It is a feasible and promising way to utilize deep neural networks learn extract valuable features from synthetic aperture radar (SAR) images for SAR automatic target recognition (ATR). However, it too difficult effectively train the with limited raw images. In this paper, we propose new approach do ATR, in which multiview learning framework was employed. Based on ATR pattern, first present flexible mean generate adequate data, can guarantee large amount of inputs network training without...

Ship target detection using large-scale synthetic aperture radar (SAR) images has important application in military and civilian fields. However, ship targets are difficult to distinguish from the surrounding background many false alarms can occur due influence of land area. False always with because most area SAR treated as clutter, considered unevenly distributing small targets. To address these issues, a method via CenterNet is proposed this article. As an anchor-free method, defines...

This paper considers the multi-target tracking (MTT) problem through use of dynamic programming based track-before-detect (DP-TBD) methods. The usual solution this is to adopt a state, which concatenation individual target states, then search estimate in expanded state space. However, involves high-dimensional joint maximization computationally intractable for most realistic problems. Additionally, dimension has be determined before implementing DP search. problematic when number targets...

This study proposes a novel non‐negative matrix factorisation (NMF) variant L 1/2 ‐NMF after visualisation and analysis of the process target recognition via NMF for synthetic aperture radar (SAR) images. has been applied to obtain pattern feature in SAR considers intrinsic character physical meaning when automatic recognition. At base obtaining linear relationship between sample be recognised train samples, whole is detailed vividly visualised. Meanwhile, lots researches have done improve...

Recently, the generalized sparse iterative covariance-based estimation algorithm was extended to allow for varying norm constraints in scanning radar applications. In this paper, further development, we introduce a wideband dictionary framework which can provide computationally efficient of signals. The technique is formed by initially introducing coarse grid constructed from integrating elements, spanning bands considered parameter space. After forming estimates activated bands, these are...

With appropriate geometry configurations, bistatic synthetic aperture radar (SAR) can break through the limitations of monostatic SAR on forward-looking imaging. Thanks to such a capability, (BFSAR) has extensive potential applications, as self-navigation and self-landing. In mode BFSAR with stationary transmitter (ST-BFSAR), two-dimensional spatial variation makes it difficult use traditional data focusing algorithms. this letter, an imaging algorithm based keystone transform nonlinear...

With the geosynchronous synthetic aperture radar (SAR) satellite as transmitter, unmanned aerial vehicle (UAV) can passively receive echo within illuminated ground area and achieve 2-D imaging of interested target. This SAR system, known GEO-UAV bistatic SAR, is capable autonomously accomplishing mission in rough terrain environments by prespecifying a path for UAV receiver. In this paper, system first investigated. The practical advantages spatial resolution are then analyzed detail....

High-resolution scanning radar mapping of the surface is an effective tool for addressing concerns in local environmental and social investigation fields. Regrettably, azimuth resolution a constrained by antenna beamwidth. Multiple super-resolution approaches have been applied to enhance resolution, but they suffer from limited improvement. In this paper, methodology derive estimates at improved proposed. We first consider truncated spectrum discarding unreliable frequencies suppress noise...

In synthetic aperture radar (SAR) imagery, images of ground moving targets (GMTs) are smeared, distorted, and shifted. Current GMT imaging methods mostly based on range-Doppler algorithms, which have two main drawbacks: 1) coupling between range cell migration correction (RCMC) Doppler parameter estimation 2) cross terms degrade the performance nonlinear methods. this paper, an optimal 2-D spectrum matching method for SAR is proposed. The innovation or advantage that problem transformed into...

Recently, a variety of super-resolution (SR) methods have been devoted to enhancing the angular resolution real beam mapping (RBM) imagery in modern microwave remote sensing applications. When addressing large-scale datasets, however, they suffer from notably high computational complexity due high-dimensional matrix inversion, multiplication, or singular value decomposition (SVD). To overcome this limitation, article presents low-complexity SR strategy based on adaptive low-rank...

Due to the division of transceiver, a bistatic synthetic aperture radar (SAR) gains many advantages, such as forward-looking imaging and powerful anti-interference capabilities. In meantime, information sharing (e.g., positions status) between transmitter receiver is required for SAR imaging. This article addresses co- use waveform design SAR-dual-functional communication (SAR-DFRC). To this end, we embed into time-frequency spectrum phase coded signal, which can be feasible solution...

Maritime surveillance is indispensable for civilian fields, including national maritime safeguarding, channel monitoring, and so on, in which synthetic aperture radar (SAR) ship target recognition a crucial research field. The core problem to realizing accurate SAR the large inner-class variance inter-class overlap of features, limits performance. Most existing methods plainly extract multi-scale features network utilize equally each feature scale classification stage. However, shallow are...

With appropriate geometry configurations, bistatic synthetic aperture radar (SAR) can break through the limitations of monostatic SAR on forward-looking imaging. Thanks to such a capability, (BFSAR) has extensive potential applications. In this paper, we categorize various airborne and ground based BFSAR give their possible applications, challenges processing approaches. After that, gradient theory, analyze resolution performance different mode present intrinsic reasons why carry out

Geosynchronous synthetic aperture radar (GEO-SAR) offers new opportunities for continuous Earth observation missions with large coverage and short revisit cycle. The unique features of GEO-SAR present huge potentials bistatic applications. In this paper, the concept advantages GEO SAR (GEO-BiSAR) are first investigated. system consists a illuminator an airborne receiver, such as airplane or near-space vehicle. Compared monostatic system, configuration can provide finer spatial resolution...

Because a scanning radar system works as noncoherent sensor, it is suitable for any geometry situation, and has significant extensive applications, such surveillance, autonomous landing of aircraft, navigation, guidance. After the pulse compression technique improving range resolution was presented, angular became crucial system. In this paper, scheme superresolution based on maximum posteriori (MAP) framework proposed. First, received signal in azimuth modeled mathematical convolution...

Particle filter (PF) based multi-target tracking (MTT) methods suffer from the curse of dimensionality. Existing strategies to combat this assume posterior independence between target states, in order then sample targets independently, or perform joint sampling closely spaced only. When many are proximity, these either poorly too computationally expensive. We make two contributions towards addressing limitations. Firstly, we advocate an alternative view use which emphasizes statistical...

Scanning radar is of notable importance for ground surveillance, terrain mapping and disaster rescue. However, the angular resolution a scanning image poor compared to achievable range resolution. This paper presents deconvolution algorithm super-resolution in based on Bayesian theory, which states that can be realized by solving corresponding problem with maximum posteriori (MAP) criterion. The considers noise composed two mutually independent parts, i.e., Gaussian signal-independent...

Airborne forward-looking radar (AFLR) imaging has raised many concerns in fields of Earth observation, independent weather and daytime. Constrained by principles, conventional high-resolution techniques such as synthetic aperture (SAR) Doppler beam sharpening (DBS) are incapable AFLR imaging. The real (RAR) can obtain images using a scanning antenna, but suffers from coarse cross-range resolution. Recently, there been much attention paid to the iterative adaptive approach (IAA), which draws...

Bistatic forward-looking synthetic aperture radar (BFL-SAR) is a kind of bistatic SAR system that can image terrain in the flight direction an aircraft. Until now, BFL-SAR imaging theories and methods have been researched for stationary targets. Unlike target, motion ground-moving target (GMT) induces unknown range cell migration additional modulation azimuth signal. Thus, to finely GMT, one must obtain its velocity parameters accurately, but they are usually unknown. In this paper, novel...

Cross-resolution enhancement for airborne high-squint radar (AHSR) imagery is mathematically equivalent to the ill-conditioned problem of inverse-scattering reconstruction. Although a variety inversion methods with regularization can be introduced advance field AHSR imagery, they turn out computationally intensive when extended 2-D (range and cross-range dimension) image formulation due range-by-range calculation space-variant operators over full range swath. To tackle efficiency, this...

Real-aperture radar can be utilized to realize forward-looking imaging by antenna scanning the region. However, low azimuth resolution seriously affects its practical application. Although traditional super-resolution methods could enhance a certain extent, effective preservation of contour information for important targets still remains problem. In this article, method total variation-sparse (TV-sparse) multiconstraint deconvolution is proposed improve as well preserve targets. Since our...

Bistatic forward-looking synthetic aperture radar (BFSAR) is a kind of bistatic SAR system that can image terrain in the flight direction receiver. Current literature and reports about BFSAR mainly concentrate on stationary scene ground-moving target imaging. Unlike targets, translational rotational movements ship targets usually lead to complicated range cell migration (RCM) Doppler frequency (DFM). Moreover, characteristics RCM DFM for different scattering points are significantly...

Clutter suppression is a challenging task in synthetic aperture radar-ground moving target indication (SAR-GMTI). In general, sufficient secondary samples are not easily acquired due to the nonstationary and nonhomogeneous characteristics of bistatic SAR (BiSAR) clutter, resulting worse clutter results. Recently, space–time adaptive processing based on sparse recovery (SR-STAP) has been developed since its better performance with less samples. However, off-grid problem domain caused by...

Convolutional neural networks (CNNs) have dominated the synthetic aperture radar (SAR) automatic target recognition (ATR) for years. However, under limited SAR images, width and depth of CNN-based models are limited, widening received field global features in images is hindered, which finally leads to low performance recognition. To address these challenges, we propose a Transformer (ConvT) ATR few-shot learning (FSL). The proposed method focuses on constructing hierarchical feature...