A phd Filter for Tracking Multiple Extended Targets Using Random Matrices

Granström, Karl; Orguner, Umut

doi:10.1109/tsp.2012.2212888

Cited by 258 publications

(279 citation statements)

References 22 publications

Supporting

Mentioning

278

Contrasting

Order By: Relevance

“…However, for many modern sensors the assumption no longer holds since targets may occupy multiple resolution cells in the sensor [59]. Commonly, such targets are referred to as extended.…”

Section: Extended Target Modelmentioning

confidence: 99%

See 1 more Smart Citation

Tracking of Animals Using Airborne Cameras

Veiback¹

2016

View full text Add to dashboard Cite

The various elements of a modern target tracking framework are covered. Background theory on pre-processing, modelling and estimation is presented as well as some novel ideas on the topic by the author. In addition, a few applications are posed as target tracking problems for which solutions are gradually constructed as relevant theory is covered.Among considered problems are how to constrain targets to a region, use stateindependent measurements to improve estimation in jump Markov models and how to incorporate observations sampled at an uncertain time into a state-space model.A framework is developed for tracking dolphins constrained to a basin using an overhead camera that suffers from occlusions. In this scenario, conventional motion models would suffer from infeasible predictions outside the basin. A motion model is developed for the dolphins where collisions with nearby walls are avoided by turning. The basin is modelled as a polygon where each point along the edge influences the turn rate of the dolphin. The proposed model results in predictions inside the basin, increasing robustness against occlusions. An extension to a Gaussian mixture background model providing a degree of confidence for detections is used to improve tracking in the presence of shadows. A probabilistic data association filter is also modified to estimate the dolphin extension as an ellipse. The proposed framework is able to maintain tracks through occlusions and poor light conditions. A framework is developed for estimating takeoff times and directions of birds in circular cages using an overhead camera. A jump Markov model is used to model the stationary and flight behaviours of the birds. A proposed extension also incorporates state-independent measurements, such as blurriness, to improve mode estimation. Takeoff times and directions are estimated from mode transitions and results are compared to manually annotated data.The cameras are inaccessible in both applications, disallowing proper calibrations. As an alternative, a method is proposed to estimate stationary camera models from available data and known features in the scene. A map of the basin and the funnel dimensions are used respectively. The method estimates a homography and distortion parameters in an invertible mapping function.An extension to the linear Gaussian state-space models is proposed, incorporating an additional observation with an uncertain timestamp. The posterior distribution of the states is derived for the model, which is shown to be a mixture of Gaussians, as well as some estimators for the distribution. The effects of incorporating the observation with an uncertain timestamp into the model are analysed for a one-dimensional scenario. The model is also applied to improve the GPS position of an orienteering sprinter by using the control position as an observation with an uncertain timestamp. Populärvetenskaplig sammanfattningMålföljning (eng. target tracking) är ett moget forskningsområde med anor tillbaka till åtminstone 30-talet. Då tävlade en ha...

show abstract

“…However, for many modern sensors the assumption no longer holds since targets may occupy multiple resolution cells in the sensor [59]. Commonly, such targets are referred to as extended.…”

Section: Extended Target Modelmentioning

confidence: 99%

“…In this thesis a model introduced in [79] is presented, where, in addition to the kinematic state vector, the extension is modelled as an ellipsoid represented by a random matrix. Although the exact shape of the target is not modelled this is sufficient for many applications [59].…”

Section: Extended Target Modelmentioning

confidence: 99%

Tracking of Animals Using Airborne Cameras

Veiback¹

2016

View full text Add to dashboard Cite

show abstract

“…Such target is referred to as an extended target in [4][5][6][7]. Recently, extended target tracking is a hot topic in the field of data fusion and has drawn a considerable attention, especially the multiple extendedtarget tracking (METT) [8][9][10][11][12][13][14].…”

Section: Introductionmentioning

confidence: 99%

“…It is considered that in this measurement model, the target is sufficiently far away from the sensor, and the measurements resemble a point cluster rather than a geometric structure [9][10][11]. Random matrix (RM) is proposed in [15], which has been used to track elliptical target extension [12][13][14]16,17]. Different methods to estimate star-convex target extension and other general shapes are the random hypersurface model (RHM) and star-convex RHM by Baum and Hanebeck [6,7].…”

Section: Introductionmentioning

confidence: 99%

Multiple extended target tracking algorithm based on GM-PHD filter and spectral clustering

Yang

Liu

et al. 2014

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

With the increase of the resolution of modern radars and other detection equipments, one target may produce more than one measurement. Such targets are referred to as extended targets. Recently, multiple extended target tracking (METT) has drawn a considerable attention. However, one crucial problem is how to partition the measurement sets accurately and rapidly. In this paper, an improved METT algorithm is proposed based on the Gaussian mixture probability hypothesis density (GM-PHD) filter and an effective partition method using spectral clustering technique. First, the density analysis technique is introduced to eliminate the disturbance of clutter, and then the spectral clustering technique based on neighbor propagation is used to partition the measurements. Finally, the GM-PHD filter is implemented to achieve the METT. Simulation results show that the proposed algorithm has a better performance, especially a better real-time performance, than the conventional distance partition and K-means++ methods.

show abstract

“…Several different approaches have been applied to the tracking of extended targets, with applications including the tracking of multiple extended targets and the joint estimation of the extent of the target as well as its motion parameters [9], [10]. An assumption made in these papers is that the observation sources are not fixed upon the target.…”

mentioning

confidence: 99%

Joint inference of dominant scatterer locations and motion parameters of an extended target in high range‐resolution radar

Freitas

Villiers

Nel

2015

IET Radar, Sonar & Navigation

View full text Add to dashboard Cite

A target of interest measured by a high range resolution radar may be modelled by multiple dominant points of reflections referred to as dominant scatterers. In this paper a non-linear state space setting is used to model the states and measurements of a target moving in the down-and cross-range dimensions. A resample-move particle filter with simulated annealing is successfully used to jointly infer the locations of the dominant scatterers and the motion parameters of the target. A novel technique for the initialization of the particle filter for the given application is presented. The location estimates of scatterers using the particle filter method are compared to those obtained using standard range-Doppler inverse synthetic aperture radar (ISAR) imaging when using the same radar returns for both cases. The particle filter infers the location of scatterers more accurately than range-Doppler ISAR processing, and the processing can be performed online as opposed to ISAR processing, which requires batching. It is relatively straightforward to extend the method to perform localisation and tracking of scatterers in three dimensions, whereas such an extension is challenging in range-Doppler ISAR processing. However, several challenges need be addressed to make this algorithm suitable for practical implementation and these challenges are discussed. This method may be used to obtain very accurate estimates of target state, which may in turn be used for accurate ISAR motion compensation. Given enough computing resources this algorithm may in future become the basis of a new radar target imaging scheme.

show abstract

A phd Filter for Tracking Multiple Extended Targets Using Random Matrices

Cited by 258 publications

References 22 publications

Tracking of Animals Using Airborne Cameras

Tracking of Animals Using Airborne Cameras

Multiple extended target tracking algorithm based on GM-PHD filter and spectral clustering

Joint inference of dominant scatterer locations and motion parameters of an extended target in high range‐resolution radar

Contact Info

Product

Resources

About