Multitemporal Synthetic Aperture Radar Images Research Articles

Change Detection in SAR Images via Ratio-Based Gaussian Kernel and Nonlocal Theory

Compared with the synthetic aperture radar (SAR) image processing theory based on local neighborhood, the nonlocal theory is not limited to a local neighborhood of an image and has great potential in change detection of SAR images. In this study, an approach using ratio-based nonlocal information (RNLI) is proposed for change detection in multitemporal SAR images. First, the RNLI is extracted from a spatial–temporal nonlocal neighborhood where the similarity of two pixels in the nonlocal neighborhood is well characterized by the proposed ratio-based Gaussian kernel function. The parameters of RNLI: noise level and matching window size are adaptively determined to avoid the uncertainty of the change detection result caused by user experience. Second, the difference image is generated by using the RNLI and the ratio operator. Finally, the change map is obtained by segmenting the difference image with a threshold. Experiments conducted on two real datasets and two simulated datasets showed that the proposed method performed better than the other advanced change detection methods, which can better retain the edge information of the changed area while reducing the overall error of the change detection results.

Change Detection in Multitemporal SAR Images Based on Slow Feature Analysis Combined With Improving Image Fusion Strategy

Change detection in multitemporal synthetic aperture radar (SAR) images has been an important research content in the field of remote sensing for a long time. In this article, based on the slow feature analysis (SFA) theory and the nonsubsampled contourlet transform (NSCT) algorithm, we propose a novel unsupervised change detection method called NSCT nonlocal means (NSCT-NLM). The powerful extraction to the changed information of SFA and the superior information fusion of NSCT are jointly adopted in this method. The main framework consists of the following parts. First, SFA and the log-ratio operator are used to generate difference images (DIs) independently. Then, the NSCT is used to fuse two DIs into a new higher quality DI. The newly fused DI combines the complementary information of the two kinds of original DI. Therefore, the contrast of the changed regions and unchanged regions is greatly enhanced, as well as the changed details are preserved more completely. Furthermore, an NLM filtering algorithm is employed to suppress the strong speckles in the fused DI. We use the fuzzy C-means algorithm to generate the final binary change map. The experiments are carried out on two public datasets and three real-world SAR datasets from different scenarios. The results demonstrate that the proposed method has higher detection accuracy by comparing with the reference methods.

Ship Detection in Complex Environment Using SAR Time Series

Ship detection in complex environment is a challenging task due to strong background inferences, for which various deep-learning-based methods have been proposed. However, they have poor performance on detecting nearshore ships for medium-resolution synthetic aperture radar (SAR) images due to the loss of typical features and the confusion with the land scatterers. The availability of multitemporal SAR images gives the opportunity to separate nearshore ships with land scatterers by using the temporal characteristics. In this article, we propose a ship detection method based on SAR time series. First, we investigate the statistical stability of the SAR time series and propose a preclassification method to identify the potential changed pixel clusters. Then, we discriminate between ship and background pixel candidates in the preclassification by combining a rotating object detector and the transition detection algorithm and generate the corresponding frozen background reference (FBR) image. In addition, a dynamic framework for ship detection is proposed based on the FBR image and a two-stage outlier detection approach. The experiments show that the proposed method enables a dynamic ship monitoring with a high accuracy in ship detection and low false alarm rate for nearshore ship targets.

Fuzzy Active Contour Model With Markov Random Field for Change Detection

The traditional active contour models are sensitive to the speckle noise in the synthetic aperture radar (SAR) images. In this paper, the Markov random field (MRF) theory is incorporated into the fuzzy active contour model to detect the changes of multitemporal SAR images. In the proposed method, neighboring information is considered to modify the pointwise prior probability for exploiting the mutual and spatial information. In addition, we incorporate MRF into the fuzzy active contour model and get the resulting MRF-based energy function. Finally, we drive the associated first variation of the energy function to compute the fuzzy membership. Due to the introduction of MRF, the proposed MRF-based fuzzy active contour model is robust to the speckle noise in the SAR images and can achieve accurate change detection results. Experiments on four SAR image datasets demonstrate that the proposed MRF-based fuzzy active contour model is able to accurately segment the difference image and has better performance in comparison with other change detection techniques.

Sentinel-2 Cloud Removal Considering Ground Changes by Fusing Multitemporal SAR and Optical Images

Publicly available optical remote sensing images from platforms such as Sentinel-2 satellites contribute much to the Earth observation and research tasks. However, information loss caused by clouds largely decreases the availability of usable optical images so reconstructing the missing information is important. Existing reconstruction methods can hardly reflect the real-time information because they mainly make use of multitemporal optical images as reference. To capture the real-time information in the cloud removal process, Synthetic Aperture Radar (SAR) images can serve as the reference images due to the cloud penetrability of SAR imaging. Nevertheless, large datasets are necessary because existing SAR-based cloud removal methods depend on network training. In this paper, we integrate the merits of multitemporal optical images and SAR images to the cloud removal process, the results of which can reflect the ground information change, in a simple convolution neural network. Although the proposed method is based on deep neural network, it can directly operate on the target image without training datasets. We conduct several simulation and real data experiments of cloud removal in Sentinel-2 images with multitemporal Sentinel-1 SAR images and Sentinel-2 optical images. Experiment results show that the proposed method outperforms those state-of-the-art multitemporal-based methods and overcomes the constraint of datasets of those SAR-based methods.

Change Detection from SAR Images Based on Convolutional Neural Networks Guided by Saliency Enhancement

Change detection is an important task in identifying land cover change in different periods. In synthetic aperture radar (SAR) images, the inherent speckle noise leads to false changed points, and this affects the performance of change detection. To improve the accuracy of change detection, a novel automatic SAR image change detection algorithm based on saliency detection and convolutional-wavelet neural networks is proposed. The log-ratio operator is adopted to generate the difference image, and the speckle reducing anisotropic diffusion is used to enhance the original multitemporal SAR images and the difference image. To reduce the influence of speckle noise, the salient area that probably belongs to the changed object is obtained from the difference image. The saliency analysis step can remove small noise regions by thresholding the saliency map, and interest regions can be preserved. Then an enhanced difference image is generated by combing the binarized saliency map and two input images. A hierarchical fuzzy c-means model is applied to the enhanced difference image to classify pixels into the changed, unchanged, and intermediate regions. The convolutional-wavelet neural networks are used to generate the final change map. Experimental results on five SAR data sets indicated the proposed approach provided good performance in change detection compared to state-of-the-art relative techniques, and the values of the metrics computed by the proposed method caused significant improvement.

다중시기 SAR 영상 간 정합 정도에 따른 변화탐지 성능분석 연구

Multi-temporal synthetic aperture radar (SAR) images, which are obtained to monitor the ground area at different observation times, are utilized in change detection (CD) to identify changing objects in the observed area. However, even with co-registration processing, it is impossible to achieve perfectly matching multi-temporal SAR images. In this study, we compared the change detection performance of four different CD techniques under various co-registration conditions. We analyzed the change detection performance of these techniques in terms of detection and false alarm rates based on pixel units. The experimental results show that the performance of classical CCD degrades significantly as the degree of co-registration degrades, while Berger-CCD achieves a sufficient change detection performance even under sub-par co-registration conditions.

Recognition of Autumn Crop Based on Polsar Data and Feature Selection

Optical remote sensed images have been intensively used to map global and regional agriculture information. However, few optical images could be collected due to cloud contamination during the crop growth period. Therefore, synthetic aperture radar (SAR) images could be used to extract crop distribution since it is capable of acquiring data without regard to bad weather conditions. Although numerous studies have been successfully carried out to highlight the potential of SAR images in crop monitoring, there are still some problems to be solved. The high dimensionality of multi-temporal SAR images remains a major issue, classification using all features is limited to efficiency. In this study, a method of autumn crop recognition based on PolSAR data and feature selection was proposed. Using Radarsat-2 PolSAR images acquired during the autumn crop growing season, the optimal subset of features was obtained by 3 feature selection methods and 15 polarimetric target decomposition methods. With optimal feature subset and train samples, crops and other ground objects were classified by SVM. The classification result was assessed by test samples.

Speckle2Void: Deep Self-Supervised SAR Despeckling With Blind-Spot Convolutional Neural Networks

Information extraction from synthetic aperture radar (SAR) images is heavily impaired by speckle noise, and hence, despeckling is a crucial preliminary step in scene analysis algorithms. The recent success of deep learning envisions a new generation of despeckling techniques that could outperform classical model-based methods. However, current deep learning approaches to despeckling require supervision for training, whereas clean SAR images are impossible to obtain. In the literature, this issue is tackled by resorting to either synthetically speckled optical images, which exhibit different properties with respect to true SAR images, or multitemporal SAR images, which are difficult to acquire or fuse accurately. In this article, inspired by recent works on blind-spot denoising networks, we propose a self-supervised Bayesian despeckling method. The proposed method is trained by employing only noisy SAR images and can, therefore, learn features of real SAR images rather than synthetic data. Experiments show that the performance of the proposed approach is very close to the supervised training approach on synthetic data and superior on real data in both quantitative and visual assessments.

A Wavelength-Resolution SAR Change Detection Method Based on Image Stack through Robust Principal Component Analysis

Recently, it was demonstrated that low-frequency wavelength-resolution synthetic aperture radar (SAR) images could be considered to follow an additive mixing model due to their backscatter characteristics. This simplification allows for the use of source separation methods, such as robust principal component analysis (RPCA) via principal component pursuit (PCP), for detecting changes in those images. In this manuscript, a change detection method for wavelength-resolution SAR images based on image stack through RPCA is proposed. The method aims to explore both the temporal and flight heading diversity of a set of wavelength-resolution multitemporal SAR images in order to detect concealed targets in forestry areas. A heuristic based on three rules for better exploring the RPCA results is introduced, and a new configurable parameter for false alarm reduction based on the analysis of image windows is proposed. The method is evaluated using real data obtained from measurements of the ultrawideband (UWB) very high-frequency (VHF) SAR system CARABAS-II. Experiments for stacks of four and seven reference images are conducted, and the use of reference images acquired with different flight headings is explored. The results indicate that a gain in performance can be achieved by using large image stacks containing, at least, one image of each possible flight heading of the data set, which can result in a probability of detection (PD) above 99% for a false alarm rate (FAR) as low as one false alarm per three square kilometers. Furthermore, it is demonstrated that high PD and low FAR can be achieved, also considering images from similar flight headings as reference images.

Unsupervised Multitemporal Building Change Detection Framework Based on Cosegmentation Using Time-Series SAR

Building change detection using remote sensing images is essential for various applications such as urban management and marketing planning. However, most change detection approaches can only detect the intensity or type of change. The aim of this study is to dig for more change information from time-series synthetic aperture radar (SAR) images, such as the change frequency and the change moments. This paper proposes a novel multitemporal building change detection framework that can generate change frequency map (CFM) and change moment maps (CMMs) from multitemporal SAR images. We first give definitions of CFM and CMMs. Then we generate change feature using four proposed generators. After that, a new cosegmentation method combining raw images and change feature is proposed to divide time-series images into changed and unchanged areas separately. Secondly, the proposed cosegmentation and the morphological building index (MBI) are combined to extract changed building objects. Then, the logical conjunction between the cosegmentation results and the binarized MBI is performed to recognize every moment of change. In the post-processing step, we use fragment removal to increase accuracy. Finally, we propose a novel accuracy assessment index for CFM. We call this index average change difference (ACD). Compared to the traditional multitemporal change detection methods, our method outperforms other approaches in terms of both qualitative results and quantitative indices of ACD using two TerraSAR-X datasets. The experiments show that the proposed method is effective in generating CFM and CMMs.

Synthetic Aperture Radar Image Change Detection via Siamese Adaptive Fusion Network

Synthetic aperture radar (SAR) image change detection is a critical yet challenging task in the field of remote sensing image analysis. The task is nontrivial due to the following challenges: First, intrinsic speckle noise of SAR images inevitably degrades the neural network because of error gradient accumulation. Furthermore, the correlation among various levels or scales of feature maps is difficult to be achieved through summation or concatenation. Toward this end, we proposed a siamese adaptive fusion (AF) network for SAR image change detection. To be more specific, two-branch CNN is utilized to extract high-level semantic features of multitemporal SAR images. Besides, an AF module is designed to adaptively combine multiscale responses in convolutional layers. Therefore, the complementary information is exploited, and feature learning in change detection is further improved. Moreover, a correlation layer is designed to further explore the correlation between multitemporal images. Thereafter, robust feature representation is utilized for classification through a fully connected layer with softmax. Experimental results on four real SAR datasets demonstrate that the proposed method exhibits superior performance against several state-of-the-art methods. Our codes are available at https://github.com/summitgao/SAR_CD_SAFNet .

MULTI-TEMPORAL SAR IMAGE DESPECKLING BASED A CONVOLUTIONAL NEURAL NETWORK

Abstract. Speckle noise is an intrinsic property of Synthetic Aperture Radar (SAR) imagery, which affects the quality of image. Single-temporal despeckling methods usually pay attention to the utilization of spatial information, but sometimes due to lack of sufficient information, the despeckling image is too smooth or losses some information about edge details. However, multi-temporal SAR images can provide extra information for despeckling resulting in better performance. Therefore, in this paper, we proposed a novel multi-temporal SAR despeckling method based a convolutional neural network (MSAR-CNN) embedded temporal and spatial attention (TSA) module to deeply mine the spatial and temporal correlation of multitemporal SAR images. The whole network, which is end-to-end trained with simulate realistic SAR data, consists of several residual blocks. In addition, the simulated and real-data experiments demonstrate that the proposed MSAR-CNN outperforms most of the mainstream methods in both the quantitative evaluation indexes and visual effects.

ESTIMATING 3D LAND SUBSIDENCE FROM MULTI-TEMPORAL SAR IMAGES AND GNSS DATA USING WEIGHTED LEAST SQUARES

Abstract. Analysis of multi-temporal synthetic aperture radar (SAR) satellite images using persistent scatterer interferometry is an effective approach for monitoring land subsidence, which is a serious issue in some urban areas. However, a drawback to this approach is that it is limited to displacement along the radar line-of-sight direction. An accurate understanding of land subsidence requires estimation of 3D displacement. One solution is to combine observations from multiple sources and directions, such as multi-temporal SAR images acquired on ascending and descending orbits, with global navigation satellite system (GNSS) data. While this approach estimates 3D displacement, other methods do not account for differences in data accuracy. Therefore, in this paper, we propose a method for estimating 3D land subsidence from multi-temporal SAR images and GNSS data by using the weighted least squares method. The weights for data sources are calculated from the PSI results and GNSS data. We apply the method to Kansai International Airport, using 13 ALOS-2/PALSAR-2 ascending images from 2014 to 2018 and 17 ALOS-2/PALSAR-2 descending images from 2015 to 2018. Root mean squared errors in the east–west, north–south and vertical directions are 6, 13, and 10 mm/year, respectively. These results demonstrate that combining PSI and geodetic results is effective for monitoring land deformation accurately with high spatial resolution.

A Novel Rayleigh Dynamical Model for Remote Sensing Data Interpretation

This article introduces the Rayleigh autoregressive moving average (RARMA) model, which is useful to interpret multiple different sets of remotely sensed data, from wind measurements to multitemporal synthetic aperture radar (SAR) sequences. The RARMA model is indeed suitable for continuous, asymmetric, and nonnegative signals observed over time. It describes the mean of Rayleigh-distributed discrete-time signals by a dynamic structure including autoregressive (AR) and moving average (MA) terms, a set of regressors, and a link function. After presenting the conditional likelihood inference for the model parameters and the detection theory, in this article, a Monte Carlo simulation is performed to evaluate the finite signal length performance of the conditional likelihood inferences. Finally, the new model is applied first to sequences of wind speed measurements, and then to a multitemporal SAR image stack for land-use classification purposes. The results in these two test cases illustrate the usefulness of this novel dynamic model for remote sensing data interpretation.

Building Change Detection in VHR SAR Images via Unsupervised Deep Transcoding

Building change detection (CD), important for its application in urban monitoring, can be performed in near real time by comparing prechange and postchange very-high-spatial-resolution (VHR) synthetic-aperture-radar (SAR) images. However, multitemporal VHR SAR images are complex as they show high spatial correlation, prone to shadows, and show an inhomogeneous signature. Spatial context needs to be taken into account to effectively detect a change in such images. Recently, convolutional-neural-network (CNN)-based transfer learning techniques have shown strong performance for CD in VHR multispectral images. However, its direct use for SAR CD is impeded by the absence of labeled SAR data and, thus, pretrained networks. To overcome this, we exploit the availability of paired unlabeled SAR and optical images to train for the suboptimal task of transcoding SAR images into optical images using a cycle-consistent generative adversarial network (CycleGAN). The CycleGAN consists of two generator networks: one for transcoding SAR images into the optical image domain and the other for projecting optical images into the SAR image domain. After unsupervised training, the generator transcoding SAR images into optical ones is used as a bitemporal deep feature extractor to extract optical-like features from bitemporal SAR images. Thus, deep change vector analysis (DCVA) and fuzzy rules can be applied to identify changed buildings (new/destroyed). We validate our method on two data sets made up of pairs of bitemporal VHR SAR images on the city of L’Aquila (Italy) and Trento (Italy).

RETRACTED ARTICLE: A feature based change detection approach using multi-scale orientation for multi-temporal SAR images

ABSTRACT Excellent operation regardless of weather conditions and superior resolution independent of sensor light are the most attractive and desired features of synthetic aperture radar (SAR) imagery. This paper proposes an exclusive multi-scale with multiple orientation approach for multi-temporal SAR images. This approach integrates pre-processing and change detection. Pre-processing is performed on the SAR imagery through speckle reducing anisotropic diffusion and discrete wavelet transform. The processed speckle-free images are designed by Log-Gabor filter bank in terms of multi-scale with multiple orientations. The maximum magnitude of multiple orientations is concatenated to obtain feature-based scale representation. Each scale is dealt with multiple orientations and is compared by band-wise subtraction to retrieve difference image (DI) coefficient. The series of the difference coefficients from each scale are add-on together to estimate a DI. Thus, the resultant image of multi-scale orientation gives perception of detailed information with specific contour. Constrained k-means clustering algorithm is preferred to achieve change and un-change map. Performance of the proposed approach is validated on three real SAR image datasets. The effective change detection is examined by using confusion matrix parameters. Experimental results are described to show the efficacy of the proposed approach.

The applications of InSAR technique for natural hazard detection in smart society

Synthetic Aperture Radar (SAR) is a potential application of remote sensing to geological and hydrometeorological hazards. This paper presents sustainability strategies for smart cities: the use of SAR Sentinel-1 for monitoring flood inundation and landslide hazards in Aceh Province, Indonesia. In this study, for flood detection, we attempt to uses Sentinel-1A (S-1A) in the same direction and acquisition through polarization of Vertical transmit and Vertical received (VV) – Vertical transmit and Horizontal received (VH) with a temporal baseline of 6 days. Those data were then analysed using the SNAP Toolbox. The results showed that the S-1A was successfully for detecting a flood inundation in which VH polarization is more sensitive than VV. For landslide monitoring, we apply multitemporal SAR images, where one of them is the Quasi-Persistent Scatterers (Q-PS) technique. Using ascending and descending orbit pass results in a better velocity map where both sides of the slope are detected due to the different sensor angle of both orbital passes. This technique has resulted in the undulating areas being monitored well and this will also fill the gap of layover and shadowing phenomena of the slant range SAR image. The Q-PS combinations were very effective to identify the deformation features associated with the land movement. For a smart city, natural hazards such as landslides and floods require continuous monitoring to ensure the potential as well as management and mitigation of natural hazards. Therefore, the InSAR technique is one of the effective ways for doing this strategy.

A Novel Active Contours Model for Environmental Change Detection from Multitemporal Synthetic Aperture Radar Images

In this paper, we propose a novel approach based on the active contours model for change detection from synthetic aperture radar (SAR) images. In order to increase the accuracy of the proposed approach, a new operator was introduced to generate a difference image from the before and after change images. Then, a new model of active contours was developed for accurately detecting changed regions from the difference image. The proposed model extracts the changed areas as a target feature from the difference image based on training data from changed and unchanged regions. In this research, we used the Otsu histogram thresholding method to produce the training data automatically. In addition, the training data were updated in the process of minimizing the energy function of the model. To evaluate the accuracy of the model, we applied the proposed method to three benchmark SAR data sets. The proposed model obtains 84.65%, 87.07%, and 96.26% of the Kappa coefficient for Yellow River Estuary, Bern, and Ottawa sample data sets, respectively. These results demonstrated the effectiveness of the proposed approach compared to other methods. Another advantage of the proposed model is its high speed in comparison to the conventional methods.

Wetland Surface Water Detection from Multipath SAR Images Using Gaussian Process-Based Temporal Interpolation

Wetlands provide society with a myriad of ecosystem services, such as water storage, food sources, and flood control. The ecosystem services provided by a wetland are largely dependent on its hydrological dynamics. Constant monitoring of the spatial extent of water surfaces and the duration of flooding of a wetland is necessary to understand the impact of drought on the ecosystem services a wetland provides. Synthetic aperture radar (SAR) has the potential to reveal wetland dynamics. Multitemporal SAR image analysis for wetland monitoring has been extensively studied based on the advances of modern SAR missions. Unfortunately, most previous studies utilized monopath SAR images, which result in limited success. Tracking changes in individual wetlands remains a challenging task because several environmental factors, such as wind-roughened water, degrade image quality. In general, the data acquisition frequency is an important factor in time series analysis. We propose a Gaussian process-based temporal interpolation (GPTI) method that enables the synergistic use of SAR images taken from multiple paths. The proposed model is applied to a series of Sentinel-1 images capturing wetlands in Okanogan County, Washington State. Our experimental analysis demonstrates that the multiple path analysis based on the proposed method can extract seasonal changes more accurately than a single path analysis.