Airborne Synthetic Aperture Radar Images Research Articles

Efficient path estimation through parallel media for wide‐beam ice‐sounding radar

The authors propose an algorithm to estimate the path followed by refracted signals from a source to a target, through a medium formed by uniform parallel layers with known different refractive indices, a common model used for ice radio-echo sounding. The analytical solution is a polynomial with a degree that exponentially depends on the number of layers, being computationally inefficient. For low incidence angles, the small-angle approximation can be used to avoid the polynomial. In their technique, they normalise the governing equations to obtain a framework where to find a narrow angular interval containing the solution, finally estimated interpolating the boundaries. The new approach improves the results regarding the small-angle approximation for a wider angular range at a slightly higher computational time. This method has been applied to focus airborne synthetic aperture radar images for deep ice sounding, reducing the calculation time and improving the detected response in wide beam and squinted geometries, used for high along-track resolution or the detection of sloping internal layers.

A Novel Attention Fully Convolutional Network Method for Synthetic Aperture Radar Image Segmentation

As an important step of synthetic aperture radar image interpretation, synthetic aperture radar image segmentation aims at segmenting an image into different regions in terms of homogeneity. Because of the deficiency of the labeled samples and the existence of speckling noise, synthetic aperture radar image segmentation is a challenging task. We present a new method for synthetic aperture radar image segmentation in this article. Due to the large size of the original synthetic aperture radar image, we first divide the input image into small slices. Then the image slices are input to the attention-based fully convolutional network for obtaining the segmentation results. Finally, the fully connected conditional random field is adopted for improving the segmentation performance of the network. The innovations of our method are as follows: 1) The attention-based fully convolutional network is embedded with the multiscale attention network which is capable of enhancing the extraction of the image features through three strategies, namely, multiscale feature extraction, channel attention extraction, and spatial attention extraction. 2) We design a new loss function for the attention fully convolutional network by combining Lovasz-Softmax and cross-entropy losses. The new loss allows us to simultaneously optimize the intersection over union and the pixel classification accuracy of the segmentation results. The experiments are performed on two airborne synthetic aperture radar image databases. It has been proved that our method is superior to other state-of- the-art image segmentation approaches.

Attention Graph Convolution Network for Image Segmentation in Big SAR Imagery Data

The recent emergence of high-resolution Synthetic Aperture Radar (SAR) images leads to massive amounts of data. In order to segment these big remotely sensed data in an acceptable time frame, more and more segmentation algorithms based on deep learning attempt to take superpixels as processing units. However, the over-segmented images become non-Euclidean structure data that traditional deep Convolutional Neural Networks (CNN) cannot directly process. Here, we propose a novel Attention Graph Convolution Network (AGCN) to perform superpixel-wise segmentation in big SAR imagery data. AGCN consists of an attention mechanism layer and Graph Convolution Networks (GCN). GCN can operate on graph-structure data by generalizing convolutions to the graph domain and have been successfully applied in tasks such as node classification. The attention mechanism layer is introduced to guide the graph convolution layers to focus on the most relevant nodes in order to make decisions by specifying different coefficients to different nodes in a neighbourhood. The attention layer is located before the convolution layers, and noisy information from the neighbouring nodes has less negative influence on the attention coefficients. Quantified experiments on two airborne SAR image datasets prove that the proposed method outperforms the other state-of-the-art segmentation approaches. Its computation time is also far less than the current mainstream pixel-level semantic segmentation networks.

New method for airborne SAR image positioning

Synthetic aperture radar (SAR) geometric positioning obtains the geodetic coordinates of each pixel in a SAR image. However, due to the changes in geometry, position and attitude angle of the sensor and the influence factors such as topography, atmospheric refraction, plus the curvature and rotation of the earth, image deformation will occur. Therefore, there is a great practical significance to explore the precise geometric positioning method for SAR images. It proposed a new positioning method combining range-Doppler model and polynomial model, realising SAR image accurate positioning with the digital elevation model data. The method considers the effects of terrain-height-induced geometric distortions and establishes a practical positioning model based on the geometric relationship of imaging. The result shows that the positioning accuracy of the proposed method is higher than the accuracy of the traditional rang-Doppler model and polynomial model with a small number of ground control points.

An Efficient Segmentation Method for Mariculture Areas on Airborne SAR Images

AbstractIn this paper, the problems of mariculture area segmentation and corresponding area value estimations are investigated on the basis of airborne synthetic aperture radar (SAR) images. In order to deal with a limited amount of noisy airborne SAR image data in an efficient way, an effective coarse-to-fine approach is proposed, consisting of three major components, including (1) an adaptive segmentation method for each local patch to remove noise from the ocean background, (2) a dynamic coarse-to-fine clustering method for grouping pixels to achieve image segments, and (3) a polygon-fitting-based algorithm to obtain regular borders for each region and corresponding area value. Some feasible experiments are operated based on the restricted airborne SAR images, and the effectiveness of the proposed algorithm is validated in terms of the provided pixel level evaluation annotations.

Refocusing Moving Ship Targets in SAR Images Based on Fast Minimum Entropy Phase Compensation.

Moving ship targets appear blurred and defocused in synthetic aperture radar (SAR) images due to the translation motion during the coherent processing. Motion compensation is required for refocusing moving ship targets in SAR scenes. A novel refocusing method for moving ship is developed in this paper. The method is exploiting inverse synthetic aperture radar (ISAR) technique to refocus the ship target in SAR image. Generally, most cases of refocusing are for raw echo data, not for SAR image. Taking into account the advantages of processing in SAR image, the processing data are SAR image rather than raw echo data in this paper. The ISAR processing is based on fast minimum entropy phase compensation method, an iterative approach to obtain the phase error. The proposed method has been tested using Spaceborne TerraSAR-X, Gaofeng-3 images and airborne SAR images of maritime targets.

Airborne SAR Imaging Algorithm for Ocean Waves Based on Optimum Focus Setting

Ocean waves are the richest texture on the sea surface, from which valuable information can be inversed. In general, the Synthetic Aperture Radar (SAR) images of surface waves will inevitably be distorted due to the intricate motion of surface waves. However, commonly used imaging algorithms do not take the motion of surface waves into consideration. Therefore, surface waves on the obtained SAR images are rather blurred. To solve this problem, an airborne SAR imaging algorithm for ocean waves based on optimum focus setting is proposed in this paper. Firstly, in order to obtain the real azimuth phase speed of dominant wave, the geometric and scanning distortion in the blurred SAR image is calibrated. Subsequently, according to the SAR integration time and wavelength of the dominant wave, a proper focus setting variation section is selected. Afterwards, all the focus settings in this variation section are used to refocus the image, which are then compared to decide the optimum focus setting for dominant wave. Finally, by redesigning the azimuth matched filter using this optimum focus setting, a well-focused SAR image for the dominant wave can be obtained. The proposed algorithm is applied to both simulation and field data, and SAR images of surface waves are obtained. Furthermore, the obtained images are compared with those obtained with a zero-focus setting. The comparison shows that the focus of surface waves is significantly improved, which verifies the effectiveness of the proposed algorithm. Finally, how to choose the appropriate focus setting variation section under different parameters and the applicability of the algorithm are analyzed.

Weakly Supervised Segmentation of SAR Imagery Using Superpixel and Hierarchically Adversarial CRF

Synthetic aperture radar (SAR) image segmentation aims at generating homogeneous regions from a pixel-based image and is the basis of image interpretation. However, most of the existing segmentation methods usually neglect the appearance and spatial consistency during feature extraction and also require a large number of training data. In addition, pixel-based processing cannot meet the real time requirement. We hereby present a weakly supervised algorithm to perform the task of segmentation for high-resolution SAR images. For effective segmentation, the input image is first over-segmented into a set of primitive superpixels. This algorithm combines hierarchical conditional generative adversarial nets (CGAN) and conditional random fields (CRF). The CGAN-based networks can leverage abundant unlabeled data learning parameters, reducing their reliance on the labeled samples. In order to preserve neighborhood consistency in the feature extraction stage, the hierarchical CGAN is composed of two sub-networks, which are employed to extract the information of the central superpixels and the corresponding background superpixels, respectively. Afterwards, CRF is utilized to perform label optimization using the concatenated features. Quantified experiments on an airborne SAR image dataset prove that the proposed method can effectively learn feature representations and achieve competitive accuracy to the state-of-the-art segmentation approaches. More specifically, our algorithm has a higher Cohen’s kappa coefficient and overall accuracy. Its computation time is less than the current mainstream pixel-level semantic segmentation networks.

Long-Term Satellite Monitoring of the Slumgullion Landslide Using Space-Borne Synthetic Aperture Radar Sub-Pixel Offset Tracking

Kinematic characterization of a landslide at large, small, and detailed scale is today still rare and challenging, especially for long periods, due to the difficulty in implementing demanding ground surveys with adequate spatiotemporal coverage. In this work, the suitability of space-borne synthetic aperture radar sub-pixel offset tracking for the long-term monitoring of the Slumgullion landslide in Colorado (US) is investigated. This landslide is classified as a debris slide and has so far been monitored through ground surveys and, more recently, airborne remote sensing, while satellite images are scarcely exploited. The peculiarity of this landslide is that it is subject to displacements of several meters per year. Therefore, it cannot be monitored with traditional synthetic aperture radar differential interferometry, as this technique has limitations related to the loss of interferometric coherence and to the maximum observable displacement gradient/rate. In order to overcome these limitations, space-borne synthetic aperture radar sub-pixel offset tracking is applied to pairs of images acquired with a time span of one year between August 2011 and August 2013. The obtained results are compared with those available in the literature, both at landslide scale, retrieved through field surveys, and at point scale, using airborne synthetic aperture radar imaging and GPS. The comparison showed full congruence with the past literature. A consistency check covering the full observation period is also implemented to confirm the reliability of the technique, which results in a cheap and effective methodology for the long-term monitoring of large landslide-induced movements.

SAR Target CFAR Detection Via GPU Parallel Operation

The constant false alarm rate with convolution and pooling (CP-CFAR) method, which can improve the detection efficiency via GPU parallel acceleration in the airborne synthetic aperture radar (SAR) images, is proposed in this paper. The constant false alarm rate (CFAR) method is one of the most widely used methods for target detection in airborne SAR images. However, since the CFAR is performed on a pixel-by-pixel basis, the time consumption will increase rapidly with the expansion of image scene. Even if the GPU is used for acceleration, the efficiency improvement is still limited, which cannot meet the real-time processing requirements. Therefore, the CP-CFAR method for the target detection of SAR images is proposed in this paper. The convolution layer uses the horizontal and vertical Sobel operators to improve the contrast between targets and background, and the pooling layer can reduce the processing dimension of the images. The convolution and pooling layers are added before the two-parameter CFAR, which can reduce the computational elements but without losing the main feature of the original image. More importantly, compared to the traditional CFAR, the proposed CP-CFAR is more suitable for GPU acceleration, which can improve the detection efficiency significantly. Experiments on the moving and stationary target acquisition and recognition SAR images with a size of 1478 × 1784 show that, compared with the traditional cell-averaging CFAR, two-parameter CFAR and their CPU, multithread CPU, and GPU acceleration modes, the proposed CP-CFAR with GPU acceleration can obtain the best detection performance with the highest acceleration ratio, and the operation time is less than 192 ms.

Despeckling of SAR Images by Finding the Expected Values Using the Probability Distribution of Speckle

Synthetic aperture radar (SAR) images are contaminated by multiplicative speckle noise, which reduces the contrast and resolution of the images. To improve the quality and the performance of quantitative image analysis, speckle reduction is a prerequisite for SAR images. In this paper, a new method is proposed by calculating the expected value of all the pixel elements in the window with respect to the centre pixel. The weighted mean of all the expected values in the window replaces the centre pixel. The weights are calculated according to the height of the probability distribution. Thus, the expected value which has higher probability has more weightage. The proposed method is applied to the air-borne and space-borne SAR images. By comparing with some well-known filters, the obtained results demonstrate that the proposed method is able to reduce the noise effectively with accurately preserving edges and fine details of the images.

HIGH RESOLUTION SAR IMAGING EMPLOYING GEOMETRIC FEATURES FOR EXTRACTING SEISMIC DAMAGE OF BUILDINGS

Abstract. Synthetic Aperture Radar (SAR) image is relatively easy to acquire but difficult for interpretation. This paper probes how to identify seismic damage of building using geometric features of SAR. The SAR imaging geometric features of buildings, such as the high intensity layover, bright line induced by double bounce backscattering and dark shadow is analysed, and show obvious differences texture features of homogeneity, similarity and entropy in combinatorial imaging geometric regions between the un-collapsed and collapsed buildings in airborne SAR images acquired in Yushu city damaged by 2010 Ms7.1 Yushu, Qinghai, China earthquake, which implicates a potential capability to discriminate collapsed and un-collapsed buildings from SAR image. Study also shows that the proportion of highlight (layover & bright line) area (HA) is related to the seismic damage degree, thus a SAR image damage index (SARDI), which related to the ratio of HA to the building occupation are of building in a street block (SA), is proposed. While HA is identified through feature extraction with high-pass and low-pass filtering of SAR image in frequency domain. A partial region with 58 natural street blocks in the Yushu City are selected as study area. Then according to the above method, HA is extracted, SARDI is then calculated and further classified into 3 classes. The results show effective through validation check with seismic damage classes interpreted artificially from post-earthquake airborne high resolution optical image, which shows total classification accuracy 89.3 %, Kappa coefficient 0.79 and identical to the practical seismic damage distribution. The results are also compared and discussed with the building damage identified from SAR image available by other authors.

TIE POINTS EXTRACTION FOR SAR IMAGES BASED ON DIFFERENTIAL CONSTRAINTS

Abstract. Automatically extracting tie points (TPs) on large-size synthetic aperture radar (SAR) images is still challenging because the efficiency and correct ratio of the image matching need to be improved. This paper proposes an automatic TPs extraction method based on differential constraints for large-size SAR images obtained from approximately parallel tracks, between which the relative geometric distortions are small in azimuth direction and large in range direction. Image pyramids are built firstly, and then corresponding layers of pyramids are matched from the top to the bottom. In the process, the similarity is measured by the normalized cross correlation (NCC) algorithm, which is calculated from a rectangular window with the long side parallel to the azimuth direction. False matches are removed by the differential constrained random sample consensus (DC-RANSAC) algorithm, which appends strong constraints in azimuth direction and weak constraints in range direction. Matching points in the lower pyramid images are predicted with the local bilinear transformation model in range direction. Experiments performed on ENVISAT ASAR and Chinese airborne SAR images validated the efficiency, correct ratio and accuracy of the proposed method.

Coherent Auto-Calibration of APE and NsRCM under Fast Back-Projection Image Formation for Airborne SAR Imaging in Highly-Squint Angle

Synthetic Aperture Radar (SAR) imaging with a non-zero (forward) squint angle is capable of providing a longer time for reaction than that of the broadside mode. However, due to the large squint angle, there will be severe coupling between range and azimuth samples in the echoed data, which is known as the problematic Range Cell Migration (RCM) in the SAR community. Especially when the SAR sensor mounted on an airborne platform encounters unexpected motion deviations/errors, the coupling becomes more complicated, and it is difficult to differentiate the systematic RCM for the SAR Image Formation Processing (IFP) and the non-systematic RCM error to be compensated. To this end, a novel and accurate SAR imaging algorithm is proposed in this paper to facilitate the processing of airborne SAR data collected at a high-squint angle. Firstly, the proposed algorithm is established under a Fast Time-Domain Back-Projection (FTDBP) framework for the SAR IFP. FTDBP paves the way to avoid the complicated processing for the systematic RCM as for the conventional SAR IFP in the Doppler processing manner. It is capable of generating a high-resolution SAR image efficiently under more general geometries and configurations. Secondly, regarding the non-systematic RCM errors, the proposed algorithm realizes the compensation by correcting both the Non-systematic Range Cell Migration (NsRCM), as well as Azimuthal Phase Error (APE) in a coherent manner. It is consequently capable of auto-calibrating the effects of the motion error completely without being dependent on the airborne navigation unit. Finally, both simulated and raw data collected by the airborne squinted SAR are applied to evaluate the proposed algorithm. Comparisons with conventional algorithms are carried out to reveal the superiority of the proposed algorithm.

Visual Saliency Modeling for River Detection in High-Resolution SAR Imagery

Accurate detection of rivers plays a significant role in water conservancy construction and ecological protection, where airborne synthetic aperture radar (SAR) data have already become one of the main sources. However, extracting river information from radar data efficiently and accurately still remains an open problem. The existing methods for detecting rivers are typically based on rivers’ edges, which are easily mixed with those of artificial buildings or farmland. In addition, pixel-based image processing approaches cannot meet the requirement of real-time processing. Inspired by the feature integration and target recognition capabilities of biological vision systems, in this paper, we present a hierarchical method for automated detection of river networks in the high-resolution SAR data using biologically visual saliency modeling. For effective saliency detection, the original image is first over-segmented into a set of primitive superpixels. A visual feature set is designed to extract a regional feature histogram, which is then quantized based on the optimal parameters learned from the labeled SAR images. Afterward, three saliency measurements based on the specificity of the rivers in the SAR images are proposed to generate a single layer saliency map, i.e., local region contrast, boundary connectivity, and edge density. Finally, by exploiting belief propagation, we propose a multi-layer saliency fusion approach to derive a high-quality saliency map. Extensive experimental results on three airborne SAR image data sets with the ground truth demonstrate that the proposed saliency model consistently outperforms the existing saliency target detection models.

Extended Autofocus Backprojection Algorithm for Low-Frequency SAR Imaging

Since the trajectory deviations of a radar platform cause serious phase errors that degrade the focusing quality of synthetic aperture radar (SAR) imagery, an autofocus method is very important for high-resolution airborne SAR imaging. In this letter, an extended autofocus backprojection (EABP) algorithm is developed to accommodate the phase errors. Under the criterion of maximum image sharpness, the traditional ABP algorithm supports a broader class of collection and imaging geometries. However, it neglects the influence of SAR image energy distribution on the estimation of phase errors that make it inapplicable for SAR imaging, which has high dynamic range, such as low-frequency SAR imaging. By choosing regions and balancing the energy distribution of the data, the EABP algorithm is more efficient and useful to avoid the estimation error caused by the unbalanced energy distribution. Its performance has been demonstrated by using the experimental data that are acquired by a P-band airborne SAR system with a low-accuracy global positioning system.

Processing Sliding Mosaic Mode Data With Modified Full-Aperture Imaging Algorithm Integrating Scalloping Correction

Modified full-aperture imaging algorithm for sliding Mosaic mode synthetic aperture radar (SAR) is presented in this paper, which includes scalloping correction and spikes suppression. The full-aperture imaging algorithm is introduced into Mosaic mode and validated by real C-band airborne SAR imaging experiments. The main idea is to substitute zeros between bursts with linear-predicted data extrapolated from adjacent bursts to suppress the spikes caused by multibursts processing. Furthermore, scalloping correction for sliding Mosaic mode is integrated with this algorithm. It is innovational to correct the azimuth beam pattern weighting altered by radar antenna rotation in azimuth with deramping preprocessing operation. Finally, experiments performed by the C-band airborne SAR system with a maximum bandwidth of 200 MHz validate the effectiveness of the approach.

Ground-Based 3D Radar Imaging of Trees Using a 2D Synthetic Aperture

Motivated by the desire to gain insight into the details of conventional airborne synthetic aperture radar (SAR) imaging of trees, a ground-based SAR system designed for short-range three-dimensional (3D) radar imaging is developed using a two-dimensional (2D) synthetic aperture. The heart of the system is a compact linear frequency modulation-continuous wave (LFM-CW) radar, a custom two-dimensional scan mechanism, and a three-dimensional time-domain backprojection algorithm that generates three-dimensional backscatter images at an over-sampled resolution of 10 cm by 10 cm by 10 cm. The backprojection algorithm is formulated directly in spatial coordinates. A new method for estimating and compensating for signal attenuation within the canopy is used that exploits the backprojection image formation approach. Several three-dimensional C-band backscatter images of different individual trees of multiple species are generated from data collected for trees both in isolation and near buildings. The trees imaged in this study are about 10 m in height. The transformation of the three-dimensional images to airborne SAR images is described and a sample result provided.

Range cell migration correction analysis of one-step and two-step motion compensation for millimeter-wave airborne SAR imaging

Conventional two-step motion compensation (MOCO) method is widely adapted for airborne synthetic aperture radar (SAR) imaging due to its conciseness combining with the SAR focusing procedure. For two-step MOCO, range-independent compensation is processed before range cell migration correction (RCMC), and the range-dependent phase correction is implemented after RCMC. However, the accuracy of RCMC would be seriously decreased by the residual range-dependent phase, which is a fatal problem for high-resolution millimeter-wave (MMW) SAR imaging. In this paper, an extensive investigation on the RCMC accuracy is provided by establishing an accurate formula expression between the range cell migration error and the residual range-dependent phase error. One-step MOCO-based SAR imaging algorithm is investigated by compensating the range-dependent motion error before RCMC, so the presence of range cell migration error would be significantly suppressed. What is more, a modified azimuth match filtering (AMF) function is given by precise topography and aperture-dependent motion compensation (PTA) method to overcome the residual azimuth-dependent phase error in the azimuth compression stage. Both simulated and real-measured MMW SAR data sets are used to validate the analysis for high-resolution airborne SAR imaging.

Adaptive and Fast Prescreening for SAR ATR via Change Detection Technique

Change detection is a process of identifying changes in the state of objects between the reference and test images. This letter presents a target prescreening method that employs the change detection technique for automatic target recognition in synthetic aperture radar (SAR) images. First, four translated versions of an original SAR image are generated, and the corresponding four likelihood ratio images are computed. Then, a robust threshold is derived from the ratio of the histogram at two adjacent gray-level values of the likelihood ratio images. Finally, the threshold is applied to perform the prescreening. The proposed method implements the procedure without any prior knowledge and overcomes the weak adaptability of traditional algorithms. Two different real X-band airborne SAR images acquired over Beijing are used to quantitatively and qualitatively demonstrate the effectiveness of the proposed method.