Refined GM-PHD tracker for tracking targets in possible subsequent missed detections

Yazdian-Dehkordi, Mahdi; Azimifar, Zohreh

doi:10.1016/j.sigpro.2015.04.008

Cited by 25 publications

(10 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is acceptable because the situation of missed detections and target disappearances are difficult to distinguish within a few time steps. For practical MTT problems, there always exists a trade-off between alleviating performance degradation caused by missed detection and providing an excellent performance for track continuity [3,24], and it is also not advisable to declare target death prematurely. Empirically, considering the inherent inertia of the CPHD filter in response to cardinality changes [12],

2 \leq N_{W} \leq 4

is suggested for the purpose of reducing the spooky effect and preventing the adverse influence on the response speed to target disappearance.…”

Section: Improved Gm-cphd Filtermentioning

confidence: 99%

“…Based on the proposed N-step attenuation function, the following heuristic gating strategy is proposed to adjust the gate threshold:

{\overset{T}{true}}_{g} = {\begin{matrix} left \\ \frac{T_{g}}{α_{normalP} (n_{miss}^{j})} & β^{j} = 1, n_{miss}^{j} \neq 0, n_{miss}^{j} \leq N_{W} \\ T_{g} & otherwise \end{matrix}

where the influence of consecutive missed detection is considered for the confirmed GM components. The basic principle behind the proposed gating method is that the undetected targets in the previous time step will have a larger statistical error in their predicted measurements than those of the detected targets to some degree [24]. Therefore, we allow a large gate size, which can be adaptive in accordance with the degree of missed detection, for the confirmed component with missed detection, such that it can be more conservative in the region where the corresponding measurement may appear.…”

Section: Improved Gm-cphd Filtermentioning

confidence: 99%

See 1 more Smart Citation

Multi-Target Tracking Using an Improved Gaussian Mixture CPHD Filter

Wang

2016

Sensors

View full text Add to dashboard Cite

The cardinalized probability hypothesis density (CPHD) filter is an alternative approximation to the full multi-target Bayesian filter for tracking multiple targets. However, although the joint propagation of the posterior intensity and cardinality distribution in its recursion allows more reliable estimates of the target number than the PHD filter, the CPHD filter suffers from the spooky effect where there exists arbitrary PHD mass shifting in the presence of missed detections. To address this issue in the Gaussian mixture (GM) implementation of the CPHD filter, this paper presents an improved GM-CPHD filter, which incorporates a weight redistribution scheme into the filtering process to modify the updated weights of the Gaussian components when missed detections occur. In addition, an efficient gating strategy that can adaptively adjust the gate sizes according to the number of missed detections of each Gaussian component is also presented to further improve the computational efficiency of the proposed filter. Simulation results demonstrate that the proposed method offers favorable performance in terms of both estimation accuracy and robustness to clutter and detection uncertainty over the existing methods.

show abstract

2 \leq N_{W} \leq 4

is suggested for the purpose of reducing the spooky effect and preventing the adverse influence on the response speed to target disappearance.…”

Section: Improved Gm-cphd Filtermentioning

confidence: 99%

“…Based on the proposed N-step attenuation function, the following heuristic gating strategy is proposed to adjust the gate threshold:

{\overset{T}{true}}_{g} = {\begin{matrix} left \\ \frac{T_{g}}{α_{normalP} (n_{miss}^{j})} & β^{j} = 1, n_{miss}^{j} \neq 0, n_{miss}^{j} \leq N_{W} \\ T_{g} & otherwise \end{matrix}

Section: Improved Gm-cphd Filtermentioning

confidence: 99%

Multi-Target Tracking Using an Improved Gaussian Mixture CPHD Filter

Wang

2016

Sensors

View full text Add to dashboard Cite

show abstract

“…Yazdian-Dehkordi and Azimifar proposed a GM-PHD filter that adopts state-dependent probability of survival. To track and report undetected targets, each Gaussian component is assigned a probability of confirming on which track management is based [ 26 ]. Wang et al presented a SMC-PHD filter that is suitable for scenarios with low probability of detection.…”

Section: Introductionmentioning

confidence: 99%

PHD Filter for Object Tracking in Road Traffic Applications Considering Varying Detectability

Törő

Bécsi

Gáspár

2021

Sensors

View full text Add to dashboard Cite

This paper considers the object detection and tracking problem in a road traffic situation from a traffic participant’s perspective. The information source is an automotive radar which is attached to the ego vehicle. The scenario characteristics are varying object visibility due to occlusion and multiple detections of a vehicle during a scanning interval. The goal is to maintain and report the state of undetected though possibly present objects. The proposed algorithm is based on the multi-object Probability Hypothesis Density filter. Because the PHD filter has no memory, the estimate of the number of objects present can change abruptly due to erroneous detections. To reduce this effect, we model the occlusion of the object to calculate the state-dependent detection probability. Thus, the filter can maintain unnoticed but probably valid hypotheses for a more extended period. We use the sequential Monte Carlo method with clustering for implementing the filter. We distinguish between detected, undetected, and hidden particles within our framework, whose purpose is to track hidden but likely present objects. The performance of the algorithm is demonstrated using highway radar measurements.

show abstract

“…Despite a lack of trajectory of the target, these filters have been applied to autonomous vehicle [12], sonar tracking [13] and radar tracking [14]. Their extensions have been developed to track the maneuvering target [15]- [17], the extended target [18]- [21], the target with glint noise [22] and the target in possible subsequent missed detections [23]. Detection-guided Bayesian filter [24] and parallel implementation of the sequential Monte Carlo PHD filter [25] were also designed to obviate the need for exact knowledge of birth objects and to improve the computational efficiency of the sequential Monte Carlo PHD filter, respectively.…”

Section: Introductionmentioning

confidence: 99%

Marginal Distribution Multi-Target Bayes Filter With Assignment of Measurements

Liu

Gan

2020

IEEE Access

View full text Add to dashboard Cite

This work proposes a marginal distribution multi-target Bayes filter with assignment of measurements to track multiple targets in the presence of an unknown and variable number of targets, clutter, and missed detections. Mathematically, the association of the measurements with either a target or clutter may be established by maximizing the joint likelihood function of the measurement partition, which leads to a two-dimensional assignment problem. By the introduction of detecting label, a handling approach for missed detections is also developed and is applied to the proposed filter. This filter greatly reduces the number of hypothesized targets or Gaussian terms by selecting the predicted probability density of a target or one of its multiple updated probability densities as its state distribution at each time step. Experimental results indicate that the proposed filter requires a less computational load than the existing filters and performs better than the efficient implementations of the δ-generalized labeled multi-Bernoulli filter for multi-target tracking at low and moderate clutter densities. INDEX TERMS multi-target tracking, marginal distribution multi-target Bayes filter, generalized labeled multi-Bernoulli filter, two-dimensional assignment.

show abstract

Refined GM-PHD tracker for tracking targets in possible subsequent missed detections

Cited by 25 publications

References 38 publications

Multi-Target Tracking Using an Improved Gaussian Mixture CPHD Filter

Multi-Target Tracking Using an Improved Gaussian Mixture CPHD Filter

PHD Filter for Object Tracking in Road Traffic Applications Considering Varying Detectability

Marginal Distribution Multi-Target Bayes Filter With Assignment of Measurements

Contact Info

Product

Resources

About