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...

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...

Ship detection from synthetic aperture radar (SAR) images is one of the crucial issues in maritime surveillance. However, due to varying ocean waves and strong echo sea surface, it very difficult detect ships heterogeneous clutter backgrounds. In this paper, an innovative ship method proposed effectively distinguish vessels complex backgrounds a SAR image. First, input image pre-screened by maximally-stable extremal region (MSER) method, which can obtain candidate regions with low...

Marine surveillance radar is widely used in marine monitoring for its ability of observing sea surface all-time and all-weather. However, the target detection performance seriously affected by existence clutter. In this article, we propose a new clutter suppression method based on machine learning approach. We first employ cyclic structure network with pair generative adversarial networks to sufficiently learn characteristics clutter, which converts problem as transformation from data domain...

Angular resolution of real aperture radar (RAR) can be improved using deconvolution methods to achieve enhanced target information based on the convolution relationship between scatterings and an antenna pattern. However, depending wide scanning scope dense sampling angular interval, computational complexity will drastically increase as dimension azimuthal data increases. In this paper, efficiently improve RAR, a generalized adaptive asymptotic minimum variance (GAAMV) estimator that relies...

Synthetic aperture radar (SAR) is an important microwave sensor that capable of high-resolution imaging. Extracting valuable features from the SAR target imagery one crucial issues in automatic recognition (ATR). In this paper, we propose a new feature extraction method named 2-D principal-component-analysis-based neighborhood virtual points discriminant embedding (2DPCA-based 2DNVPDE) for ATR. The projected into space by 2DPCA and 2DNVPDE approach. able to preserve global spatial structure...

Synthetic aperture radar (SAR) is an important means for target surveillance through reconstructing the microwave image of observation area. However, under condition low signal-to-clutter ratio (SCR), such as a strong sea clutter situation, it difficult to surveil targets from SAR images acquired by traditional matched filter-based imaging methods. To improve performance SAR, this article proposes target-oriented method, which can enhance desired and SCR in reconstructed images. By...

Multiview synthetic aperture radar (SAR) images contain much richer information for automatic target recognition (ATR) than a single-view one. It is desirable to establish reasonable multiview ATR scheme and design effective algorithm thoroughly learn extract that classification information, so superior SAR performance can be achieved. Hence, general processing framework applicable pattern first given in this paper, which provide an approach system design. Then, new method using deep feature...

In radar forward-looking super-resolution imaging, improving the azimuth resolution while acquiring contour information of target has significant research value. this letter, an approach based on balanced Tikhonov and total variation (TV) deconvolution is proposed for imaging. We combine regularization TV to construct objective function resolve respective cost using alternating direction method multipliers (ADMM). each iteration, gradient scattering coefficient used as adaptive weighted...

Spaceborne-airborne multistatic synthetic aperture radar (SA-MuSAR) has the ability to provide high-resolution forward-looking imagery for receivers, but it relies on careful design of geometric configuration (GC). In this article, a GC optimization method is proposed obtain high-quality fused image with limited observation time. First, relationship between spatial resolution and illustrated by wavenumber spectrum distribution SA-MuSAR. Second, evaluators depending multiple data are...

Radar operation mode recognition holds an increasingly critical place in electronic countermeasure (ECM) as well remote sensing. However, the overlapped waveform parameters pose huge challenges to performing radar task severe electromagnetic environments, particularly with large measurement errors or small sample lengths. By analyzing timing patterns of a single pulse parameter and correlation characteristics multiple parameters, this paper first provides revolutionary representation...

Synthetic aperture radar (SAR) can provide high-resolution electromagnetic backscattering images of the illuminated area, playing a significant role in various applications. However, achieving focused SAR is challenging under sparse sampling and phase error conditions. By exploiting sparsity or compressibility priors, state-of-the-art sparsity-driven imaging methods reconstruct condition sampling. handcrafted priors used these limit performance, iterative solution schemes reduce...

Automatic target recognition (ATR) holds a crucial position in synthetic aperture radar (SAR) image interpretation. Despite deep learning advancements have significantly propelled SAR ATR, addressing the challenge of with few training data remains vital concern applications. Two main issues still exist: 1) In few-shot depth and width CNN-based models are limited, which restricts its modeling capacity, thus extracting discriminative generalized features challenging. 2) With only labeled...

A novel reduced-dimension (RD) MUSIC algorithm based on augmented correlation matrix is proposed to estimate the direction of arrival (DOA) and departure (DOD) for bistatic co-prime multiple-input multiple-output (MIMO) radar. Firstly, by reference difference co-array receiving array, we construct an which then utilized DOA estimation one-dimension (1D) MUSIC. Secondly, depending each estimated DOA, designed spatial filtering performed original received signal after matched filtering, a...

Synthetic aperture radar (SAR) is an advanced microwave imaging system of great importance. The recognition real-world targets from SAR images, i.e., automatic target (ATR), attractive but challenging issue. majority existing ATR methods are designed for single-view images. However, multiview images contain more abundant classification information than which benefits and recognition. This paper proposes end-to-end deep feature extraction fusion network (FEF-Net) that can effectively exploit...

Radar countermeasure (RCM) is increasingly important in modern warfare. Fast and accurate jamming decision made by RCM systems can provide timely electronic protection for or high-value targets. To quickly find a policy against the multifunctional radar, generation scheme via heuristic programming reinforcement learning proposed this article, where strategy be dynamically adjusted using interactive self-learning. First, relationships between radar operation modes are investigated under...

Superresolution methods can be applied to real aperture radar (RAR) improve its angular resolution by solving an inverse problem. However, traditional superresolution are achieved after batch data collection, which requires extensive operational complexity and storage space. To solve this problem for RAR, online detect-before-reconstruct (DBR) framework is proposed in article based on the sparse property of targets. First, along range direction, each sample echo detected reduce computational...

Real aperture radar (RAR) usually sweeps a wide sector to continuously observe the scenario of interest. Because its angular resolution is limited by size antenna aperture, target reconstruction methods are widely applied obtain super-resolution images. However, wide-sector processing mode suffers from high operational complexity because high-dimensional matrix inversion. Even worse, for targets located at scene edge, echo data received less than half beamwidth. The incomplete will lead...

Convolutional neural networks (CNN) show superior potential in synthetic aperture radar automatic target recognition (SAR ATR). However, due to the difficulty of obtaining SAR images and scarcity labeled images, supervised learning has poor performance this area is not widely applicable. To address problem, a semi-supervised conditional generative adversarial network with multi-discriminator (SCGAN-MD) proposed paper. In our method, (CGAN) adopted two discriminators for training generated...

Waveform design has become an attractive topic in the field of colocated multiple-input multiple-output (MIMO) radar that allows antennas to transmit different waveforms. properties MIMO space, time and Doppler domains determine performances resource utilization, interference suppression, moving target detection. Therefore, simultaneous optimization multi-domain through waveform is significant improve performance radar. In this paper, a novel framework constrains beampattern while maximizing...

Finding out interested targets from synthetic aperture radar (SAR) imagery is an attractive but challenging problem in SAR application. Traditional target detection independent on imaging process, which purposeless and unnecessary. Hence, a new processing approach for simultaneous image formation proposed this paper. This based time domain human visual saliency detection. First, series of sub-aperture images with resolutions low to high are generated by the method. Then, those...