&lt;title&gt;Multicriterion vehicle pose estimation for SAR ATR&lt;/title&gt;

Voicu, Letitia; Patton, R.J.; Myler, Harley R.

doi:10.1117/12.357666

Cited by 32 publications

(19 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This paper reports our results of pose estimation for MSTAR/IU database, based on our proposed method of pose estimation using information theoretic features [2], [6]. The need for such an effective algorithm stems from our goal of The preliminary results of our method show a maximum mean error of 8 degrees of azimuth, when the training images are taken at 3.5 degree increments.…”

Section: Discussionmentioning

confidence: 98%

See 1 more Smart Citation

<title>Statistical pose estimation of land targets in SAR</title>

Prı́ncipe

Learn

et al. 2000

SPIE Proceedings

View full text Add to dashboard Cite

This paper explores statistical pose estimation in SAR ATR using a recently proposed training method based on information theory. The theory of training with information theoretic learning is briefly summarized. Different pose estimator topologies and training criteria are employed. Experimental results in the MSTAR I and II show that our proposed method is capable of producing 1-DOF and 2-DOF pose estimations and we show the dependence of the training parameters on performance.

show abstract

Section: Discussionmentioning

confidence: 98%

“…This method may be accurate but it is very time consuming. Alternatively some features of the target can be identified such as the leading bright edge of the target and its shadow and pose computed from them [2]. These methods have the problem of using heuristics to determined the features used for pose estimation.…”

Section: Introductionmentioning

confidence: 99%

<title>Statistical pose estimation of land targets in SAR</title>

Prı́ncipe

Learn

et al. 2000

SPIE Proceedings

View full text Add to dashboard Cite

show abstract

“…By eqs. (4) and (6), the estimation of X AO and X BO actually provides the information of the rotating center [13]. For radar A, from eq.…”

Section: Rotating Center Estimationmentioning

confidence: 99%

“…According to the ISAR imaging theory, the resolution along RLOS can be improved by signals with larger bandwidth, and the cross-range resolution can be improved by wider aspect observations. It is known that high resolution imaging results of multi-aspect can improve the performance of target identification [4]. Also, observations from multi-aspect can be used to form 2D images with better resolution [5] or to perform 3D reconstruction of a target [6][7][8], and enable more reliable target identification [6].…”

Section: Introductionmentioning

confidence: 99%

Key parameter estimation for radar rotating object imaging with multi-aspect observations

Chunmao

Peng

et al. 2010

Sci. China Inf. Sci.

View full text Add to dashboard Cite

Rotating object model is commonly used for imaging analysis in high resolution radars such as the inverse synthetic aperture radar (ISAR). For a rotating object, it is known that multi-aspect observations can improve cross-range resolution with the known imaging geometry. For the non-cooperative rotating object with unknown imaging geometry, this paper proposes an integrated scheme to estimate the key parameters, e.g., the rotating velocity and the aspect angle difference between every two observations. Furthermore, convolution back-projection (CBP) method is applied to provide fused imaging result with improved resolution. Also, the accuracy of the ultimate parameter estimation is analyzed, which is strongly related with several important factors like position extraction error of scattering centers and so on. Finally, the results of numerical experiments are provided to demonstrate the effectiveness of the proposed method. CitationYe C M, Xu J, Peng Y N, et al. Key parameter estimation for radar rotating object imaging with multi-aspect observations.

show abstract

“…This method results in a tedious process of pose estimation, and the accuracy is not prefectP [4] P.…”

Section: Introductionmentioning

confidence: 99%

A Practical Kernel Criterion for Feature Extraction and Recognition of MSTAR SAR Images

Cheng

Zhao

Zhang

et al. 2006

2006 8th International Conference on Signal Processing

View full text Add to dashboard Cite

Complete kernel fisher discriminant analysis (CKFDA) is essentially a practical nonlinear feature extraction criterion based on kernel trick. The process is divided into two phases, i.e., kernel principal component analysis (KPCA) and linear discriminant analysis (LDA). This work uses two different kinds of CKFDA methods to extract the features of MSATR SAR images: one only obtains the regular information in "single discriminant space", the other gains regular and irregular information in "double discriminant subspaces". The inspiring recognition results verify that the features not only overcome aspect sensitivity existent in SAR images, but also are robust to variants within the target classes which have small configuration differences.

show abstract

<title>Multicriterion vehicle pose estimation for SAR ATR</title>

Cited by 32 publications

References 0 publications

<title>Statistical pose estimation of land targets in SAR</title>

<title>Statistical pose estimation of land targets in SAR</title>

Key parameter estimation for radar rotating object imaging with multi-aspect observations

A Practical Kernel Criterion for Feature Extraction and Recognition of MSTAR SAR Images

Contact Info

Product

Resources

About