Synthetic Aperture Radar Segmentation Research Articles

SAR Image Multiclass Segmentation Using a Multiscale TMF Model in Wavelet Domain

The triplet Markov field (TMF) model recently proposed is suitable for dealing with nonstationary synthetic aperture radar (SAR) image segmentation. In this letter, we propose a multiscale TMF model in wavelet domain, named as the wavelet-domain TMF (WTMF) model. In the WTMF model, a multiscale causal WTMF energy function is constructed to capture the intra- and interscale dependences in random fields <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$(X, U)$</tex> </formula> . Moreover, multiscale likelihoods of the WTMF model are derived based on a wavelet hidden Markov tree to capture the statistical properties of wavelet coefficients. The proposed model can integrate the global and local information in terms of spatial configuration and image features in a more complete manner. The coarser scale information is utilized to guide the finer scale segmentation, and the coarse-to-fine causal interactions are considered using a Markov chain. Experimental results prove that the proposed model can segment SAR images better than several models previously proposed.

Fast threshold selection algorithm for segmentation of synthetic aperture radar images

Automatic detection of disjoint regions in synthetic aperture radar (SAR) images for applications requiring localisation and identification of objects is complicated by the nature of the speckle. A multi-level ratio of exponential weighted averages (MROEWA) together with a fast algorithm for optimal threshold selection is proposed for SAR segmentation. A non-parametric and unsupervised principle using the grey level histogram is utilised for producing the regions that are homogenous. An optimal threshold is automatically selected by maximising the separability of the classes in grey level by incorporating a simple search strategy. Experimental results on both synthetic and real images verify the effectiveness of the proposed method. It is validated that the proposed method outperforms greatly by reducing the number of arithmetic operations required for computing the optimal threshold.

On the Use of a Shape Constraint in a Pixel-Based SAR Segmentation Algorithm

A variety of competing algorithms exist for segmentation of both single-channel and multichannel synthetic aperture radar (SAR) images. Among the most successful of these algorithms is the approach presented by Stewart et al. This algorithm defines a cost which is a weighted sum of a likelihood term that estimates the statistical likelihood of the membership of pixels to neighboring segments and a shape term that is intended to provide a smoothing constraint on segment boundaries. The shape term in the original implementation of the Stewart algorithm was rather rudimentary, and in this paper, we explore the performance of a shape term based on Sethian-Osher curvature-flow theory. We demonstrate the performance of the refined curvature-cost (CC) shape term on a set of simulated images as well as an ASAR scene. We assess the segmentation performance using a hybridized shape metric and on the number of segments produced. We find that the CC shape term significantly improves the performance of the Stewart segmentation algorithm, particularly for high-contrast edges. In spite of this success, we argue that further improvements to the algorithm will be difficult due to the architecture of the system.

Feature extraction of dual-pol SAR imagery for sea ice image segmentation

Dual-polarization synthetic aperture radar (SAR) image data, such as that available from RADARSAT-2, provide additional information for discriminating sea ice types compared with single-polarization data. We performed a thorough investigation of published feature extraction and fusion techniques to make optimal use of this additional information for unsupervised sea ice image segmentation. Segmentation was performed by transforming the dual-pol data (i) into a new two-channel feature space (multivariate) and (ii) into a fused single-channel feature space (univariate). Both real and synthetic dual-polarization SAR sea ice images were transformed using a variety of methods and segmented using a recognized SAR segmentation algorithm (IRGS). The results indicated that the untransformed data provides consistent and high segmentation accuracy, avoids feature extraction pre-processing, and is thus recommended for SAR sea ice image segmentation using dual-pol imagery.

Triplet Markov fields with edge location for fast unsupervised multi-class segmentation of synthetic aperture radar images

Triplet Markov fields (TMF) model is suitable for dealing with multi-class segmentation of non-stationary synthetic aperture radar (SAR) images. In this study, an algorithm using TMF with edge location for fast unsupervised multi-class segmentation of SAR images is proposed. The new segmentation algorithm can locate edge accurately with reasonable computational cost. First for the statistical characteristics of multiplicative speckle noise in SAR image, an edge strength based on the ratio of exponentially weighted averages operator is introduced into the Turbopixels algorithm to obtain a superpixel graph with accurate edge location in SAR images. To enhance the computational efficiency and suppress the speckle, the TMF model on pixel is generalised to that on the superpixel graph. Then, the new corresponding potential energy function and maximisation of posterior marginal segmentation formula are derived. The experimental results on synthetic and real SAR images show that the proposed algorithm can obtain accurate edge location in multi-class segmentation of SAR images, as well as enhance the computational efficiency. Especially when dealing with SAR images in large size, the proposed algorithm can give a robust and efficient result of segmentation.

SAR image segmentation based on Artificial Bee Colony algorithm

Due to the presence of speckle noise, segmentation of Synthetic Aperture Radar (SAR) images is still a challenging problem. This paper proposes a fast SAR image segmentation method based on Artificial Bee Colony (ABC) algorithm. In this method, threshold estimation is regarded as a search procedure that searches for an appropriate value in a continuous grayscale interval. Hence, ABC algorithm is introduced to search for the optimal threshold. In order to get an efficient fitness function for ABC algorithm, after the definition of grey number in Grey theory, the original image is decomposed by discrete wavelet transform. Then, a filtered image is produced by performing a noise reduction to the approximation image reconstructed with low-frequency coefficients. At the same time, a gradient image is reconstructed with some high-frequency coefficients. A co-occurrence matrix based on the filtered image and the gradient image is therefore constructed, and an improved two-dimensional grey entropy is defined to serve as the fitness function of ABC algorithm. Finally, by the swarm intelligence of employed bees, onlookers and scouts in honey bee colony, the optimal threshold is rapidly discovered. Experimental results indicate that the proposed method is superior to Genetic Algorithm (GA) based and Artificial Fish Swarm (AFS) based segmentation methods in terms of segmentation accuracy and segmentation time.

SAR Image Segmentation using Statistical Techniques

is performed in recognition applications as a primary step towards extraction of interesting regions of an image. In this paper, the characteristic effects of Weibull and Fractal parameters in the segmentation of Synthetic Aperture Radar (SAR) and Optical images acquired from satellite platform is studied. The algorithms are tested for different window sizes and different number of classes to bring out the effect of these parameters in the segmentation process.

SAR imagery segmentation by statistical region growing and hierarchical merging

This paper presents an algorithm to segment synthetic aperture radar (SAR) images, corrupted by speckle noise. Most standard segmentation techniques may require speckle filtering previously. Our approach performs radar image segmentation using the original noisy pixels as input data, i.e. without any preprocessing step. The algorithm includes a statistical region growing procedure combined with hierarchical region merging. The region growing step oversegments the input radar image, thus enabling region aggregation by employing a combination of the Kolmogorov-Smirnov (KS) test with a hierarchical stepwise optimization (HSWO) algorithm for performance improvement. We have tested and assessed the proposed technique on artificially speckled image and real SAR data.

Spectral Clustering Ensemble Applied to SAR Image Segmentation

Spectral clustering (SC) has been used with success in the field of computer vision for data clustering. In this paper, a new algorithm named SC ensemble (SCE) is proposed for the segmentation of synthetic aperture radar (SAR) images. The gray-level cooccurrence matrix-based statistic features and the energy features from the undecimated wavelet decomposition extracted for each pixel being the input, our algorithm performs segmentation by combining multiple SC results as opposed to using outcomes of a single clustering process in the existing literature. The random subspace, random scaling parameter, and Nystrom approximation for component SC are applied to construct the SCE. This technique provides necessary diversity as well as high quality of component learners for an efficient ensemble. It also overcomes the shortcomings faced by the SC, such as the selection of scaling parameter, and the instability resulted from the Nystrom approximation method in image segmentation. Experimental results show that the proposed method is effective for SAR image segmentation and insensitive to the scaling parameter.

Multi-region segmentation of SAR image by a multiphase level set approach

In this letter, a multiphase level set approach unifying region and boundary-based information for multi-region segmentation of Synthetic Aperture Radar (SAR) image is presented. An energy functional that is applicable for SAR image segmentation is defined. It consists of two terms describing the local statistic characteristics and the gradient characteristics of SAR image respectively. A multiphase level set model that explicitly describes the different regions in one image is proposed. The purpose of such a multiphase model is not only to simplify the way of denoting multi-region by level set but also to guarantee the accuracy of segmentation. According to the presented multiphase model, the curve evolution equations with respect to edge curves are deduced. The multi-region segmentation is implemented by the numeric solution of the partial differential equations. The performance of the approach is verified by both simulation and real SAR images. The experiments show that the proposed algorithm reduces the speckle effect on segmentation and increases the boundary alignment accuracy, thus correctly divides the multi-region SAR image into different homogenous regions.

A segmentation algorithm for SAR images based on the anisotropic heat diffusion equation

A method toward unsupervised segmentation of synthetic aperture radar (SAR) images is proposed. In this method, the distribution of SAR intensity image and the maximum a posteriori (MAP) algorithm is used to obtain an initial segmentation. Then according to the equivalence between the solid heat diffusion model and image scale-space, multiscale anisotropic smoothing of the posterior probability matrixes is introduced to remove the influence of speckle and to preserve important structure information. The effectiveness of this algorithm is demonstrated by application to simulated and real SAR images.

Multiscale Unsupervised Segmentation of SAR Imagery Using the Genetic Algorithm.

A valid unsupervised and multiscale segmentation of synthetic aperture radar(SAR) imagery is proposed by a combination GA-EM of the Expectation Maximization(EM) algorith with the genetic algorithm (GA). The mixture multiscale autoregressive(MMAR) model is introduced to characterize and exploit the scale-to-scale statisticalvariations and statistical variations in the same scale in SAR imagery due to radar speckle,and a segmentation method is given by combining the GA algorithm with the EMalgorithm. This algorithm is capable of selecting the number of components of the modelusing the minimum description length (MDL) criterion. Our approach benefits from theproperties of the Genetic and the EM algorithm by combination of both into a singleprocedure. The population-based stochastic search of the genetic algorithm (GA) exploresthe search space more thoroughly than the EM method. Therefore, our algorithm enablesescaping from local optimal solutions since the algorithm becomes less sensitive to itsinitialization. Some experiment results are given based on our proposed approach, andcompared to that of the EM algorithms. The experiments on the SAR images show that theGA-EM outperforms the EM method.

Integration of synthetic aperture radar image segmentation method using Markov random field on region adjacency graph

A novel approach to obtain precise segmentation of synthetic aperture radar (SAR) images using Markov random field model on region adjacency graph (MRF-RAG) is presented. First, to form a RAG, the watershed algorithm is employed to obtain an initially over-segmented image. Then, a novel MRF is defined over the RAG instead of pixels so that the erroneous segmentation caused by speckle in SAR images can be avoided and the number of configurations for the combinatorial optimisation can be reduced. Finally, a modification method based on Gibbs sampler is proposed to correct edge errors, brought by the over-segmented algorithm, in the segmentations obtained by MRF-RAG. The experimental results both on simulated and real SAR images show that the proposed method can reduce the computational complexity greatly as well as increase the segmentation precision.

Unsupervised segmentation of synthetic aperture Radar sea ice imagery using a novel Markov random field model

Environmental and sensor challenges pose difficulties for the development of computer-assisted algorithms to segment synthetic aperture radar (SAR) sea ice imagery. In this research, in support of operational activities at the Canadian Ice Service, images containing visually separable classes of either ice and water or multiple ice classes are segmented. This work uses image intensity to discriminate ice from water and uses texture features to identify distinct ice types. In order to seamlessly combine image spatial relationships with various image features, a novel Bayesian segmentation approach is developed and applied. This new approach uses a function-based parameter to weight the two components in a Markov random field (MRF) model. The devised model allows for automatic estimation of MRF model parameters to produce accurate unsupervised segmentation results. Experiments demonstrate that the proposed algorithm is able to successfully segment various SAR sea ice images and achieve improvement over existing published methods including the standard MRF-based method, finite Gamma mixture model, and K-means clustering.

Utilisation of contour criteria in micro-segmentation of SAR images

The segmentation of SAR (Synthetic Aperture Radar) images is greatly complicated by the presence of coherent speckle. To carry out this process a hierarchical segmentation algorithm based on stepwise optimization is used. It starts with each individual pixel as a segment and then sequentially merges the segment pair that minimizes the criterion. In a hypothesis testing approach, we show how the stepwise merging criterion is derived from the probability model of image regions. The Ward criterion is derived from the Gaussian additive noise model. A new criterion is derived from the multiplicative speckle noise model of SAR images. The first merging steps produce micro-regions. With standard merging criteria, the high noise level of SAR images results in the production of micro-regions that have unreliable mean and variance values and irregular shapes. If the micro-segments are not correctly delimited then the following steps will merge segments from different fields. In examining the evolution of the initial segments, we see that the merging should take into account spatial aspects. In particular, the segment contours should have good shapes. We present three measures based on contour shapes, using the perimeter, the area and the boundary length of segments. These measures are combined with the SAR criterion in order to guide correctly the segment merging process. The new criterion produces good micro-segmentation of SAR images. The criterion is also used in the following merges to produce larger segments. This is illustrated by synthetic and real image results.

Baltic Sea ice SAR segmentation and classification using modified pulse-coupled neural networks

A method for segmentation and classification of Baltic Sea ice synthetic aperture radar (SAR) images, based on pulse-coupled neural networks (PCNNs), is presented. Also, automated training, which is based on decomposing the total pixel value distribution into a mixture of class distributions, is presented and discussed. The algorithms have been trained and tested using logarithmic scale Radarsat-1 ScanSAR Wide mode images over the Baltic Sea ice. Before the decomposition into mixture of class distributions, an incidence angle correction, specifically designed for these Baltic Sea ice SAR images, is applied. Because the data distributions in the uniform areas of these images are very close to Gaussian distributions, the data are decomposed into a mixture of Gaussian distributions, using the Expectation-Maximazation algorithm. Only uniform image areas are used in the decomposition phase. The mixture of distributions is compared to the distributions of the Baltic Sea ice classes, based on earlier scatterometer measurements and visual video interpretations of the sea ice classes. The parameter values for the PCNN segmentation are defined based on this mixture of distributions. The PCNN segmentation results are also compared to the operational sea ice information of digitized ice charts and to visual interpretation of the sea ice class.

Edge location in SAR images: performance of the likelihood ratio filter and accuracy improvement with an active contour approach

The likelihood ratio edge detector is an efficient filter for the segmentation of synthetic aperture radar (SAR) images. We show that this filter provides biased location of the edge, when the window does not have the same orientation as the edge. A phenomenological model is proposed to characterize this bias. We then introduce an efficient technique to refine edge location: the statistical active contour. The combination of these two methods permits to achieve accurate and regularized edge location.

Discontinuity-adaptive Markov random field model for the segmentation of intensity SAR images

An approach for the segmentation of synthetic aperture radar (SAR) intensity images is discussed. The method integrates the gamma distribution in an objective function that exploits a discontinuity-adaptive Markov random field (DA-MRF) model that allows the user to determine the degree of homogeneity of the results according to the nature of the phenomena of interest. Both visual and numerical results illustrate its potential.

Airborne line scanner measurements for ERS-1 SAR interpretation of sea ice

A new method is presented for interpretation of Synthetic Aperture Radar (SAR) images recorded with the first European Remote Sensing Satellite (ERS-1) with respect to classification of sea ice. Conventional interpretation of spaceborne radar images is often based on subjective classification of radar signatures. Themost promising approach is a region-based segmentation with subsequent interpretation of the segments. The definition of regions with common properties enables a characterisation by texture parameters besides backscatter coefficients. The major drawback of this method is the subjective selection and characterisation of the training segments on which the classification is based. To avoid this subjective influence, an attempt is made of an independent classification by the use of two airborne line scanner systems, one for the visible and the other for the infrared spectral range. The systems were installed on research aircraft (aeroplane and helicopter) during an experiment covering the marginal ice zone north of Spitsbergen in late winter 1993 and in the North East Water Polynya in summer 1993. Several underflights of ERS-1 were carried out during the SAR mode. Typical areal coverage of the line scanner images during a flight is 1.5% of a SAR scene. The line scanner images are classified using a two-dimensional feature space. By this procedure various ice types are identified. After matching the locations of the line scanner and the segmented SAR images, it is then possible to derive the characteristics of different ice types with respect to radar signatures. Backscatter coefficient and texture parameters are used for the discrimination of ice types. Texture parameters used are moments of the grey-level distribution and values of the co-occurrence matrices. The results of this work show, that the common method of a subjective characterisation of training regions runs into difficulties. SAR signatures are essentially dependent on the conditions during ice development. Regions with different ice types can have the same SAR signatures and different SAR signatures may be found for the same ice type. With the presented method it is possible to achieve classification accuracies up to 78% with a discrimination of young ice, first-year ice and old ice during winter situations. In summer, it is difficult to discriminate wind exposed water and sea ice. In this case, the classification of SAR images is derived exclusively by means of texture parameters since the backscatter coefficient is of little use. With the discrimination of water, snow-covered ice, and brash ice an accuracy of 73% is achieved under summer conditions.

Quantitative comparison of the performance of SAR segmentation algorithms

Methods to evaluate the performance of segmentation algorithms for synthetic aperture radar (SAR) images are developed, based on known properties of coherent speckle and a scene model in which areas of constant backscatter coefficient are separated by abrupt edges. Local and global measures of segmentation homogeneity are derived and applied to the outputs of two segmentation algorithms developed for SAR data, one based on iterative edge detection and segment growing, the other based on global maximum a posteriori (MAP) estimation using simulated annealing. The quantitative statistically based measures appear consistent with visual impressions of the relative quality of the segmentations produced by the two algorithms. On simulated data meeting algorithm assumptions, both algorithms performed well but MAP methods appeared visually and measurably better. On real data, MAP estimation was markedly the better method and retained performance comparable to that on simulated data, while the performance of the other algorithm deteriorated sharply. Improvements in the performance measures will require a more realistic scene model and techniques to recognize oversegmentation.