ISAR Compressive Sensing Imaging Using Convolution Neural Network with Interpretable Optimization

Compressive sensing (CS) has been widely utilized in inverse synthetic aperture radar (ISAR) imaging, since ISAR measured data are generally non-completed in cross-range direction, and CS-based imaging methods can obtain high-quality imaging results using under-sampled data. However, the traditional CS-based methods need to pre-define parameters and sparse transforms, which are tough to be hand-crafted. Besides, these methods usually require heavy computational cost with large matrices operation. In this paper, inspired by the adaptive parameter learning and rapidly reconstruction of convolution neural network (CNN), a novel imaging method, called convolution iterative shrinkage-thresholding (CIST) network, is proposed for ISAR efficient sparse imaging. CIST is capable of learning optimal parameters and sparse transforms throughout the CNN training process, instead of being manually defined. Specifically, CIST replaces the linear sparse transform with non-linear convolution operations. This new transform and essential parameters are learnable end-to-end across the iterations, which increases the flexibility and robustness of CIST. When compared with the traditional state-of-the-art CS imaging methods, both simulation and experimental results demonstrate that the proposed CIST-based ISAR imaging method can obtain imaging results of high quality, while maintaining high computational efficiency. CIST-based ISAR imaging is tens of times faster than other methods.

Inverse synthetic aperture radar (ISAR) imaging is typically useful when there is a need to classify, recognize, or identify a moving target of interest. In fact, an ISAR image highlights two-dimensional (2-D) geometric features of a target, which can provide indications of target's type, size, and other salient information. Such information can be then used for target classification, recognition, and identification. In this chapter, we will provide five case studies on applications of ISAR imaging. Such case studies are chosen to diversify the type of application and to give additional indications about how to interpret ISAR images. Specifically, in the first case study (Section 12.1), we show an example of ground-based ISAR images of a noncooperative sailing ship. The ISAR techniques used to form the ISAR images are the range-Doppler (Chapter 3) and the ICBA (Chapter 5) techniques. Different time windows are chosen to show the effects of the coherent processing interval (CPI) length on the ISAR image. Since the radar is ground based, this represents a case where the ISAR image is formed by exploiting only the sea surface induced target's motions. The second case study (Section 12.2) refers to a scenario where the radar is carried by an aircraft and the target is a ship at sea. In this case, both the target and the platform motions concur to the ISAR image formation. Also in this case, the range-Doppler and ICBA are used to form the ISAR image sequence. The third case study (Section 12.3) concerns dual ground-based/satellite ISAR imaging of a noncooperative sailing ship. In this experiment two radars are used to image a sailing ship at the same time. Two techniques are here used to form the ISAR images: a refocus technique (see Chapter 8) to form the satellite ISAR image; and a range-Doppler with ICBA to form the ground-based ISAR image. In this case study, we show how a dual system is able to form ISAR images of the same target with different target's views and projections. This is a simple case of multiple-perspective ISAR images, which may be looked at as a way to improve a target's classification and recognition [1]. The fourth case study (Section 12.4) shows ISAR images of four aircrafts obtained by processing data acquired with a ground-base radar. This case study may be regarded as the classic ISAR imaging example. Four different airplanes have been selected to show how their characteristics are mapped onto ISAR images. Such characteristics can then be used by classifiers for recognition purposes.

Sub-Nyquist sampling is a common means of interfering with inverse synthetic aperture radar (ISAR) imaging, which has been validated for the feasibility of 1-D compressive sensing (CS) ISAR imaging. In order to overcome the high computational complexity of 1-D CS-based ISAR imaging, the coupled 2-D CS algorithms are utilized to achieve ISAR imaging directly which has been proposed recently. In this paper, sub-Nyquist sampling jamming signals are utilized to induce multiple false-target ISAR images by using the coupled 2-D-CS algorithm. The jamming signals, which are formed by the intercepted radar signals under the sub-Nyquist sampling theorem and scattered by moving targets, are collected by ISAR to establish a 2-D super resolution model, and the finally reconstructed ISAR images are induced and show that the deceptive false-target images retaining the real target information. Simulated trails and real data of a Yak-42 plane are used to validate the correctness of the analyses and the finally well-focused false-target images greatly support the effectiveness of the sub-Nyquist sampling idea in the countermeasures of ISAR with coupled 2-D CS.

In this paper, performance efficiency analysis of compressive inverse synthetic aperture radar (ISAR) imaging for a complex, electrically large target is presented. In this context, physical optics (PO) is used as a high-frequency scattered field prediction technique and compressive sensing (CS) theory is applied to ISAR imaging by exploiting the sparse point scatterer structure of the target. Results prove that CS provides high resolution and accuracy in ISAR imaging even under the Nyquist sampling rate and linearly decreases computational time.

Inverse synthetic aperture radar (ISAR) imaging for the target with micro-motion parts is influenced by the micro-Doppler (m-D) effects. In this case, the radar echo is generally decomposed into the components from the main body and micro-motion parts of target, respectively, to remove the m-D effects and derive a focused ISAR image of the main body. For the sparse aperture data, however, the radar echo is intentionally or occasionally under-sampled, which defocuses the ISAR image by introducing considerable interference, and deteriorates the performance of signal decomposition for the removal of m-D effects. To address this issue, this paper proposes a novel m-D effects removed sparse aperture ISAR (SA-ISAR) imaging algorithm. Note that during a short interval of ISAR imaging, the range profiles of the main body of target from different pulses are similar, resulting in a low-rank matrix of range profile sequence of main body. For the range profiles of the micro-motion parts, they either spread in different range cells or glint in a single range cell, which results in a sparse matrix of range profile sequence. From this perspective, the low-rank and sparse properties are utilized to decompose the range profiles of the main body and micro-motion parts, respectively. Moreover, the sparsity of ISAR image is also utilized as a constraint to eliminate the interference caused by sparse aperture. Hence, SA-ISAR imaging with the removal of m-D effects is modeled as a triply constrained underdetermined optimization problem. The alternating direction method of multipliers (ADMM) and linearized ADMM (L-ADMM) are further utilized to solve the problem with high efficiency. Experimental results based on both simulated and measured data validate the effectiveness of the proposed algorithm.

Combining traditional inverse synthetic aperture radar (ISAR) imaging, interferometry technique and compressed sensing (CS), we studied the problem of interferometric ISAR imaging with sparse aperture. Based on the elaborate analysis of smoothed l0 algorithm in CS, we proposed joint smoothed l0 algorithm which is capable of maintaining coherence between ISAR images of different radars under sparse aperture, thus it could be applied to interferometric ISAR imaging. Besides, we discussed the relationship between sparsity degree and the precision of interferometric ISAR imaging result. Simulation results validate the effectiveness of proposed algorithm and rationality of our conclusions.

For the Inverse Synthetic Aperture Radar (ISAR) imaging, the ISAR image obtained by the Range-Doppler (RD) or time-frequency analysis methods can not display the target's real shape due to its azimuth relating to the target Doppler frequency, thus the cross-range scaling is required for ISAR image. In this paper, a fast cross-range scaling method for ISAR is proposed to estimate the Rotational Angular Velocity (RAV). Firstly, the proposed method utilizes efficient Pseudo Polar Fast Fourier Transform (PPFFT) to transform the rotational motion of two ISAR images from two different instant time into translation in the polar angle direction. Then, a new cost function called integrated correction is defined to obtain the RAV coarse estimation. Finally, the optimal RAV can be estimated using the Bisection method to realize the cross-range scaling. Compared with the available algorithms, the proposed method avoids the problems of precision loss and high computational complexity caused by interpolation operation. The results of computer simulation and real data experiments are provided to demonstrate the validity of the proposed method.

Inverse synthetic aperture radar (ISAR) imaging has the competence of generating two-dimensional (2-D) high-quality images of remote moving targets, while interferometric ISAR (InISAR) is a technique to reproduce three-dimensional (3-D) high-resolution shapes of targets by employing several ISAR images obtained from different antennas. For the target moving with an invariable acceleration, the Doppler frequency shift of its echoes is proved to be time variant; consequently, the Range-Doppler (RD) algorithm will not work properly for acquiring the clear ISAR image of it. To produce high-resolution ISAR/InISAR image of the maneuvering target, a novel algorithm via the peak extraction technique is proposed. The main contributions made by this paper can be listed as follows: 1) by analyzing ISAR imaging with the RD technique for the target moving with a constant acceleration, it is elaborately demonstrated that each scattering point still corresponds to a unique peak in the defocused ISAR image; 2) the expression of the peak location in the defocused ISAR image corresponding to each scattering point is precisely derived; 3) a novel peak extraction technique is developed for forming a high-resolution radar image of the maneuvering target; 4) due to the discovery that the ISAR images obtained in this way can completely preserve the phase information, a new InISAR imaging method via the peak extraction technique is presented consequently. The correctness of the theoretical derivation and the performance of the new algorithms are illustrated through several simulation experiments.

Enhanced sparse ISAR imaging by jointly using sparsity and low-rank properties

Most existing imaging studies are mainly used for a single target or a composite scene consisting of a target and a rough surface (TRS) below it, such as the target-ocean scene. This article focuses on inverse synthetic aperture radar (ISAR) imaging of a special composite scene consisting of a target and layered rough surfaces (TLRS) below it, such as a target and a snow-soil layered scene. Differently from previous imaging algorithms based on the frequency-domain (FD) method, this article combines the time-domain ray bouncing (TDRB) method with imaging algorithms to develop an accurate ISAR imaging framework to obtain composite images of the TLRS scene. To improve the simulation efficiency of the time-domain (TD) echo matrix for ISAR imaging, the multi-thread parallel is used to accelerate the TD scattering algorithm, then, the ISAR imaging formulas based on scattering field expressions in the TLRS scene are derived. Since the linear frequency modulation pulse used in traditional radar imaging has longer pulsewidth than the Gaussian, the matched filter is modified to realize pulse compression under modulating Gaussian pulse excitations to enhance the imaging efficiency. Simulation results show that the proposed ISAR imaging framework ensures the image quality and improves the simulation efficiency significantly. Finally, the ISAR images of the target-snow-soil scene with different parameters are simulated. This article addresses the correlation between the scattering mechanism and image features for ISAR images based on different scattering components, a topic that the literatures do lack.

Multi-view inverse synthetic aperture radar (ISAR) image sequences provide multi-dimensional observation information about space targets, enabling precise attitude estimation that is fundamental to both non-cooperative target monitoring and critical space operations including active debris removal and space collision avoidance. However, directly utilizing all images within an ISAR sequence for attitude estimation can result in a substantial data preprocessing workload and reduced algorithm efficiency. Given the inherent overlap and redundancy in the target information provided by these ISAR images, this paper proposes a novel space target attitude estimation method based on the selection of multi-view ISAR image sequences. The proposed method begins by establishing an ISAR imaging projection model, then characterizing the target information differences through variations in imaging plane normal, and proposing an image selection method based on the uniform sampling across elevation and azimuth angles of the imaging plane normal. On this basis, the method utilizes a high-resolution network (HRNet) to extract the feature points of typical components of the space target. This method enables simultaneous feature point extraction and matching association within ISAR images. The attitude estimation problem is subsequently modeled as an unconstrained optimization problem. Finally, the particle swarm optimization (PSO) algorithm is employed to solve this optimization problem, thereby achieving accurate attitude estimation of the space target. Experimental results demonstrate that the proposed methodology effectively filters image data, significantly reducing the number of images required while maintaining high attitude estimation accuracy. The method provides a more informative sequence than conventional selection strategies, and the tailored HRNet + PSO estimator resists performance degradation in sparse-data conditions, thereby ensuring robust overall performance.

Inverse synthetic aperture radar (ISAR) images are frequently used in target classification and recognition applications. Nevertheless, the interpretation of ISAR images remains problematic for several reasons. One of these is the fact that the image plane cannot be defined by the user but instead depends on the target's own motions and on its relative position with respect to the radar. In order to overcome the problem of interpreting two-dimensional (2D) ISAR images, a method for three-dimensional (3D) reconstruction of moving targets is presented. This method is based on the use of a dual interferometric ISAR system. The interferometric phases measured from two orthogonal baselines are used to jointly estimate the target's effective rotation vector and the heights of the scattering centers with respect to the image plane. The scattering center extraction from the ISAR image is performed by applying a multichannel CLEAN technique. Finally, a 3D image of the moving target is reconstructed from the 3D spatial coordinates of the scattering centers. The effectiveness and robustness of the proposed algorithm is first proven theoretically and then tested against several radar-target scenarios as well as in the presence of noise.

Distributed array radar provides new prospects for three-dimensional (3D) inverse synthetic aperture radar (ISAR) imaging. The accuracy of image registration, as an essential part of 3D ISAR imaging, affects the performance of 3D reconstruction. In this paper, the imaging process of distributed array ISAR is proposed according to the imaging model. The ISAR images of distributed array radar at different APCs have different distribution of scatters. When the local distribution of scatters for the same target are quite different, the performance of the existing ISAR image registration methods may not be optimal. Therefore, an image registration method is proposed by integrating the feature-based method and the area-based method. The proposed method consists of two stages: coarse registration and fine registration. In the first stage, a dominant scatters model is established based on scale-invariant feature transform (SIFT). In the second stage, sub-pixel precision registration is achieved using the local correlation matching method. The effectiveness of the proposed method is verified by comparison with other image registration methods. The 3D reconstruction of the registered experimental data is carried out to assess the practicability of the proposed method.

The ionosphere causes serious phase error in P-band inverse synthetic aperture radar (ISAR) systems, which makes it difficult to obtain a high-quality image. Recently, the convolutional neural network (CNN) has gained much attention in signal processing, and it can automatically extract features to realise an image super-resolution reconstruction. As a popular CNN-based network, U-net can work with less training samples. Hence, the authors are interested in exploiting and modifying the U-net to enhance the P-band ISAR imaging. In this study, in light of the analysis of the effect of the ionospheric total electron content on the ground-based P-band radar echo signal, a novel ISAR imaging method is proposed for the ionospheric effect correction based on the modified U-net and polar formatting algorithm (PFA). The PFA is performed for the phase error coarse compensation. Then, the phase error fine compensation is exploited by the trained U-net. The proposed method can adapt the ionosphere disturbances and show high performance in imaging quality and computational efficiency. The simulation results show the effectiveness of the proposed method.

In this study, we propose an efficient strategy to exploit cross-range resolutions (CRRs) of inverse synthetic aperture radar (ISAR) images to improve classification performance. The proposed method preprocesses an ISAR image of an unknown target using both the CRR of an unknown ISAR image, which can be obtained using the cross-range scaling algorithm, and that of the ISAR images in a training database (DB). Specifically, the proposed method adjusts the unknown ISAR image so that the CRR of the adjusted unknown ISAR image is identical to that of the ISAR images in the training DB. The adjusted unknown ISAR image is then directly used for the subsequent classification task, thus considerably improving classification performance.