A MIMO-SAR Tomography Algorithm Based on Fully-Polarimetric Data

Kong, Lingyu; Xu, Xiaojian

doi:10.3390/s19224839

“…To validate the feasibility of the proposed technique, this section presents some 3-D reconstruction results from both numerical and measurement datasets. The following algorithms, i.e., the superfast line spectral estimation (Superfast LSE) [39] and the polarimetric UMUSIC (P-UMUSIC) [19], are used for comparison. Notice that all the reference algorithms are only applied for the elevation inversion process since all experimental datasets need to be preprocessed by the steps described in Section III-A and III-B.…”

Section: Resultsmentioning

confidence: 99%

A Fast ISAR Tomography Technique for Fully Polarimetric 3-D Imaging of Man-Made Targets

Wu

¹

,

Xu

²

2022

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

The 3-D inverse synthetic aperture radar (ISAR) tomography is an enabling technique for applications such as the exact diagnosis of scattering mechanisms for complex targets. Nevertheless, current ISAR tomography solutions still suffer from problems such as the great computational complexity and optimal utilization of the signatures acquired from limited baselines. In this work, we propose a fast ISAR tomography technique for fully polarimetric 3-D imaging of man-made targets. A stack of 2-D complex-valued images with different baselines and polarizations is first obtained through a phase error calibration (PEC) process and graphic processing unit accelerated polarimetric filtered backprojection. A polarimetric state-space decomposition (P-SSD) algorithm is then developed which could provide joint 3-D reconstruction results with low computational complexity. Examples from both numerical multibaseline data for the Sandia laboratories implementation of cylinders (SLICY) benchmark model and the outdoor range dataset collected by the Georgia Tech Research Institute (GTRI) are presented to demonstrate the superior performance and the usefulness of the proposed technique.

show abstract

“…One of its primary applications is the detection and identi-22 fication of various military targets [1,2]. With the enhancement of SAR data acquisition 23 capability, Synthetic Aperture Radar Automatic Target Recognition (SAR ATR) [3] has 24 become a key technology and research hotspot of radar signal processing. Traditional 25 SAR target recognition methods [4] merely rely on artificial experience for feature extrac-26 tion and selection, which lead to a certain degree of subjectivity and bias.…”

Section: Introduction 19mentioning

confidence: 99%

“…However, directly replacing some pixels with black may produce high-frequency signal processing, we generally consider the noise of SAR image as multiplicative noise 361 [3,6]. Figure 10 shows the above processing of the same image.…”

mentioning

confidence: 99%

C-RISE: A Post-hoc Interpretation Method of Black-box Models for SAR ATR

Zhu¹,

Cheng²,

Lei³

et al. 2023

Preprint

1

0

View full text Add to dashboard Cite

The integration of deep learning methods, especially Convolutional Neural Networks (CNN), and Synthetic Aperture Radar Automatic Target Recognition (SAR ATR) has been widely deployed in the field of radar signal processing. Nevertheless, these methods are frequently regarded as black-box models due to the limited visual interpretation of their internal feature representation and parameter organization. In this paper, we propose an innovative approach named C-RISE, which builds upon the RISE algorithm to provide a post-hoc interpretation technique for black-box models used in SAR Images Target Recognition. C-RISE generates saliency maps that effectively visualize the significance of each pixel. Our algorithm outperforms RISE by clustering masks that capture similar fusion features into distinct groups, enabling more appropriate weight distribution and increased focus on the target area. Furthermore, we employ Gaussian blur to process the masked area, preserving the original image structure with optimal consistency and integrity. C-RISE has been extensively evaluated through experiments, and the results demonstrate superior performance over other interpretation methods based on perturbation when applied to neural networks for SAR image target recognition. Furthermore, our approach is highly robust and transferable compared to other interpretable algorithms, including white-box methods.

show abstract

“…One of its primary applications is the detection and identification of various military targets [1,2]. With the enhancement of SAR data acquisition capability, Synthetic Aperture Radar Automatic Target Recognition (SAR ATR) [3] has become a key technology and research hotspot of radar signal processing. Traditional SAR image recognition methods, such as template matching [4], feature-based approaches [5,6], and CAD model-based methods [7], predominantly rely on the statistical and physical characteristics inherent in the image data.…”

Section: Introductionmentioning

confidence: 99%

C-RISE: A Post-Hoc Interpretation Method of Black-Box Models for SAR ATR

Zhu

¹

,

Cheng

²

,

Lei

³

et al. 2023

Remote Sensing

3

0

View full text Add to dashboard Cite

The integration of deep learning methods, especially Convolutional Neural Networks (CNN), and Synthetic Aperture Radar Automatic Target Recognition (SAR ATR) has been widely deployed in the field of radar signal processing. Nevertheless, these methods are frequently regarded as black-box models due to the limited visual interpretation of their internal feature representation and parameter organization. In this paper, we propose an innovative approach named C-RISE, which builds upon the RISE algorithm to provide a post-hoc interpretation technique for black-box models used in SAR Images Target Recognition. C-RISE generates saliency maps that effectively visualize the significance of each pixel. Our algorithm outperforms RISE by clustering masks that capture similar fusion features into distinct groups, enabling more appropriate weight distribution and increased focus on the target area. Furthermore, we employ Gaussian blur to process the masked area, preserving the original image structure with optimal consistency and integrity. C-RISE has been extensively evaluated through experiments, and the results demonstrate superior performance over other interpretation methods based on perturbation when applied to neural networks for SAR image target recognition. Furthermore, our approach is highly robust and transferable compared to other interpretable algorithms, including white-box methods.

show abstract

A MIMO-SAR Tomography Algorithm Based on Fully-Polarimetric Data

Cited by 5 publications

References 20 publications

A Fast ISAR Tomography Technique for Fully Polarimetric 3-D Imaging of Man-Made Targets

A Fast ISAR Tomography Technique for Fully Polarimetric 3-D Imaging of Man-Made Targets

C-RISE: A Post-hoc Interpretation Method of Black-box Models for SAR ATR

C-RISE: A Post-Hoc Interpretation Method of Black-Box Models for SAR ATR

Contact Info

Product

Resources

About