Fast Multiscale Superpixel Segmentation for SAR Imagery

We address the application of model-supported exploitation techniques to synthetic aperture radar (SAR) imagery. The emphasis is on monitoring SAR imagery using wide area 2D and/or 3D site models along with contextual information. We consider here the following tasks useful in monitoring: (a) site model construction using segmentation and labeling techniques, (b) target detection, (c) target classification and indexing, and (d) SAR image-site model registration. The 2-D wide area site models used here for SAR image exploitation differ from typical site models developed for RADIUS applications, in that they do not model specific facilities, but constitute wide area site models of cultural features such as urban clutter areas, roads, clearings, fields, etc. These models may be derived directly from existing site models, possibly constructed from electro-optical (EO) observations. When such models are not available, a set of segmentation and labeling techniques described here can be used for the construction of 2D site models. The use of models can potentially yield critical information which can disambiguate target signatures in SAR images. We address registration of SAR and EO images to a common site model. Specific derivations are given for the case of registration within the RCDE platform. We suggest a constant false alarm rate (CFAR) detection scheme and a topographic primal sketch (TPS) based classification scheme for monitoring target occurrences in SAR images. The TPS of an observed target is matched against candidate targets TPSs synthesized for the preferred target orientation, inferred from context (e.g. road or parking lot targets). Experimental results on real and synthetic SAR images are provided.

This paper proposes using a backpropagation (BP) neural network for the classification of ship targets in airborne synthetic aperture radar (SAR) imagery. The ship targets consisted of 2 destroyers, 2 cruisers, 2 aircraft carriers, a frigate and a supply ship. A SAR image simulator was employed to generate a training set, a validation set, and a test set for the BP classifier. The features required for classification were extracted from the SAR imagery using three different methods. The first method used a reduced resolution version of the whole SAR image as input to the BP classifier using simple averaging. The other two methods used the SAR image range profile either before or after a local-statistics noise filtering algorithm for speckle reduction. Performance on an extensive test set demonstrated the performance and computational advantages of applying the neural classification approach to targets in airborne SAR imagery. Improvements due to the use of multi-resolution features were also observed.

The relationship between synthetic aperture radar (SAR) imagery of ocean waves and the ocean surface structure is discussed. The effects of ocean surface motion on SAR images is given. A generalization of conventional SAR correlation techniques is developed to provide imagery with the minimum possible degradation and distortion. The modeling of such imagery is discussed and an analytically tractable example given. Ongoing work towards the determination of the relevant components of surface microstructure is described. The future use of SAR phase information to supplement information obtainable from SAR imagery is suggested.

The SEASAT satellite system, using an onboard 23-cm-wavelength synthetic aperture radar (SAR), collected approximately 25- to 40-m-resolution radar images of the ocean in 100-km-wide swaths ranging in length from about 300 to 3000 km. Here we report results from the 18 SEASAT SAR passes during the Joint Air-Sea Interaction (JASIN) experiment conducted off the west coast of Scotland in the summer of 1978. These many SAR images, when coupled with the intensive ship, buoy, and aircraft measurements of the JASIN experiment, provide a unique opportunity to assess the ability of satellite SAR to measure ocean surface phenomena, particularly surface wave fields. In this study we use only optically processed SAR images. Although gravity waves of length approximately 80 to 300 m are often seen in SAR imagery, they are not always seen. We find that SAR resolution and wave height are important criteria for determining wave visibility and suggest that wind velocity is also. Comparisons between SAR and buoy estimates of dominant wavelength and direction (including data from several other experiments) agree to within about ±14% and ± 10°, respectively. We use a focus sharpness algorithm to resolve the 180° directional ambiguity of SAR image directional estimates. Correlation of buoy measurements of significant wave height (H 1/3) with peak signal- to-noise ratio in Fourier transforms of SAR images (r = 0.7) suggests that H 1/3 can be estimated to an accuracy of about ± 1 m. However, a similar, but weaker, correlation exists between signal-to-noise and wave length, i.e., H 1/3 and wavelength are correlated in the JASIN data set. SAR image power spectra are in rough agreement with buoy measurements of omnidirectional ocean wave height spectra Ψ(K) and correspond less closely with ocean wave slope spectra Ψ(K) However, the dominant wavenumber (Kpeak.) in SAR spectra typically falls below the dominant wavenumber of corresponding buoy spectra. Further, the slope of SAR spectra for K K Peak is typically less steep than corresponding buoy spectra Ψ(K). These differences between SAR spectra and buoy measurements of form qualitative support for the imaging mechanisms developed by Alpers, Ross, and Rufenach. Analysis algorithms based on this theory, but including empirical modifications, should substantially improve estimates of Ψ (K,θ) using SAR images.

Synthetic aperture radar (SAR) image segmentation is the key to SAR image automatic interpretation. However, speckle noise, intensity inhomogeneity, and irregular shaped objects with changing edge often make the SAR image segmentation very difficult, and existing algorithms have high computational complexity. We propose a region-based level set method using the local image intensity information. To represent the statistical characteristics of speckle noise, we first use a gamma statistical distribution to model every segmented SAR image. We then apply a modified region mean estimation formula to efficiently segment SAR images with inhomogeneity. Finally, Gaussian filtering is employed to regularize the level set function, which can avoid reinitialization. The experimental results on synthetic and real-world SAR images demonstrate that the proposed method has less computation cost, faster convergence rate, and more accurate segmentation results.

A theoretical model of waves propagating into an ice cover composed of frazil and pancake ice is developed and compared with measurements of wavelength and direction derived from synthetic aperture radar (SAR) imagery obtained from Seasat in October 1978. The theoretical model is based on the concept that ice of these types, which consists of small crystals or cakes, has only a mass‐loading effect on the water surface. We derive the dispersion relation for phase and group velocities, finding that there is an upper frequency limit for propagation into the ice. From the reflection coefficient at the ice edge we derive the wave radiation pressure exerted on the ice, showing that it will cause a slick of frazil ice backed by thicker floes to become more dense or thick with increasing penetration. The implications for radar scattering enabling detection on SAR are that the Bragg resonant wavelength corresponds to waves above the frequency limit for propagation, so that a frazil slick appears dark on an SAR image. When the frazil ice becomes transformed into pancake ice, through slick compression or other means, the raised edges of the pancakes cause the ice to appear bright despite the fact that there are no waves present at the Bragg wavelength. These results are applied to a Seasat SAR image obtained from the Chukchi Sea. The appearance of the ice in the image corresponds to what we expect for frazil ice gradually transforming itself into pancake ice, backed by thicker floes. We derive directional wave number spectra outside and inside the ice cover by digital Fourier analysis of image subscenes, and we find that the change of wavelength and angle of refraction of the dominant wave entering the ice field are both characteristic of the dispersion relation derived theoretically. Mean ice thicknesses extracted from the theory correspond to thicknesses expected for such slicks. The technique offers a possible means of extracting the thickness of fields of frazil and pancake ice from SAR imagery; this may be of considerable utility when ERS 1 SAR is used to study the advancing winter ice edge in the Antarctic, which consists of vast areas of these ice types.

Finding ships in water using high-resolution radar images is goal of synthetic aperture radar (SAR) ship detection, which works in any lighting or weather scenario. By taking advantage of ships' distinct scattering characteristics against water's surface, it is an essential tool for border security, maritime surveillance, and monitoring illicit fishing or trafficking. To accurately recognize ships in synthetic aperture radar (SAR) images in real-time, this study presents AMTNet, a new attention-enhanced multiscale transformer network. Despite abundance of structural information in SAR images, object detection remains a tough task due to speckle noise and diverse backdrops. suggested AMTNet integrates a ResNet-based multiscale convolutional backbone, a spatial attention module (SAM), a channel attention module (CAM), and transformer-based contextual modeling to solve these problems. In addition, a feature exchange system is implemented to address issue of domain gaps that might be generated by differences in environment and sensors in bi-temporal SAR images. To improve accuracy of change detection and localization, this Siamese design allows cooperative exploitation of spatial, temporal, and contextual signals. Compared to top models like MSDFF-Net, GLDet, and LEAD-YOLO, AMTNet outperforms them all on SSDD dataset, with metrics like detection speed (138.5 FPS), accuracy (94.60 percent), recall (95.15 percent), and F1-score (74.87%). Attention processes and feature exchange have been proven to play a key role in ablation research. Another proof of model's practicality is its resilience in face of occlusion and noise. With its excellent performance in a variety of SAR imaging settings and its high interpretability, AMTNet is a computationally economical and effective option for marine surveillance.

The usage of a hypersonic platform for remote sensing application has promising prospects, especially for hypersonic platform-borne synthetic aperture radar (SAR) imaging. However, the high-speed of hypersonic platform will lead to extreme friction between the platform and air, which will cause the ionization of air. The ionized gas forms the plasma sheath wrapped around the hypersonic platform. The plasma sheath will severely affect the propagation of SAR signal and further affect the SAR imaging. Therefore, hypersonic platform-borne SAR imaging should be studied from a physical perspective. In this paper, hypersonic platform-borne SAR imaging under plasma sheath is analyzed. The SAR signal propagation in plasma sheath is computed using scatter matrix method. The proposed SAR signal model is verified by using a ground experiment system. Moreover, the effect of attenuation caused by plasma sheath on SAR imaging is studied under different SAR parameters and plasma sheath. The result shows that attenuation caused by plasma sheath will degrade the SAR imaging quality and even cause the point and area targets to be submerged into the noise. The real SAR images under plasma sheath also illustrate this phenomenon. Furthermore, by studying imaging results under different SAR and plasma parameters, it can be concluded that the severe degradation of SAR imaging quality appears at condition of high plasma sheath electron density and low SAR carrier frequency. The work in this paper will be beneficial for the study of hypersonic platform-borne SAR imaging and design of hypersonic SAR imaging systems in the future.

Recognition of occluded objects in synthetic aperture radar (SAR) images is a significant problem for automatic target recognition. In this paper, we present a hidden Markov modeling (HMM) based approach for recognizing objects in synthetic aperture radar (SAR) images. We identify the peculiar characteristics of SAR sensors and using these characteristics we develop feature based multiple models for a given SAR image of an object. The models exploiting the relative geometry of feature locations or the amplitude of SAR radar return are based on sequentialization of scattering centers extracted from SAR images. In order to improve performance we integrate these models synergistically using their probabalistic estimates for recognition of a particular target at a specific azimuth. Experimental results are presented using both synthetic and real SAR images.

Successful implementation of oil spill segmentation in synthetic aperture radar (SAR) images is vital for marine environmental protection. In this article, we develop an effective segmentation framework named DGNet, which performs oil spill segmentation by incorporating the intrinsic distribution of backscatter values in SAR images. Specifically, our proposed segmentation network is constructed with two deep neural modules running in an interactive manner, where one is the inference module to achieve latent feature variable inference from SAR images and the other is the generative module to produce oil spill segmentation maps by drawing the latent feature variables as inputs. Thus, to yield accurate segmentation, we take into account the intrinsic distribution of backscatter values in SAR images and embed it in our segmentation model. The intrinsic distribution originates from SAR imagery, describing the physical characteristics of oil spills. In the training process, the formulated intrinsic distribution guides efficient learning of optimal latent feature variable inference for oil spill segmentation. The efficient learning enables the training of our proposed DGNet with a small amount of image data. This is economically beneficial to oil spill segmentation where the availability of oil spill SAR image data is limited in practice. Additionally, benefiting from optimal latent feature variable inference, our proposed DGNet performs accurate oil spill segmentation. We evaluate the segmentation performance of our proposed DGNet with different metrics, and experimental evaluations demonstrate its effective segmentations.

The importance of synthetic aperture radar (SAR) images has increased exponentially in recent years due to its ability to capture images in all-weather, day and night conditions. SAR can generate target images in scale and cross-range with enough resolution for identification. The main demerit with SAR images is the multiplicative or speckle noise appearance in captured images, which can affect classification performance. To achieve a good classification result, it is crucial to reduce the speckle noise from the SAR imagery. The size of the available SAR imagery dataset is limited. There is always a chance that a classification model may overfit due to the small size dataset and less feature compared to optical imagery. In this work, the authors anticipated a new convolutional neural network (CNN) for SAR object classification based of despeckling. The experiment is performed on the MSTAR dataset. Experimental results of the current study showed that the proposed CNN model has resulted in an excellent training accuracy of 99.67 % and validation accuracy of 98.98%.

This paper proposes a novel method on Synthetic Aperture Radar (SAR) Image denoising, which is based on context modeling combined with multiscale orthogonal bandlet coefficients. For the main influence of SAR image is multiplicative speckle noise, the logarithm uniform transform is used to convert it to additive noise. In this paper, the multiscale orthogonal bandlet transform is applied to noisy SAR Image. The contextual model is used to compute contextual values of bandlet coefficients on each scale. Therefore, bandlet coefficients are reconstructed into some parts by contextual values. The spatially adaptive soft-threshold (T) is to denoise SAR image in each part. Experimental results demonstrate that this method outperforms the methods of Ehance-Lee (E-lee), contextual model based on wavelet domain (Wavelet-context), and Hidden Markov Tree model bases on wavelet domain (Wavelet-hmt) on there indexes: Floating index (FI), Standard deviation and Edge saving index of vertical direction, by using a real SAR image. However, the method can not only reduce the speckle noise but also preserve edges and high frequency information, and gets considerable visual effect on SAR Image denoising.

This tutorial presents a state-of-the-art survey of advanced applications of Synthetic Aperture Radar (SAR). Applications include: the detection and recognition of targets in SAR imagery, and the detection of targets under trees (and in the open) using change detection (coherent change detection, full-polarization change detection, etc). A fully polarimetric SAR sensor is described. Models of clutter and targets are developed from analysis of full-polarization SAR imagery. Algorithms for optimum (GLRT) processing of full-polarization data are derived, including the polarimetric whitening filter (PWF) and the polarimetric matched filter (PMF). Polarization-based discrimination features (polarization entropy, anisotropy, etc.) are developed and applied to SAR image terrain classification. Model-based and template-based target recognition systems are described and classifier performance results versus resolution and polarization are summarized for 10-and 20-target classifiers. Super-resolution algorithms are applied to SAR imagery and target recognition performance improvements demonstrated. Multi-polarization and multi-pass change detection algorithms are developed and demonstrated using UHF SAR imagery of targets under trees; fusion of multi-polarization coherent and non-coherent change detection algorithms is demonstrated. The effect of SAR image compression on change detection performance is quantified. Change detection performance using multi-pass VHF SAR imagery of targets under trees is demonstrated using data gathered by the Swedish CARABAS SAR system.

Monitoring the marginal ice zone (MIZ) is becoming increasingly important due to recent evidence that the width of the MIZ is changing with climate. A method to automatically detect the MIZ in synthetic aperture radar (SAR) imagery is proposed. The method utilizes the curve-like features of MIZ in SAR images. A multiscale strategy, the curvelet transform, is chosen to extract features from the SAR images. The statistical and co-occurrence features of curvelet coefficients at an appropriate scale are used to identify the MIZ from open water and consolidated ice. Experimental results show a significant increase in classification accuracy (89.7%) compared with the most commonly used MIZ definition from passive microwave sea ice concentration (74%), especially in the diffuse MIZ.

The availability of synthetic aperture radar (SAR) data offers great potential for environmental monitoring due to the insensitiveness of SAR imagery to atmospheric and sun- light-illumination conditions. In addition, the short revisit time provided by future SAR-based missions will allow a huge amount of multitemporal SAR data to become systematically available for monitoring applications. In this paper, the problem of detecting the changes that occurred on the ground by analyzing SAR im- agery is addressed by a completely unsupervised approach, i.e., by developing an automatic thresholding technique. The image- ratioing approach to SAR change detection is adopted, and the Kittler and Illingworth minimum-error thresholding algorithm is generalized to take into account the non-Gaussian distribution of the amplitude values of SAR images. In particular, a SAR-specific parametric modeling approach for the ratio image is proposed and integrated into the thresholding process. Experimental results, which confirm the accuracy of the method for real X-band SAR and spaceborne imaging radar C-band images, are presented.