitoring during stereotactic body radiotherapy (SBRT) for lung tumors can help to ensure that the tumor is only irradiated when it is inside the planning target volume. This can maximize tumor control and reduce uncertainty margins, lowering organ-at-risk dose. Conventional tracking methods are prone to errors, or have a low tracking rate, especially for small tumors that are in close vicinity to bonystructures. We investigated patient-specific deep Siamese networks for real-time tumor tracking, during VMAT. Due to lack of ground truth tumor locations in the kilovoltage (kV) images, each patient-specific model was trained on synthetic data (DRRs), generated from the 4D planning CT scans, and evaluated on clinical data (x-rays). Since there are no annotated datasets with kV images, we evaluated the model on a 3D printed anthropomorphic phantom but also on six patients by computing the correlation coefficient with the breathing-related vertical displacement of the surface-mounted marker (RPM). For each patient/phantom, we used 80% of DRRs for training and 20% forvalidation. The proposed Siamese model outperformed the conventional benchmark template matching-based method (RTR): (1) when evaluating both methods on the 3D phantom, the Siamese model obtained a 0.57-0.79-mm mean absolute distance to the ground truth tumor locations, compared to 1.04-1.56 mm obtained by RTR; (2) on patient data, the Siamese-determined longitudinal tumor position had a correlation coefficient of 0.71-0.98 with the RPM, compared to 0.07-0.85 for RTR; (3) the Siamese model had a 100% tracking rate, compared to 62%-82% forRTR. Based on these results, we argue that Siamese-based real-time 2D markerless tumor tracking during radiation delivery is possible. Further investigation and development of 3D tracking iswarranted.

Tracking the sources of dissolved organic matter under bio- and photo-transformation conditions using fluorescence spectrum-based machine learning techniques

Dissolved organic matter (DOM) from various sources can lead to environmental issues such as eutrophication in agricultural watersheds. Effective source-tracking tools are needed to implement proper management practices. Fluorescence excitation–emission matrix (EEM) spectroscopy has been widely used to probe DOM composition. We explored optimal fluorescence EEM-based machine learning (ML) tools to quantify the proportions of different DOM sources in mixture samples under natural transformation conditions. Bulk DOM samples were prepared from soil and compost at various ratios and treated to simulate biogeochemical transformations. ML models based on all the EEM data outperformed those based on defined fluorescence indices. The trained support vector regression model (SVR) outperformed the conventional source tracking method of end-member mixing analysis (EMMA) with an R2 of 0.88 versus 0.83. Among the five suitable ML algorithms tested, SVR explained 90% and 85% of the variability in the proportions of soil and compost sources in the DOM mixture, with the mean squared errors of 0.004 and 0.007, respectively. The predicted capacity revealed a close relationship or causality between the specific mixing ratios of the bulk samples and the EEM spectra. The ML technique with EEM data was not constrained by the identification of all major sources, which is a required condition for the EMMA method. This study highlights the significant potential of EEM-based ML for tracing the source of DOM and establishes a basis for the future development of EEM data-driven models capable of tracking multiple DOM sources, even in the absence of all possible end-members.

Convolutional Neural Network-Based Speckle Tracking for Ultrasound Strain Elastography: An Unsupervised Learning Approach

Accurate and computationally efficient motion estimation is a critical component of real-time ultrasound strain elastography (USE). With the advent of deep-learning neural network models, a growing body of work has explored supervised convolutional neural network (CNN)-based optical flow in the framework of USE. However, the above-said supervised learning was often done using simulated ultrasound data. The research community has questioned whether simulated ultrasound data containing simple motion can train deep-learning CNN models that can reliably track complex in vivo speckle motion. In parallel with other research groups' efforts, this study developed an unsupervised motion estimation neural network (UMEN-Net) for USE by adapting a well-established CNN model named PWC-Net. Our network's input is a pair of predeformation and postdeformation radio frequency (RF) echo signals. The proposed network outputs both axial and lateral displacement fields. The loss function consists of a correlation between the predeformation signal and the motion-compensated postcompression signal, smoothness of the displacement fields, and tissue incompressibility. Notably, an innovative correlation method known as the globally optimized correspondence (GOCor) volumes module developed by Truong et al. was used to replace the original Corr module to enhance our evaluation of signal correlation. The proposed CNN model was tested using simulated, phantom, and in vivo ultrasound data containing biologically confirmed breast lesions. Its performance was compared against other state-of-the-art methods, including two deep-learning-based tracking methods (MPWC-Net++ and ReUSENet) and two conventional tracking methods (GLUE and BRGMT-LPF). In summary, compared with the four known methods mentioned above, our unsupervised CNN model not only obtained higher signal-to-noise ratios (SNRs) and contrast-to-noise ratios (CNRs) for axial strain estimates but also improved the quality of the lateral strain estimates.

A Novel Approach for GNSS Signal Tracking Based on Sliding Mode Control

Designing a robust carrier tracking loop for GNSS receivers is a challenging task under strong scintillation, multipath, clock uncertainty, interference, and vehicle high dynamics. The conventional tracking methods such as proportional integral filter (PIF), Wiener filter (WF), and Kalman filter (KF) are known to perform well under given GNSS signal model parameters and noise characteristics. If it is not the case, then performance deteriorates and may result in loss of the carrier lock. To circumvent this problem, we propose a robust control and filtering method based on the well-known sliding mode approach. In this paper, a sliding mode control and state estimation framework is developed for GNSS carrier tracking. The efficacy of the proposed method is proved by simulation studies for GPS(L1) signal. From static to high dynamics operating cases are simulated for proving its usage on the high dynamics vehicles such as high speed trains, flighter jets, and missile systems. The achieved results are promising for use of the proposed method on these vehicles and the receiver operating in the adverse (weak GNSS signal) conditions.

Passive Electro-Optical Tracking of Resident Space Objects for Distributed Satellite Systems Autonomous Navigation

Autonomous navigation (AN) and manoeuvring are increasingly important in distributed satellite systems (DSS) in order to avoid potential collisions with space debris and other resident space objects (RSO). In order to accomplish collision avoidance manoeuvres, tracking and characterization of RSO is crucial. At present, RSO are tracked and catalogued using ground-based observations, but space-based space surveillance (SBSS) represents a valid alternative (or complementary asset) due to its ability to offer enhanced performances in terms of sensor resolution, tracking accuracy, and weather independence. This paper proposes a particle swarm optimization (PSO) algorithm for DSS AN and manoeuvring, specifically addressing RSO tracking and collision avoidance requirements as an integral part of the overall system design. More specifically, a DSS architecture employing hyperspectral sensors for Earth observation is considered, and passive electro-optical sensors are used, in conjunction with suitable mathematical algorithms, to accomplish autonomous RSO tracking and classification. Simulation case studies are performed to investigate the tracking and system collision avoidance capabilities in both space-based and ground-based tracking scenarios. Results corroborate the effectiveness of the proposed AN technique and highlight its potential to supplement either conventional (ground-based) or SBSS tracking methods.

New Approach-based MPP Tracking Design for Standalone PV Energy Conversion Systems

In searching for a maximum power point (MPP) using a DC boost converter for photovoltaic (PV) energy conversion systems, we realised that the fast and accurate way to find the suitable duty ratio value is the core problem to enhance the energy conversion efficiency of the PV system. Under uniform irradiation, the panels will generate the same values, so they have only one peak on the P-V curve; conventional MPP tracking methods easily obtain this MPP. However, under partial shading conditions, many peaks are created, traditional MPP tracking methods can fall into the local MPP, and this issue will cause energy loss and reduce PV energy conversion efficiency. To avoid this disadvantage, this paper proposes a hybrid method (HM) by combining the improved chicken swarm optimisation (CSO) method and the incremental conductance (InC) algorithm for a DC standalone PV energy conversion system. In this hybrid method, the improved CSO modified approach is used to search the global region, and the InC algorithm is responsible for capturing the top of this global region. MATLAB simulation and experimental results were performed to demonstrate that the proposed method has achieved the global MPP under uniform solar irradiance and partial shadow effects.

Rethinking Multi-Object Tracking Based on Re-Identification and Appearance Model Management

Conventional multi-object tracking (MOT) methods often fail to track objects that repeatedly disappear and reappear in the scene due to occlusion or moving off the screen. Previous studies have used motion and appearance information to address these issues but have limitations in dealing with complex scenarios. To overcome these limitations, we propose a post-hoc framework for MOT, a so-called rethinking MOT. The proposed framework efficiently combines MOT and re-identification methods to handle tracking failures in conventional MOT methods. To this end, we proposed a track reliability checking method to determine unreliable tracks and performed re-identification to correct incorrect tracking results. For re-identification, we proposed appearance aspect-ratio matching that is practical and effective. In addition, we designed an online appearance model management for accurate and efficient re-identification. The proposed framework does not require any offline training stages, such as feature or metric learning, but it is very practical for comparing appearances in real time. In addition, it is a flexible and pluggable framework that can adopt any baseline methods for MOT and Re-id. To validate the proposed methods, we collected 12 new challenging indoor datasets. Unlike the existing benchmark MOT datasets, we consider complex indoor sequences where targets reappear many times in the scene. The experimental results reveal that the proposed methods are promising for MOT under various challenging sequences and outperform state-of-the-art methods. Compared to baseline of the proposed rethinking MOT, the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$IDF_{1}$ </tex-math></inline-formula> score was increased by 36.9% in own dataset and 50.4% in the benchmark dataset.

Maximum power point tracking based on improved spotted hyena optimizer for solar photovoltaic

&lt;span lang="EN-US"&gt;The conventional maximum power point tracking (MPPT) method such as perturb and observe (P&amp;amp;O) under partial shading conditions with non-uniform irradiation, can get trapped on local maximum power point (LMPP) and cannot reach global maximum power point (GMPP). This study proposes a bio-inspired metaheuristic algorithm spotted hyena optimizer (SHO) and improved SHO as a new MPPT technique. The proposed SHO-MPPT and improved SHO-MPPT are used to extract GMPP from solar photovoltaic (PV) arrays operated under uniform irradiation and non-uniform irradiation. Simulation with Powersim (PSIM) and experimental with the emulated PV source were presented. Furthermore, to evaluate the performance of the proposed algorithm, SHO-MPPT is compared with P&amp;amp;O-MPPT and particle swarm optimization (PSO)-MPPT. The SHO-MPPT has an accuracy of 99% and has the good capability, but there are power fluctuations before reaching MPP. Therefore, improved SHO-MPPT was developed to get better results. The improved SHO-MPPT proved high accuracy of 99% and faster than SHO-MPPT and PSO-MPPT in tracking the maximum power point (MPP). Furthermore, there are minor power fluctuations.&lt;/span&gt;

Hybrid MPPT approach using Cuckoo Search and Grey Wolf Optimizer for PV systems under variant operating conditions

Photovoltaic (PV) systems are adversely affected by partial shading and non-uniform conditions. Meanwhile, the addition of a bypass shunt diode to each PV module prevents hotspots. It also produces numerous peaks in the PV array’s power-voltage characteristics, thereby trapping conventional maximum power point tracking (MPPT) methods in local peaks. Swarm optimization approaches can be used to address this issue. However, these strategies have an unreasonably long convergence time. The Grey Wolf Optimizer (GWO) is a fast and more dependable optimization algorithm. This renders it a good option for MPPT of PV systems operating in varying partial shading. The conventional GWO method involves a long conversion time, large steady-state oscillations, and a high failure rate. This work attempts to address these issues by combining Cuckoo Search (CS) with the GWO algorithm to improve the MPPT performance. The results of this approach are compared with those of conventional MPPT according to GWO and MPPT methods based on perturb and observe (P&O). A comparative analysis reveals that under non-uniform operating conditions, the hybrid GWO CS (GWOCS) approach presented in this article outperforms the GWO and P&O approaches.

Passive tracking of underwater acoustic targets based on multi-beam LOFAR and deep learning.

Conventional passive tracking methods for underwater acoustic targets in sonar engineering generate time azimuth histogram and use it as a basis for target azimuth and tracking. Passive underwater acoustic targets only have azimuth information on the time azimuth histogram, which is easy to be lost and disturbed by ocean noise. To improve the accuracy of passive tracking, we propose to adopt the processed multi-beam Low Frequency Analysis and Recording (LOFAR) as the dataset for passive tracking. In this paper, an improved LeNet-5 convolutional neural network model (CNN) model is used to identify targets, and a passive tracking method for underwater acoustic targets based on multi-beam LOFAR and deep learning is proposed, combined with Extended Kalman Filter (EKF) to improve the tracking accuracy. The performance of the method under realistic conditions is evaluated through simulation analysis and validation using data obtained from marine experiments.

An ordered active parameter tracking method for efficient multiphase field simulations

The multiphase field (MPF) method has been widely used for microstructural evolution and immiscible multiphase flow simulations. In this study, we have developed an ordered active parameter tracking (OAPT) method for efficient simulations of the multiple phases using the MPF method. The algorithm and performance of the OAPT method are presented and compared to those of the conventional active parameter tracking (CAPT) method. The grain growth problems were simulated for demonstration, and the results show good agreement with the conventional one and about four to seven-fold performance improvement. Moreover, we demonstrate the application of the proposed method for the multiphase flow problem, namely the turbulent bubbly downflow. The OAPT method can speed up the calculation by four folds while maintaining the original accuracy.

Metaheuristic Algorithm Based Maximum Power Point Tracking Technique Combined With One Cycle Control For Solar Photovoltaic Water Pumping Systems

Solar photovoltaic technology has become eminent in the world because of its clean and abundant nature and can be effectively used for water pumping applications. A maximum power point tracking method is indispensable to achieve the best benefit from photovoltaic systems. Conventional maximum power tracking methods become successful only under uniform irradiance conditions and fail to track the maximum power under partial shading conditions (PSC). Hence, nature-inspired metaheuristic algorithms were proposed to track the optimum power under varying environmental conditions. This study proposes a method for improving the performance of the nature-inspired maximum power point tracking algorithms by using a nonlinear control technique called one cycle control. Based on the duty cycle obtained from the tracking algorithm the one cycle control technique generates pulses for the DC–DC converter, which is connected to a brushless DC motor pump system through a voltage source inverter. The performance of the proposed system under various PSCs is validated using the Cuckoo search, particle swarm optimization, and grey wolf algorithms. Simulation results for various shading patterns prove the supremacy of the system with respect to convergence speed, tracking efficiency, robustness, steady-state oscillations at maximum power point, and initial exploration oscillations in comparison with systems without one cycle control. In addition, the introduction of a KY converter as a DC–DC converter reduces the output voltage ripple in the system. Thus, the proposed system with one cycle control overcomes the disadvantages of the existing methods and can be effectively utilized for water pumping applications.

A rapid method for prediction of airborne disease infection risks in an intercity bus

In an intercity bus, respiratory infectious diseases put passengers at high risk of getting infected by the droplets exhaled by an infected person, and the risk increases when exposed to more droplets. Here, to quickly determine the concentration distribution of droplets, to predict the infection risks in a closed space, and to enhance the reliability of the conventional steady-state particle tracking method for predicting the trajectory of droplets released by coughing or sneezing, an improved steady-state particle tracking method is proposed. In it, the momentum of released droplets previously ignored in the conventional steady-state particle tracking method was specifically incorporated using experimental data. Then, the improved method was combined with a random walk model and applied to investigate all possible trajectories of droplets released by different passengers inside a bus. Consequently, the concentration distribution of droplets was obtained from the trajectory information. Finally, the Wells–Riley equation was used to predict the infection risk of every passenger based on the evaluated number of droplets inhaled per passenger. The results show that the improved steady-state tracking method performs more accurately at predicting the concentration field of droplets and associated infection risk than the conventional steady-state particle tracking method. Furthermore, the relative cost of the improved steady-state tracking method is just 1% of the transient calculation method currently considered the most accurate.

A Fast-Speed GMPPT Method for PV Array Under Gaussian Laser Beam Condition in Wireless Power Transfer Application

The laser power transfer system is one of the most promising systems in the long-range wireless power transfer field. However, the low efficiency at photovoltaic (PV) receiver limits the performance of an implemented system due to the Gaussian laser beam. Since the irradiance profile of the Gaussian laser beam is not uniform, the PV array exhibits complex output characteristics with multiple peaks, leading that the conventional maximum power point tracking (MPPT) method is not acceptable either on tracking accuracy or on tracking speed. In this article, a simple and rapid voltage-location global MPPT (VL-GMPPT) method is proposed for PV array under Gaussian laser beam condition (GLBC) in terms of reducing the voltage searching range. In order to do so, the mathematical model of PV array's global maximum power point (GMPP) is developed under GLBC. Then, the VL-GMPPT method incorporates voltage-location mechanism to directly move the operating point to the vicinity of GMPP and employs IncCond method to accurately find the GMPP. The experimental and simulation results are shown to verify the proposed GMPPT method. It was found that the proposed method can reduce 90% of tracking time that is consumed by the conventional global searching method.

Modified honey badger algorithm based global MPPT for triple-junction solar photovoltaic system under partial shading condition and global optimization

In the recent era, Metaheuristic Algorithms (MHs) are developed to tackle many and different optimization problems. The merit of the MHs is not only its simplicity to understand but also its easiness of implementation which helps in tackling different and various real-world applications. Honey Badger Algorithm (HBA) is one of the recent MHs. Although the success of the HBA algorithm in solving complex problems with high dimensions, it suffers from several drawbacks such as; 1) being trapped in local optima issue, 2) low convergence, and 3) the imbalance between exploration and exploitation stages. Therefore, an efficient local search called Dimensional Learning Hunting (DLH) is injected to the HBA to tackle these drawbacks, the proposed method named mHBA. On the other side, the demand of the PV systems (PVSs)’ installations is rapidly increasing from a day to the next. Due to the intermittency of the weather, the PV array performs as a nonlinear component in terms of its output characteristics. Partial shading is an environmental phenomenon that prevents all or part of the light to be exposed to the PV cells. This phenomenon makes the cell's output power curve fluctuates and hence contains multiple peaks and has an essential impact on the entire output of the PV system. However, identifying the peak that has the global maximum output power is the main target of most of the conventional maximum power point tracking (MPPT) methods. In this context, the higher efficiency of the global MPPT technique becomes a must to operate the PV cells as close as possible to the global MPPT. Accordingly, the performance of the proposed mHBA is assessed based on the complex 2020 IEEE Congress on Evolutionary Computation (CEC′20) test suite. Then, it is applied to track the global MPPT of PV system-based triple-junction solar cells (TJSCs) under partial shading conditions, four shading scenarios have been considered. The results of the proposed mHBA are compared with common MHs including Gray wolf optimizer (GWO), Moth-flame optimizer (MFO), Whale optimization algorithm, Sine cosine algorithm (SCA), Salp swarm algorithm (SSA), Tunicate swarm algorithm (TSA), Sooty Tern Optimization Algorithm (STOA), and the original HBA. The findings of this research proved the effectiveness and robustness of the proposed mBHA for solving the global optimization problems of the MPPT as an engineering application.

Underwater Bearing Only Tracking Using Optimal Observer Maneuver Strategies

This paper considers the problem of tracking a uniform moving source using noisy bearing measurements obtained from a distant observer. Observer trajectory optimization plays a central role in this problem, with the objective to minimize the estimation error of the target state. The Bearing Only Tracking (BOT) of passive targets is mainly focused on the observer maneuver with known trajectories and rarely focused on the future prediction of observer states using adaptive optimization strategies. In this paper, observer paths using one-step ahead optimization based on a performance index are devised which are potentially useful for longer horizon observer trajectory planning in passive tracking. This performance index is the function of source parameters termed as the determinant of Error Covariance Matrix (ECM) which is numerically more efficient than the determinant of Fisher Information Matrix (FIM). The determinant of the FIM requires the calculation of future values for target states and measurements rather than the current values, which is not feasible for Kalman like filters. Therefore, in this paper, the optimization technique is implemented using the state error covariance which is readily available through Kalman filter equations and does require separate numerical calculations. Due to optimal observer maneuver, the performance of the proposed algorithm does not depend on the initial conditions as compared to the conventional tracking methods. The efficiency of the evolutionary algorithm is calculated in terms of range, position and velocity errors and simulation results show 4% fewer estimation errors for ECM based optimization than the determinant of the FIM method.

A Fast and Accurate Frequency Tracking Method for Ultrasonic Cutting System via the Synergetic Control of Phase and Current.

Ultrasonic cutting is a superior machining process for brittle materials, owing to its capability to reduce the cutting force and improve the surface quality. To avoid the destructive instability of ultrasonic vibration induced by the cutting force, the excitation frequency of the ultrasonic system must reliably track its resonance frequency. However, it remains challenging for the conventional frequency tracking methods via one parameter to simultaneously achieve both high response rate and high tracking accuracy. This study proposes that more than one parameter could be coupled to get advantages from each parameter. A frequency tracking method via the synergetic control of circuit phase and current of the ultrasonic system was proposed as an example. This method utilizes the phase to responsively determine the tracking direction and uses the characteristic current as the endpoint frequency to ensure accuracy. Theoretical analyses and numerical simulations were conducted to demonstrate that the proposed method can accurately track the frequency of maximum vibration amplitude with a higher response rate than conventional methods. Moreover, ultrasonic cutting tests were performed on Nomex composites to evaluate the machining performance of the ultrasonic system with the proposed method. The experimental results verify that the proposed frequency tracking method enables the ultrasonic system to reach a stable state with a shorter response time, which is beneficial for the reduction of cutting-induced defects.

Optimized Fuzzy Controller Based on Cuckoo Optimization Algorithm for Maximum Power-Point Tracking of Photovoltaic Systems

The performance of a photovoltaic (PV) array depends on temperature, radiation, shading and load size. Conventional maximum power point tracking (MPPT) methods have acceptable efficiencies under uniform conditions (irradiance = 1000 W/m² and temprature = 25 &#x00B0;C), but in dynamic weather conditions, load changes, and also in partial shading conditions due to the presence of several local maximum power points (MPP) in the P-V characteristic, the conventional tracking method does not work well in finding the main MPP. To extract maximum power in all condi-tions, many algorithms have been proposed, all of which have limitations in terms of convergence speed, output power ripple and efficiency. This research proposes an optimized Fuzzy Logic Controller (FLC) based on the Cuckoo Opti-mization Algorithm (COA) for MPPT under uniform conditions, dynamic weather conditions, partial shading and under load changes. Finally, the research compared the simulation results with four other popular methods. According to the simulation observations and the result, COA-FLC overcomes the mentioned limitations such as low convergence speed, output power ripple and low tracking efficiency in all conditions. Simulations are performed with MATLAB / Simulink software.

Underwater Detection of Small-Volume Weak Target Echo in Harbor Scene Under Multisource Interference

Owing to the strong reverberation and various obstacle echoes interference in harbor scenes, the avtive detection performance of tracking methods for weak targets will be seriously reduced. Moreover, the conventional dereverberation and tracking methods generally cannot effectively separate the weak target under the overlapping echoes of multiple scattering sources. Focusing on solving the above problems, the correlation residual cumulative clustering (CRCC) algorithm is proposed to extract scattering features of weak target motion. There are two innovations in this paper: Firstly, the complex cepstrum filter is improved to suppress strong reverberation. Secondly, the correlation residual accumulation of adjacent frames is extracted to detect the reduction matrix containing the target trajectory. Finally, the FCM objective function is optimized via spatial-temporal constraint, thus the weak target can be effectively separated from the overlapping giant interference. The experimental results indicate the strong anti-jamming, low detection loss and short-term accumulation of our proposed model, which are remarkably superior to the traditional posterior probability models.