Automatic target recognition using enhanced resolution SAR data

Using advanced technology, a new automatic target recognition (ATR) system has been developed that provides significantly improved target recognition performance compared with ATR systems that use conventional synthetic aperture radar (SAR) image-processing techniques. This significant improvement in target recognition performance is achieved by using a new superresolution image-processing technique that enhances SAR image resolution (and image quality) prior to performing target recognition. A computationally efficient two-level implementation of a template-based classifier is used to perform target recognition. The improvement in target recognition performance achieved using superresolution image processing in this new ATR system is quantified     

一种基于改进核主成分分析的SAR图像识别方法研究

针对传统核主成分分析方法识别SAR图像时,存在图像像素之间关联性差、对目标姿态角依赖性强等局限性,研究了一种基于改进核主成分分析的SAR图像识别方法。其研究思想是,结合SAR图像的特点提出了一种基于局部特征核主成分分析的特征提取方法,并设计了一种基于灰关联分析的双分类器对提取特征进行分类。NSTAR仿真实验表明：该方法不仅可以增强图像像素之间的相关性,而且对目标姿态角不存在依赖性,仿真结果验证了方法的有效性和可行性。     

VSAR: a high resolution radar system for ocean imaging

The velocity synthetic aperture radar (VSAR) is a conceptual synthetic aperture radar (SAR)-based sensor system for high resolution ocean imaging. The VSAR utilizes data collected by a multielement SAR system, to extract information not only about the radar reflectivity of the observed area, but also about the radial velocity of the scatterers in each pixel. This is accomplished by making use of the phase information contained in multiple SAR images, and not just the magnitude information as in conventional SAR. Using this velocity information, the VSAR attempts to compensate for the velocity distortion inherent in conventional SAR and to reconstruct the ocean reflectivity. We present the basic theory of the VSAR system and its performance. We also provide an analysis of the VSAR imaging mechanism for a statistical model of the radar returns, designed to capture the effects of speckle and of resolution degradation due to the decorrelation of the radar returns     

一种提高SAR目标识别率的有效方法

在合成孔径雷达自动目标识别SAR ATR中，SAR像的预处理是提高识别率的关键技术之一。给出了一种简单有效的SAR图像预处理方法，该方法首先对SAR目标像进行对数变换后，再做傅立叶变换。经预处理后的SAR像用支持矢量机SVM分类器进行目标识别。实验结果表明：本方法不但有效地提高了目标识别率，而且保证了目标的平移不变性并具有良好的推广能力。     

基于快速提升KLDA准则的MSTARSAR目标特征提取与识别研究

 核线性判别准则(KLDA)是一种非线性特征提取准则。利用KLDA提取MSTAR SAR图像特征，既达到较理想的识别概率，又可克服SAR图像对方位的敏感性。但此时训练样本最多，KLDA的计算代价高。为了解决这一问题，提出一种快速特征向量选择法（FFVS）。FFVS把类别和方位相似的SAR图像分成若干组，然后快速选择各组中部分图像组成一个集合且其到高维特征空间的映射作为一组基。利用该组基的线性组合表示任一样本和投影算子，降低了KLDA中核矩阵的阶数，达到降低计算代价的目的。实验结果表明，FFVS与KLDA组合能达到理想的识别结果。     

An extended chirp scaling algorithm for spaceborne sliding spotlight synthetic aperture radar imaging

A system impulse response with low sidelobes is critical in synthetic aperture radar（SAR） images because sidelobes contribute to noise and interfere with nearby scatterers. However,the conventional tricks of sidelobe suppression are unable to be exactly applied to the case of spaceborne sliding spotlight SAR due to great azimuth shifts in both time and frequency domains. In this paper, an extended chirp scaling algorithm is presented for spaceborne sliding spotlight SAR data imaging. The proposed algorithm firstly uses the spectral analysis（SPECAN） technique to avoid the azimuth spectrum folding effect and then employs the chirp scaling（CS） algorithm to achieve data focusing, i.e., the so-called two-step approach. To suppress the sidelobe level, an efficient strategy for the azimuth spectral weighting which only involves matrix multiplications and short fast Fourier transformations（FFTs） is proposed, which is a post-process executed on the focused SAR image and particularly simple to be implemented. The SAR image processed by the proposed extended CS algorithm is very precise and perfectly phase-preserving. In the end, computer simulation results verify the analysis and confirm the validity of the proposed algorithm.     

A Model for Generating Synthetic VHF SAR Forest Clutter Images

We propose a model for generating low-frequency synthetic aperture radar (SAR) clutter that relates model parameters to physical characteristics of the scene. The model includes both distributed scattering and large-amplitude discrete clutter responses. The model also incorporates the SAR imaging process, which introduces correlation among image pixels. The model may be used to generate synthetic clutter for a range of environmental operating conditions for use in target detection performance evaluation of the radar and automatic target detection/recognition algorithms. We derive a statistical representation of the proposed clutter model's pixel amplitudes and compare with measured data from the CARABAS-II SAR. Simulated clutter images capture the structure and amplitude responses seen in the measured data. A statistical analysis shows an order of magnitude improvement in model fit error compared with standard maximum-likelihood (ML) density fitting methods.     

Superresolution HRR ATR with high definition vector imaging

A new 1-D template-based automatic target recognition (ATR) algorithm is developed and tested on high range resolution (HRR) profiles formed from synthetic aperture radar (SAR) images of targets taken from the Moving and Stationary Target Acquisition and Recognition (MSTAR) data set. In this work, a superresolution technique known as High Definition Vector Imaging (HDVI) is applied to the HRR profiles before the profiles are passed through ATR classification. The new I-D ATR system using HDVI demonstrates significantly improved target recognition compared with previous I-D ATR systems that use conventional image processing techniques. This improvement in target recognition is quantified by improvement in probability of correct classification (PCC). More importantly, the application of HDVI to HRR profiles helps to maintain the same ATR performance with reduced radar resource requirements     

基于边缘相似性的异源图像匹配

异源图像匹配是视觉导航、多源图像融合分析的关键步骤之一。对于成像机理差别较大的异源图像,如SAR图像和可见光图像,采用传统的异源图像匹配算法难以得到满意结果。本文提出一种基于边缘相似性的异源图像匹配方法,首先分别提取待匹配图像的边缘特征点集;然后计算基准图的边缘距离场;最后基于边缘相似性模型,通过实时图边缘图和基准图边缘距离场计算边缘相似度,寻找相似度最大的变换参数即为最终匹配参数。采用SAR与可见光图对方法进行了测试,结果表明,这种方法能够快速可靠地实现异源图像匹配。     

Improved SAR target detection via extended fractal features

The utility of the extended fractal (EF) feature is evaluated for the enhancement of the focus of attention (FOA) stage of a synthetic aperture radar (SAR) automatic target recognition (ATR) system. Unlike more traditional SAR detection features that distinguish target pixels from the background only on the basis of contrast, the EF feature is sensitive to both the contrast and size of objects. Furthermore, the structure for the EF feature computational algorithm lends itself to very fast implementation, and it can be shown that the new feature has a CFAR-like (constant false alarm rate) property. We demonstrate the improved performance using the new feature by testing a number of different detection approaches over two databases of SAR imagery     

A technique for enhanced slant range resolution for SAR systems inknowledge-based environment

Synthetic Aperture Radar (SAR) is an airborne (or spaceborne) radar mapping technique for generating high resolution maps of surface target areas including terrain. High resolution is achieved by coherently combining the returns from a number of radar transmissions. The resolution of the images is determined by the parameters of the emissions, with more data giving greater resolution. A requirement of the Microwave Radar Division's SAR radar is to provide classification of targets. This paper presents a technique for enhancing slant range resolution in SAR images by dithering the carrier centre frequency of the transmitted signal. The procedure controls the radar waveforms so they will optimally perform the classification function, rather than provide an image of best quality. It is shown that a Knowledge-Based engineering approach to determining the waveform of the radar gives considerably improved performance as a classifier of targets (of large radar cross-section), even though the corresponding image is degraded     

保持弱细结构特征的SAR图象模拟退火重构方法

模拟退火 ( SA)算法最先由 White R G.用于合成孔径雷达 ( SAR)图象的降斑处理。该算法在重构均匀区域和强结构区效果有很大提高 ,但也有缺点 ,尤其是过分模糊弱细结构。本文提出了一种改进的方法 ,在 SA算法中融入了边沿检测和增强步骤 ,使弱细结构得以增强并在退火过程中保持。为配合此方法 ,采用平稳下降的指数温度规划取代对数形式。通过仔细调整算法过程 ,可使新方法保留 SAR图象中的很多细小结构 ,而不使其他均匀的和强结构场景性能恶化 ,同时也没有引入其他缺陷。改进的算法更加适于中、低分辨率的 SAR图象降斑处理     

基于压缩协作表示的辐射源识别算法

针对低信噪比（SNR）条件下传统辐射源识别算法性能下降的问题,提出了基于压缩协作表示的识别算法,分别从特征提取和分类器设计两方面进行描述。首先将时域辐射源信号变换到二维时频域,通过图像处理方法提取高维特征列向量。经随机矩阵压缩到一定维度后,输入到提出的压缩协作表示分类器中得到识别结果。进而,对协作表示系数进行非负约束,提出了更符合实际应用场景的算法。仿真结果验证了所提算法的可行性与有效性,且在低信噪比条件下稳健性强、抗噪声干扰性能好、计算量较小、易于工程实现。     

基于深度卷积网络的SAR图像目标检测识别

在SAR图像解译应用领域,目标的自动检测与识别一直是该领域的研究重点和热点,也是该领域的研究难点。针对SAR图像的目标检测与识别方法一般由滤波、分割、特征提取和目标识别等多个相互独立的步骤组成。复杂的流程不仅限制了SAR图像目标检测识别的效率,多步骤处理也使模型的整体优化难以进行,进而制约了目标检测识别的精度。采用近几年在计算机视觉领域表现突出的深度学习方法来处理SAR图像的目标检测识别问题,通过使用CNN、Fast RCNN以及Faster RCNN等模型对MSTAR SAR公开数据集进行目标识别及目标检测实验,验证了卷积神经网络在SAR图像目标识别领域的有效性及高效性,为后续该领域的进一步研究应用奠定了基础。     

Performance improvement for constellation SAR using signal processing techniques

A new concept of spaceborne synthetic aperture radar (SAR) implementation has recently been proposed - the constellation of small spaceborne SAR systems. In this implementation, several formation-flying small satellites cooperate to perform multiple space missions. We investigate the possibility to produce high-resolution wide-area SAR images and fine ground moving-target indicator (GMTI) performance with constellation of small spaceborne SAR systems. In particular, we focus on the problems introduced by this particular SAR system, such as Doppler ambiguities, high sparseness of the satellite array, and array element errors. A space-time adaptive processing (STAP) approach combined with conventional SAR imaging algorithms is proposed which can solve these problems to some extent. The main idea of the approach is to use a STAP-based method to properly overcome the aliasing effect caused by the lower pulse-repetition frequency (PRF) and thereby retrieve the unambiguous azimuth wide (full) spectrum signals from the received echoes. Following this operation, conventional SAR data processing tools can be applied to focus the SAR images fully. The proposed approach can simultaneously achieve both high-resolution SAR mapping of wide ground scenes and GMTI with high efficiency. To obtain array element errors, an array auto-calibration technique is proposed to estimate them based on the angular and Doppler ambiguity analysis of the clutter echo. The optimizing of satellite formations is also analyzed, and a platform velocity/PRF criterion for array configurations is presented. An approach is given to make it possible that almost any given sparse array configuration can satisfy the criterion by slightly adjusting the PRF. Simulated results are presented to verify the effectiveness of the proposed approaches.     

A Generalized Approach to Multiresolution Complex SAR Signal Processing

Multiresolution synthetic aperture radar (SAR) image formation has been proven to be beneficial in a variety of applications such as improved imaging and target detection as well as speckle reduction. SAR signal processing traditionally carried out in the Fourier domain has inherent limitations in the context of image formation at hierarchical scales. We present a generalized approach to the formation of multiresolution SAR images using biorthogonal shift-invariant discrete wavelet transform (SIDWT) in both range and azimuth directions. Particularly in azimuth, the inherent subband decomposition property of wavelet packet transform is introduced to produce multiscale complex matched filtering without involving any approximations. This generalized approach also includes the formulation of multilook processing within the discrete wavelet transform (DWT) paradigm. The efficiency of the algorithm in parallel form of execution to generate hierarchical scale SAR images is shown. Analytical results and sample imagery of diffuse backscatter are presented to validate the method.     

适用于非合作发射源的合成孔径雷达无源成像方法

 合成孔径雷达(SAR)具有突出的高分辨率成像能力,在民用和军用领域体现出重要的应用价值。在复杂的作战环境中,提高SAR的电子对抗性能成为SAR技术发展考虑的重要方面。提出一种新的SAR无源成像方法,该成像方法将单部机载SAR接收机或多部机载SAR接收机在不同位置处接收到的回波信号相关,对相关后的信号采用滤波反投影方法进行成像,可重建场景的辐射率。该成像方法不需要已知发射源的波形和位置信息,适用于非合作发射源,具有良好的电子对抗性,并且适用于任意载机飞行轨迹、多基SAR等情形。仿真验证以双基情形为例,验证了该SAR无源成像方法的有效性,并分析了影响成像质量的各关键因素。     

Super resolution SAR imaging via parametric spectral estimationmethods

Super resolution synthetic aperture radar (SAR) image formation via sophisticated parametric spectral estimation algorithms is considered. Parametric spectral estimation methods are devised based on parametric data models and are used to estimate the model parameters. Since SAR images rather than model parameters are often used in SAR applications, we use the parameter estimates obtained with the parametric methods to simulate data matrices of large dimensions and then use the fast Fourier transform (FFT) methods on them to generate SAR images with super resolution. Experimental examples using the MSTAR and Environmental Research Institute of Michigan (ERIM) data illustrate that robust spectral estimation algorithms can generate SAR images of higher resolution than the conventional FFT methods and enhance the dominant target features     

SAR image formation via semiparametric spectral estimation

A new algorithm, referred to as the SPAR (Semiparametric) algorithm, is presented herein for target feature extraction and complex image formation via synthetic aperture radar (SAR). The algorithm is based on a flexible data model that models each target scatterer as a two-dimensional (2-D) complex sinusoid with arbitrary unknown amplitude and constant phase in cross-range and with constant amplitude and phase in range. By attempting to deal with one corner reflector, such as one dihedral or trihedral, at a time, the algorithm can be used to effectively mitigate the artifacts in the SAR images due to the flexible data model. Another advantage of SPAR is that it can be used to obtain initial conditions needed by other parametric target feature extraction methods to reduce the total amount of computations needed. Both numerical and experimental examples are provided to demonstrate the performance of the proposed algorithm     

Algorithms for Fast Image Registration

The automatic determination of local similarity between two images (image registration) is one of the most fundamental problems of image processing and pattern recognition. A class of registration algorithms that are reasonably efficient and robust for translational displacement has been considered to determine relative shift between reference and search images. Stochastic image models defined on a rectangular region of support are used to determine feature vectors associated with reference and search images. A new measure, namely, coefficient of variation, is defined to take into account effects of contrast and sharpness of the images. Based upon this measure, a computationally efficient two-stage algorithm is obtained by combining the image-model based algorithm with a template matching technique. Simulation results with several synthetic and real images are presented to evaluate the performance of the algorithms.