scholarly journals Using particle filter to track horizontal variations of atmospheric duct structure from radar sea clutter

2012 ◽  
Vol 5 (4) ◽  
pp. 6059-6082
Author(s):  
X. F. Zhao ◽  
S. X. Huang
Keyword(s):  
Particle Filter ◽  
Real Data ◽  
Forward Simulation ◽  
Sea Clutter ◽  
Radar Clutter ◽  
Atmospheric Duct ◽  
Basic Ideas ◽  
Recursive Bayesian Filter ◽  
Horizontal Variations ◽  
Recursive Bayesian

Abstract. This paper addresses the problem of estimating range-varying parameters of the height-dependent refractivity over the sea surface from radar sea clutter. In the forward simulation, the split-step Fourier parabolic equation (PE) is used to compute the radar clutter power in the complex refractive environments. Making use of the inherent Markovian structure of the split-step Fourier PE solution, the refractivity from clutter (RFC) problem is formulated within a nonlinear recursive Bayesian state estimation framework. Particle filter (PF) that is a technique for implementing a recursive Bayesian filter by Monte Carlo simulations is used to track range-varying characteristics of the refractivity profiles. Basic ideas of employing PF to solve RFC problem are introduced. Both simulation and real data results are presented to check up the feasibility of PF-RFC performances.

scholarly journals Using particle filter to track horizontal variations of atmospheric duct structure from radar sea clutter

2012 ◽  
Vol 5 (11) ◽  
pp. 2859-2866 ◽  
Author(s):  
X. F. Zhao ◽  
S. X. Huang ◽  
D. X. Wang
Keyword(s):  
Particle Filter ◽  
Real Data ◽  
Forward Simulation ◽  
Sea Clutter ◽  
Radar Clutter ◽  
Atmospheric Duct ◽  
Basic Ideas ◽  
Recursive Bayesian Filter ◽  
Horizontal Variations ◽  
Recursive Bayesian

Abstract. This paper addresses the problem of estimating range-varying parameters of the height-dependent refractivity over the sea surface from radar sea clutter. In the forward simulation, the split-step Fourier parabolic equation (PE) is used to compute the radar clutter power in the complex refractive environments. Making use of the inherent Markovian structure of the split-step Fourier PE solution, the refractivity from clutter (RFC) problem is formulated within a nonlinear recursive Bayesian state estimation framework. Particle filter (PF), which is a technique for implementing a recursive Bayesian filter by Monte Carlo simulations, is used to track range-varying characteristics of the refractivity profiles. Basic ideas of employing PF to solve RFC problem are introduced. Both simulation and real data results are presented to confirm the feasibility of PF-RFC performances.

scholarly journals Estimation of Atmospheric Duct Structure Using Radar Sea Clutter

2012 ◽  
Vol 69 (9) ◽  
pp. 2808-2818 ◽  
Author(s):  
Xiaofeng Zhao ◽  
Sixun Huang
Keyword(s):  
Approximate Solutions ◽  
Propagation Loss ◽  
Sea Surface ◽  
Model Parameters ◽  
Sea Clutter ◽  
Radar Clutter ◽  
Atmospheric Duct ◽  
Analytical Transformation ◽  
Ill Posed ◽  
Adjoint Approach

Abstract Retrieving atmospheric refractivity profiles from the sea surface backscattered radar clutter is known as the refractivity-from-clutter (RFC) technique. Because the relationship between refractivity and radar sea clutter is clearly nonlinear and ill posed, it is difficult to get analytical solutions according to current theories. Previous works treat this problem as a model parameter estimation issue and some optimization algorithms are selected to get approximate solutions. Two main factors that limit the accuracy of the estimation are that 1) the refractive environments are described by using some idealized refractivity parameter models that cannot describe the exact information of the refractivity profile, and 2) accurate modeling of the sea surface radar cross section (RCS) is very difficult. Rather than estimating a few model parameters, this paper puts forward possibilities of using the variational adjoint approach to jointly retrieve the every-height refractivity values and sea surface RCS using radar clutter data. The derivation of the adjoint model is accomplished by an analytical transformation of the parabolic equation (PE) in the continuous domain. Numerical simulations including range-independent and range-dependent RCS cases are presented to demonstrate the ability of this method for RFC estimations. Making use of the refractivity retrievals, propagation loss predictions are also presented.

scholarly journals Atmospheric Duct Estimation Using Radar Sea Clutter Returns by the Adjoint Method with Regularization Technique

2014 ◽  
Vol 31 (6) ◽  
pp. 1250-1262 ◽  
Author(s):  
Xiaofeng Zhao ◽  
Sixun Huang
Keyword(s):  
Genetic Algorithm ◽  
Parabolic Equation ◽  
Adjoint Method ◽  
Regularization Parameter ◽  
Background Field ◽  
Sea Clutter ◽  
Regularization Technique ◽  
Atmospheric Duct ◽  
Adjoint Approach ◽  
Space Range

Abstract This paper focuses on retrieving the atmospheric duct structure from radar sea clutter returns by the adjoint approach with the regularization technique. The adjoint is derived from the split-step Fourier parabolic equation method, and the regularization term is constructed by the background refractivity field. To ensure successful implementations of the regularization, the L-curve criterion is used to find the optimal regularization parameter. The feasibility of the proposed method is validated by the numerical simulations of different noise-level clutter returns, as well as a real clutter profile measured by the S-Band Space Range Radar located in Wallops Island. In the process of inversions, the refractivity profile is first obtained by genetic algorithm, and then it is used as the background field for the adjoint method. The retrieved results indicate that, with an appropriate regularization parameter, the structure of the background refractivity profile can be improved by the proposed method.

scholarly journals Artificial intelligence techniques for small boats detection in radar clutter. Real data validation

2018 ◽  
Vol 67 ◽  
pp. 296-308 ◽  
Author(s):  
Nerea del-Rey-Maestre ◽  
David Mata-Moya ◽  
María-Pilar Jarabo-Amores ◽  
Pedro-Jose Gomez-del-Hoyo ◽  
Jose-Luis Barcena-Humanes
Keyword(s):  
Artificial Intelligence ◽  
Real Data ◽  
Data Validation ◽  
Radar Clutter ◽  
Artificial Intelligence Techniques

scholarly journals 3D Human Pose Estimation with a Catadioptric Sensor in Unconstrained Environments Using an Annealed Particle Filter

Sensors ◽  
2020 ◽  
Vol 20 (23) ◽  
pp. 6985
Author(s):  
Fakhreddine Ababsa ◽  
Hicham Hadj-Abdelkader ◽  
Marouane Boui
Keyword(s):  
Particle Filter ◽  
Vision System ◽  
Real Data ◽  
Human Detection ◽  
Human Model ◽  
Svm Classifier ◽  
Likelihood Functions ◽  
Human Pose ◽  
Spherical Images ◽  
Omnidirectional Images

The purpose of this paper is to investigate the problem of 3D human tracking in complex environments using a particle filter with images captured by a catadioptric vision system. This issue has been widely studied in the literature on RGB images acquired from conventional perspective cameras, while omnidirectional images have seldom been used and published research works in this field remains limited. In this study, the Riemannian varieties was considered in order to compute the gradient on spherical images and generate a robust descriptor used along with an SVM classifier for human detection. Original likelihood functions associated with the particle filter are proposed, using both geodesic distances and overlapping regions between the silhouette detected in the images and the projected 3D human model. Our approach was experimentally evaluated on real data and showed favorable results compared to machine learning based techniques about the 3D pose accuracy. Thus, the Root Mean Square Error (RMSE) was measured by comparing estimated 3D poses and truth data, resulting in a mean error of 0.065 m when walking action was applied.

Sign in / Sign up

Export Citation Format

Share Document