Multihypothesis tracking using incoherent signal-strength information

Keuk, G. van

doi:10.1109/7.532278

Cited by 21 publications

(13 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the spatial feature is not sufficient for the association cases where objects are closely spaced or clutter is densely distributed in the object vicinity. Therefore, for more accurate association, an amplitude is used as an extra feature in [17]- [21]. The basic idea of these methods is that an amplitude from an object is usually stronger than it from a clutter.…”

Section: Related Workmentioning

confidence: 99%

“…The basic idea of these methods is that an amplitude from an object is usually stronger than it from a clutter. The extended MHT [17] and Viterbi data association [18] using the amplitude are provided. In order to exploit the amplitude without the pre-knowledge of signalto-noise ratios (SNRs), a marginalization method [19] which computes an object amplitude likelihood within any SNR boundary is presented.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Online Multi-Object Tracking With Visual and Radar Features

Bae

2020

IEEE Access

View full text Add to dashboard Cite

Multi-object tracking (MOT) constructs multiple object trajectories by associating detections between consecutive frames while maintaining object identities. In many autonomous systems equipped with a camera and a radar, an amplitude and visual features can be measured. Therefore, our goal is to solve a MOT problem by associating detections with both features. To achieve it, we propose a unified MOT framework based on object model learning and confidence-based association. For improving discriminability between different objects, we present a method to learn several visual and amplitude object models during online tracking. By applying the learned object models for the affinity evaluation, we improve the confidence-based association further. In addition, we present a practical track management method to initialize and terminate tracks, and eliminate duplicated false tracks. We implement several MOT systems with different object model learning and association methods, and compare our system with them on challenging visual MOT datasets. We further compare our method with the recent deep appearance learning methods. These comparisons verify that our method can achieve the competitive tracking accuracy while maintaining a low MOT complexity.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Online Multi-Object Tracking With Visual and Radar Features

Bae

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In many sensors, such as radar and sonar, the measurements of targets position would also come with a signal strength, or amplitude, which, if be used as a measurement in target tracking in addition to the conventional measurements, can improve the performance of algorithms effectively [10][11][12]. [11] is the first literature to incorporate the target amplitude in the PHDF and gives its Gaussian mixture implementation.…”

Section: Introductionmentioning

confidence: 99%

PPHDF-AI for multitarget tracking with unknown target SNR

Tan

2013

2013 IEEE International Conference on Signal Processing, Communication and Computing (ICSPCC 2013)

View full text Add to dashboard Cite

The performance of multitarget tracking can be improved by using target amplitude information (AI) for that targets can be identified earlier through the enhanced discrimination between target and false alarm. However, one of the limitations of this application is that the signal-to-noise ratio (SNR) of target is usually unknown in practice. This paper proposes a particle probability hypothesis density filter (PPHDF) with amplitude information (PPHDF-AI) by extending the state vector and measurement vector of particle filter (PF) can be easily. By the extension of dynamic equation and measurement equation, the proposed algorithm can estimate targets SNR as well as their individual states. Simulation results demonstrate that the proposed method is quite suitable for situation that target SNR is unknown, and is superior to traditional methods.Index Terms-Probability hypothesis density filter (PHDF), multitarget tracking, amplitude information (AI), signal-to-noise ratio (SNR)

show abstract

“…Brekke et al [19,20] proposed a new conservative amplitude likelihood for the PDA filter with improved robustness, which attempted to detect and track moderately observable targets such as small boats or human divers in environments. Van Keuk [21] used amplitude information in MHT to improve the performance of MTT. For MTT approaches in the framework of FISST, Xu et al [22] proposed a method which involves using amplitude information in PHD filter to improve MTT performance, while Clark et al [23] illustrated an approach using the target amplitude for both the PHD and CPHD filters.…”

Section: Introductionmentioning

confidence: 99%

Amplitude-Aided CPHD Filter for Multitarget Tracking in Infrared Images

Qiu¹,

Zhang²,

Lü³

et al. 2014

PIER B

View full text Add to dashboard Cite

Abstract-The cardinalized probability hypothesis density (CPHD) filter is a powerful tool for multitarget tracking (MTT). However, conventional CPHD filter discriminates targets from clutter only via the motion information, which is not reasonable in the situation of dense clutter. In the tracking, the amplitude of target returns is usually stronger than those coming from clutter, so the amplitude information can be used to enhance the discrimination between targets and clutter. Based on this idea, this paper proposes an amplitude-aided CPHD filter for the MTT in distant infrared (IR) surveillance. First, we model the amplitude of targets and clutter in IR scenarios respectively. For distant IR scenarios, the point spread function (PSF) is used to model the imaging of the point target. The center intensity of the PSF is unknown in practice, and the maximum likelihood estimation (MLE) method is adopted to estimate the target center intensity via the intensities of the latest target detections. Then a likelihood function for MTT is established, and using this likelihood function, a new CPHD recursion is derived, which can distinguish different targets and clutter by the correspondence weight. In the implementation, we adopt the Gaussian mixture (GM) approach to implement the amplitude-aided CPHD filter to achieve efficient performance. In numerical experiments, the results show that the proposed method attains a significant improvement in performance over that only using location measurements.

show abstract

Multihypothesis tracking using incoherent signal-strength information

Cited by 21 publications

References 8 publications

Online Multi-Object Tracking With Visual and Radar Features

Online Multi-Object Tracking With Visual and Radar Features

PPHDF-AI for multitarget tracking with unknown target SNR

Amplitude-Aided CPHD Filter for Multitarget Tracking in Infrared Images

Contact Info

Product

Resources

About