Enhanced particle-filtering framework for vessel segmentation and tracking

The paper aims at discussing techniques for administering one implementation issue that often arises in the application of particle filters: sample impoverishment. Dealing with such problem can significantly improve the performance of particle filters and can make the difference between success and failure. Sample impoverishment occurs because of the reduction in the number of truly distinct sample values. A simple solution can be to increase the number of particles, which can quickly lead to unreasonable computational demands, which only delays the inevitable sample impoverishment. There are more intelligent ways of dealing with this problem, such as roughening and prior editing, procedures to be discussed herein. The nonlinear particle filter is based on the bootstrap filter for implementing recursive Bayesian filters. The application consists of determining the orbit of an artificial satellite using real data from the GPS receivers. The standard differential equations describing the orbital motion and the GPS measurements equations are adapted for the nonlinear particle filter, so that the bootstrap algorithm is also used for estimating the orbital state. The evaluation will be done through convergence speed and computational implementation complexity, comparing the bootstrap algorithm results obtained for each technique that deals with sample impoverishment.

A lead-acid battery's remaining useful life prediction by using electrochemical model in the Particle Filtering framework

Particle Filter (PF) is considered as an extension of the Kalman filter to deal with non-Gaussian non-linear dynamic systems. The key idea is to construct a posterior probability by a set of hypotheses representing a potential state of the system. Sample impoverishment is the flaw of the PF techniques in the case of noisy observations. Recently, evolutionary algorithms are added to PF based methods to improve their functionality in noisy environments. A new PF based method called Adaptive Velocity Particle Filter (AVPF) is introduced to overcome this flaw of the PF. The experimental results show that the proposed method has a superior performance in relation to other evolutionary PF in cases of noisy observations.

BackgroundThe location of retinal vessels is an important prerequisite for Central Serous Chorioretinopathy (CSC) Laser Surgery, which does not only assist the ophthalmologist in marking the location of the leakage point (LP) on the fundus color image but also avoids the damage of the laser spot to the vessel tissue, as well as the low efficiency of the surgery caused by the absorption of laser energy by retinal vessels. In acquiring an excellent intra- and cross-domain adaptability, the existing deep learning (DL)-based vessel segmentation scheme must be driven by big data, which makes the densely annotated work tedious and costly.MethodsThis paper aims to explore a new vessel segmentation method with a few samples and annotations to alleviate the above problems. Firstly, a key solution is presented to transform the vessel segmentation scene into the few-shot learning task, which lays a foundation for the vessel segmentation task with a few samples and annotations. Then, we improve the existing few-shot learning framework as our baseline model to adapt to the vessel segmentation scenario. Next, the baseline model is upgraded from the following three aspects: (1) A multi-scale class prototype extraction technique is designed to obtain more sufficient vessel features for better utilizing the information from the support images; (2) The multi-scale vessel features of the query images, inferred by the support image class prototype information, are gradually fused to provide more effective guidance for the vessel extraction tasks; and (3) A multi-scale attention module is proposed to promote the consideration of the global information in the upgraded model to assist vessel localization. Concurrently, the integrated framework is further conceived to appropriately alleviate the low performance of a single model in the cross-domain vessel segmentation scene, enabling to boost the domain adaptabilities of both the baseline and the upgraded models.ResultsExtensive experiments showed that the upgraded operation could further improve the performance of vessel segmentation significantly. Compared with the listed methods, both the baseline and the upgraded models achieved competitive results on the three public retinal image datasets (i.e., CHASE_DB, DRIVE, and STARE). In the practical application of private CSC datasets, the integrated scheme partially enhanced the domain adaptabilities of the two proposed models.

Intelligent Trigonometric Particle Filter for visual tracking

Particle degeneracy and its poor ability to track saltatory states are two main problems when a particle filter is applied to fault prediction.The sequential importance re-sampling particle filter can abate the influence of particle degeneracy but will easily lead to another problem-sample impoverishment.The extended particle filter can resolve the problem of particle degeneracy to a certain extent but can not track the saltatory state.A strong tracking particle filter is put forward by introducing a strong tracking filter into a particle filter to resolve the above problems,in which the strong tracking filter is used to update particles and produce importance densities.As a result, the problems of particle degeneracy and sample impoverishment are ameliorated,and the tracking ability is improved. Simulation results demonstrate that the strong tracking particle filter is feasible and system fault can be predicted in time and correctly.

Particle filter is commonly used in various indoor positioning schemes, but sample impoverishment and weight degradation generally exist in particle filter. To solve this problem, this article uses the firefly algorithm to optimize particle filter. Using the global optimal value in the particle swarm to guide the remaining particles to update their positions, the particles tend to move to the high likelihood area, which can more accurately describe the true state of the target observation. At the same time, to avoid particles falling into local optimum and oscillating repeatedly at the extreme point, the mathematical model of the firefly algorithm is reconstructed in this article, which is named as adaptive optimization firefly algorithm (AOFA). The experimental results show that the positioning scheme can provide the positioning accuracy with an average error of less than 0.5 m. Compared with the traditional particle filter, the positioning accuracy is improved by 120%.

Particle Filter (PF) is a data assimilation method to solve recursive state estimation problem which does not depend on the assumption of Gaussian noise, and is able to be applied for various systems even with non-linear and non-Gaussian noise. However, while applying PF in dynamic systems, PF undergoes particle degeneracy, sample impoverishment, and problems of high computational complexity. Rapidly developing sensing technologies are providing highly convenient availability of real-time big traffic data from the system under study like never before. Moreover, some sensors can even receive control commands to adjust their monitoring parameters. To address these problems, a bidirectional dynamic data-driven improvement framework for PF (B3DPF) is proposed. The B3DPF enhances feedback between the simulation model and the big traffic data collected by the sensors, which means the execution strategies (sensor data management, parameters used in the weight computation, resampling) of B3DPF can be optimized based on the simulation results and the types and dimensions of traffic data injected into B3DPF can be adjusted dynamically. The first experiment indicates that the B3DPF overcomes particle degeneracy and sample impoverishment problems and accurately estimates the state at a faster speed than the normal PF. More importantly, the new method has higher accuracy for multidimensional random systems. In the rest of experiments, the proposed framework is applied to estimate the traffic state on a real road network and obtains satisfactory results. More experiments can be designed to validate the universal properties of B3DPF.

Surveillance of underwater for maritime warfare is traditionally being carried out by bearings-only tracking from many decades. The measurements used for state estimation here are nonlinear. Also the noise in the measurements and the process cannot be always Gaussian. The traditional nonlinear filtering algorithms like extended Kalman filter and modified gain extended Kalman filter use the linearisation of the system. The unscented Kalman filter (UKF) uses the sigma point approach based on Gaussian distribution to deal with nonlinearity. The particle filter (PF) uses the randomly generated particles based on the pdf of the state. PF is highly complex to implement and it also suffers from sample impoverishment. Hence, ensemble Kalman filter (EnKF) which is a simplified form of PF will be tried out for bearings-only tracking in this research work. The performance of EnKF is compared with PF and UKF and the results obtained using these filters in Matlab are presented.

The need for accurate, fast, and reliable indoor localization using wireless sensor networks (WSNs) has recently grown in diverse areas of industry. Accurate localization in cluttered and noisy environments is commonly provided by means of a mathematical algorithm referred to as a state estimator or filter. The particle filter (PF), which is the most commonly used filter in localization, suffers from the sample impoverishment problem under typical conditions of real-time localization based on WSNs. This paper proposes a novel hybrid particle/finite impulse response (FIR) filtering algorithm for improving reliability of PF-based localization schemes under harsh conditions causing sample impoverishment. The hybrid particle/FIR filter detects the PF failures and recovers the failed PF by resetting the PF using the output of an auxiliary FIR filter. Combining the regularized particle filter (RPF) and the extended unbiased FIR (EFIR) filter, the hybrid RP/EFIR filter is constructed in this paper. Through simulations, the hybrid RP/EFIR filter demonstrates its improved reliability and ability to recover the RPF from failures.

Joint 3-D vessel segmentation and centerline extraction using oblique Hough forests with steerable filters.

We demonstrate experimentally that inference in a complex hybrid dynamic Bayesian network (DBN) is possible using the 2-time slice DBN (2T-DBN) from (D. Koller et al., Sequential Monte Carlo Methods in Practice: p.445-464, Springer-Verlag, NY, 2000) to model fault detection in a watertank system. In this, a generic particle filter (PF) is used for inference. We extend the experiment and perform approximate inference using The extended Kalman filter (EKF) and the unscented Kalman filter (UKF). Furthermore, we combine these techniques in a 'non-strict' Rao-Blackwellisation framework and apply it to the watertank system. We show that UKF and UKF in a PF framework outperform the generic PF, EKF and EKF in a PF framework with respect to accuracy and robustness in terms of estimation RMSE (root-mean-square error). Especially, we demonstrate the superiority of UKF in a PF framework when our beliefs of how data was generated are wrong. We also show that the choice of network structure is very important for the performance of the generic PF and the EKF algorithms, but not for the UKF algorithms. Furthermore, we investigate the influence of data noise in the watertank simulation. Theory and implementation is based on the theory presented in (R. v.d. Merwe et al., Tech. Rep. CUED/F-INFENG/TR-380, Dept. of Eng., Cambridge Uni., 2000).

A hybrid sampling strategy is considered in multimode sampling based particle filter to alleviate the degeneracy as one of the most typical problems in the particle filter. However, to achieve high accuracy, expensive computation cost is inevitable when generating the hybrid distribution. To overcome this problem, a novel framework of particle filter is proposed in this paper with an improved hybrid sampling strategy. The main novelty is that this framework can simplify the generation of the hybrid distribution and makes the selection of particles more reasonable, in which the likelihood of particle is used to select the particles and determine the weights of multiple basic proposal distributions. Two simulation examples are implemented to test performances of the proposed filter algorithm. The obtained results show that the proposed framework has several superior performances in comparison with the standard particle filter, the unscented particle filter and the multimode sampling based particle filter.

Particle filter tracking, a type of sequential Monte Carlo method, has long been considered to be a very promising but time-consuming tracking technique. Methods have been developed to include a particle filter as part of a Variable Structure, Interactive Multiple Model (VS-IMM) structure and to integrate it into the Multiple Hypothesis Tracker (MHT) scoring structure. By integrating a particle filter as just one of many filters in Raytheon's MHT, the particle filter is applied sparingly on difficult off-road targets. This dramatically reduces the computation time as well as improves tracking performance in circumstances in which the other filters do not excel. Moreover, terrain information may be taken into account in the particle propagation process. In particular, an Unscented Particle Filter (UPF) was implemented in order to address the potential dominance of a small set of degenerate particles and/or poor prior distribution sampling from hampering the ability of the particle filter to accurately handle a maneuver. The Unscented Particle Filter treats every particle as its own Kalman filter. After the distribution of particles is adjusted in order to take into account the terrain, each particle is divided into sigma point states. These sigma points are propagated forward in time and then recombined to form a new composite particle state and covariance. These reformed particles are used in scoring and can be updated with a new observation. Since the Unscented Particle Filter includes the covariances in these calculations, this particle filter approach is more accurate and potentially requires fewer particles than an ordinary particle filter. By adding an Unscented Particle Filter to the other filters in an MHT tracker, the advantages of the UPF can be utilized in an efficient manner in order to enhance tracking performance.

A particle filter is a sequential Monte Carlo method that is superior in estimating the state of a dynamic system under nonlinear/non-Gaussian circumstance. Due to its nature, a particle filter has been regarded as an appropriate algorithm for localization. However, conventional problems, such as sample impoverishment and degeneracy problem, have not been perfectly solved yet. To solve these problems in mobile wireless environment, we propose an enhanced localization scheme, called Gaussian kernel density estimation-based particle filtering (GKPF), which calculates the target distribution (for location estimation) based on nonparametric technique. In order to estimate the target distribution, the GKPF algorithm creates both unimodal and multimodal distributions based on particle representations, and it calculates a pdf for each distribution with Gaussian kernel-density estimation. Simulation study indicates that the proposed GKPE scheme can accurately estimate the location in mobile wireless environment.