Synthetic Aperture Radar Sea Ice Research Articles

Multi-Featured Sea Ice Classification with SAR Image Based on Convolutional Neural Network

Sea ice is a significant factor in influencing environmental change on Earth. Monitoring sea ice is of major importance, and one of the main objectives of this monitoring is sea ice classification. Currently, synthetic aperture radar (SAR) data are primarily used for sea ice classification, with a single polarization band or simple combinations of polarization bands being common choices. While much of the current research has focused on optimizing network structures to achieve high classification accuracy, which requires substantial training resources, we aim to extract more information from the SAR data for classification. Therefore we propose a multi-featured SAR sea ice classification method that combines polarization features calculated by polarization decomposition and spectrogram features calculated by joint time-frequency analysis (JTFA). We built a convolutional neural network (CNN) structure for learning the multi-features of sea ice, which combines spatial features and physical properties, including polarization and spectrogram features of sea ice. In this paper, we utilized ALOS PALSAR SLC data with HH, HV, VH, and VV, four types of polarization for the multi-featured sea ice classification method. We divided the sea ice into new ice (NI), first-year ice (FI), old ice (OI), deformed ice (DI), and open water (OW). Then, the accuracy calculation by confusion matrix and comparative analysis were carried out. Our experimental results demonstrate that the multi-feature method proposed in this paper can achieve high accuracy with a smaller data volume and computational effort. In the four scenes selected for validation, the overall accuracy could reach 95%, 91%, 96%, and 95%, respectively, which represents a significant improvement compared to the single-feature sea ice classification method.

Region-Based Sea Ice Mapping Using Compact Polarimetric Synthetic Aperture Radar Imagery with Learned Features and Contextual Information

Operational sea ice maps are usually generated manually using dual-polarization (DP) synthetic aperture radar (SAR) satellite imagery, but there is strong interest in automating this process. Recently launched satellites offer compact polarimetry (CP) imagery that provides more comprehensive polarimetric information compared to DP, which compels the use of CP for automated classification of SAR sea ice imagery. Existing sea ice scene classification algorithms using CP imagery rely on handcrafted features, while neural networks offer the potential of features that are more discriminating. We have developed a new and effective sea ice classification algorithm that leverages the nature of CP data. First, a residual-based convolutional neural network (ResCNN) is implemented to classify each pixel. In parallel, an unsupervised segmentation is performed to generate regions based on CP statistical properties. Regions are assigned a single class label by majority voting using the ResCNN output. For testing, quad-polarimetric (QP) SAR sea ice scenes from the RADARSAT Constellation Mission (RCM) are used, and QP, DP, CP, and reconstructed QP modes are compared for classification accuracy, while also comparing them to other classification approaches. Using CP achieves an overall accuracy of 96.86%, which is comparable to QP (97.16%), and higher than reconstructed QP and DP data by about 2% and 10%, respectively. The implemented algorithm using CP imagery provides an improved option for automated sea ice mapping.

Uncertainty-Incorporated Ice and Open Water Detection on Dual-Polarized SAR Sea Ice Imagery

Algorithms designed for ice–water classification of synthetic aperture radar (SAR) sea ice imagery produce only binary (ice and water) output typically using manually labeled samples for assessment. This is limiting because only a small subset of labeled samples are used, which, given the nonstationary nature of the ice and water classes, will likely not reflect the full scene. To address this, we implement a binary ice–water classification in a more informative manner considering the uncertainty associated with each pixel in the scene. To accomplish this, we have implemented a Bayesian convolutional neural network (CNN) with variational inference to produce both aleatoric (data-based) and epistemic (model-based) uncertainty. This valuable information provides feedback as to regions that have pixels more likely to be misclassified and provides improved scene interpretation. Testing was performed on a set of 21 RADARSAT-2 dual-polarization SAR scenes covering a region in the Beaufort Sea captured regularly from April to December. The model is validated by demonstrating: 1) a positive correlation between misclassification rate and model uncertainty and 2) a higher uncertainty during the melt and freeze-up transition periods, which are more challenging to classify. By incorporating the iterative region growing with semantics (IRGS) segmentation algorithm and an uncertainty value-based thresholding algorithm, the Bayesian CNN classification outputs are improved significantly via both numerical analysis and visual inspection.

Sea Ice–Water Classification of RADARSAT-2 Imagery Based on Residual Neural Networks (ResNet) with Regional Pooling

Sea ice mapping plays an integral role in ship navigation and meteorological modeling in the polar regions. Numerous published studies in sea ice classification using synthetic aperture radar (SAR) have reported high classification rates. However, many of these focus on numerical results based on sample points and ignore the quality of the inferred sea ice maps. We have designed and implemented a novel SAR sea ice classification algorithm where the spatial context, obtained by the unsupervised IRGS segmentation algorithm, is integrated with texture features extracted by a residual neural network (ResNet) and, using regional pooling, classifies ice and water. This algorithm is trained and tested on a published dataset and cross-validated using leave-one-out (LOO) strategy, obtaining an overall accuracy of 99.67% and outperforming several existing algorithms. In addition, visual results show that this new method produces sea ice maps with natural ice–water boundaries and fewer ice and water errors.

An Improved Sea Ice Classification Algorithm with Gaofen-3 Dual-Polarization SAR Data Based on Deep Convolutional Neural Networks

The distribution of sea ice is one of the major safety hazards for sea navigation. As human activities in polar regions become more frequent, monitoring and forecasting of sea ice are of great significance. In this paper, we use SAR data from the C-band synthetic aperture radar (SAR) Gaofen-3 satellite in the dual-polarization (VV, VH) fine strip II (FSII) mode of operation to study the Arctic sea ice classification in winter. SAR data we use were taken in the western Arctic Ocean from January to February 2020. We classify the sea ice into four categories, namely new ice (NI), thin first-year ice (tI), thick first-year ice (TI), and old ice (OI), by referring to the ice maps provided by the Canadian Ice Service (CIS). Then, we use the deep learning model MobileNetV3 as the backbone network, input samples of different sizes, and combine the backbone network with multiscale feature fusion methods to build a deep learning model called Multiscale MobileNet (MSMN). Dual-polarization SAR data are used to synthesize pseudocolor images and produce samples of sizes 16 × 16 × 3, 32 × 32 × 3, and 64 × 64 × 3 as input. Ultimately, MSMN can reach over 95% classification accuracy on testing SAR sea ice images. The classification results using only VV polarization or VH polarization data are tested, and it is found that using dual-polarization data could improve the classification accuracy by 10.05% and 9.35%, respectively. When other classification models are trained using the training data from this paper for comparison, the accuracy of MSMN is 4.86% and 1.84% higher on average than that of the model built using convolutional neural networks (CNNs) and ResNet18 model, respectively.

Evaluation of a Neural Network With Uncertainty for Detection of Ice and Water in SAR Imagery

Synthetic aperture radar (SAR) sea ice imagery is a promising source of data for sea ice data assimilation. Classification of SAR sea ice imagery into ice and water is of particular relevance due to its relationship with ice concentration, a key variable in sea ice data assimilation systems. With increasing volumes of SAR data, automated methods to carry out these classifications are of particular importance. Although several automated approaches have been proposed, none look at the impact of including an estimate of uncertainty of the model parameters and input features on the classification output. This article uses an established database of SAR image features to train a multilayer perceptron (MLP) neural network to classify pixel locations as either ice, water, or unknown. The classification accuracies are benchmarked using a recently developed logistic regression approach for the same database. The two methods are found to be comparable. The MLP approach is then enhanced to allow uncertainty to be estimated at each pixel location. Following methods proposed in the deep learning community, two kinds of uncertainty are considered. The first, epistemic uncertainty, is that due to uncertainty in the MLP weights. The second kind of uncertainty, aleatoric uncertainty, is that which cannot be explained by the model, and is therefore associated with the input data. It is found that including these uncertainties in the MLP models reduces their accuracies slightly, but also reduces misclassification rates. This is of particular importance for data assimilation applications, where misclassifications could severely degrade the analysis.

Fractal Analysis and Texture Classification of High-Frequency Multiplicative Noise in SAR Sea-Ice Images Based on a Transform- Domain Image Decomposition Method

Texture in synthetic aperture radar (SAR) images is a combination of the intrinsic texture of scene backscattering and the texture due to noncoherent high-frequency multiplicative noise (HMN) interactions that reflect erroneous information and lead to observation misinterpretation. The focus of this paper is the fractal analysis of KOMPSAT-5 SAR images of noncoherent sea-ice textures while being decomposed by discrete wavelet transform (DWT) processing. As a novel approach, fractal analysis relies on SAR sea-ice spatial backscattering data generation and time-frequency domain (TFD) formulations from the perspective of uncorrelated HMN. To the best of our knowledge, this is the first time that the extraction of the resolution profile and raw data for the reference KOMPSAT-5 SAR sea-ice image have been derived, evaluated and compared both qualitatively and quantitatively at each scale of DWT decomposition on the basis of the presence of HMN. This paper also presents a novel detailed modeling of the multiresolution probability distribution function of the HMN and its power spectral density function modeling at each scale of the decomposition. Other quality assessment techniques, such as two K-means clustering algorithms and several visualized verification methods, such as gradient vector field, advection mapping and tensor field mapping, have been implemented in this regard to investigate embedded HMN suppression and its adverse effects on the presence of pixel anomalies. As a result, as the decomposition continues, the HMN at each scale of decomposition is constantly altering from high-frequency uncorrelated anomalies to low-frequency joint spatial information within the observed 2-D data. In other words, excessive multiscale HMN suppression will result in spatial information loss that makes the DWT scale selection quite important for texture classification. The results also show that the importance of HMN suppression in SAR sea-ice images in the form of pixel anomaly decomposition for the purpose of further texture investigation should be in accordance with the spectral behavior of the HMN. The results are helpful for SAR remote sensing image restoration and data preservation when dealing with high-resolution SAR images, such as in time series analysis, sea-ice texture change detection, and polar structural mapping. The proposed approach is implemented on real KOMPSAT-5 SAR satellite sea-ice images, while fractal spatial resolution profile simulations are carried out based on the inversed equalized hybrid domain image formation algorithm.

Incidence Angle Correction of SAR Sea Ice Data Based on Locally Linear Mapping

Radar backscatter variations that occur because of incidence angle effects constrain the application of Scanning Synthetic Aperture Radar (ScanSAR) data for sea ice monitoring and observations. In this paper, a class-based correction is proposed for normalizing each class in ScanSAR data to a nominal incidence angle. Two tested sea ice synthetic aperture radar (SAR) data sets were acquired: a data set for the Gulf of Saint Lawrence, which was obtained by the RADARSAT-2 satellite, and a data set for the Bohai Sea, which was obtained by the ENVISAT Advanced Synthetic Aperture Radar. An unsupervised classification is performed on each image block prior to normalization, and the incidence angle range of each image block is approximately 5°. Because the distribution of the backscatter coefficients in the azimuth band is discrete and nonlinear, the class-based locally linear mapping (LLM) technique is implemented, based on the assumption that a small quantity of sorted backscatter coefficients is locally linear. This algorithm is a transplantable and easily applied method that requires limited ground data, and it is also a semiautomated technique because nearly all of its parameters can be adaptively determined during the image analysis. The results demonstrate that LLM-corrected ScanSAR images appear to have more detailed textures, and the natural signal variability in the radar data is preserved, which indicates that the LLM produces better results compared with the histogram-based-alike (HIST-alike) technique when correcting the incidence angle in the sea ice SAR data. The results of the data analysis in this paper show that the width of the azimuth band should be selected based on the extent of variation in the incidence angle, and the reference band can be calculated based on the maximum interclass distance principle. The intercomparisons also reveal that the proposed algorithm can improve the accuracy of supervised classifications.

Comparison of feature based segmentation of full polarimetric SAR satellite sea ice images with manually drawn ice charts

Abstract. In this paper we investigate the performance of an algorithm for automatic segmentation of full polarimetric, synthetic aperture radar (SAR) sea ice scenes. The algorithm uses statistical and polarimetric properties of the backscattered radar signals to segment the SAR image into a specified number of classes. This number was determined in advance from visual inspection of the SAR image and by available in situ measurements. The segmentation result was then compared to ice charts drawn by ice service analysts. The comparison revealed big discrepancies between the charts of the analysts, and between the manual and the automatic segmentations. In the succeeding analysis, the automatic segmentation chart was labeled into ice types by sea ice experts, and the SAR features used in the segmentation were interpreted in terms of physical sea ice properties. Utilizing polarimetric information in sea ice charting will increase the efficiency and exactness of the maps. The number of classes used in the segmentation has shown to be of significant importance. Thus, studies of automatic and robust estimation of the number of ice classes in SAR sea ice scenes will be highly relevant for future work.

Classification of Sea Ice Types in ENVISAT Synthetic Aperture Radar Images

In this paper, sea ice in the Central Arctic has been classified in synthetic aperture radar (SAR) images from ENVISAT using a neural network (NN)-based algorithm and a Bayesian algorithm. Since different sea ice types can have similar backscattering coefficients at C-band HH polarization, it is necessary to use textural features in addition to the backscattering coefficients. The analysis revealed that the most informative texture features for the classification of multiyear ice (MYI), deformed first-year ice (FYI) (DFYI), and level FYI (LFYI) and open water/nilas are correlation, inertia, cluster prominence, energy, homogeneity, and entropy, as well as third and fourth central statistical moments of image brightness. The optimal topology of the NN, trained for ENVISAT wide-swath SAR sea ice classification, consists of nine neurons in input layer, six neurons in hidden layer, and three neurons in output layer. The classification results for a series of 20 SAR images, acquired in the central part of the Arctic Ocean during winter months, were compared to expert analysis of the images and ice charts. The results of the NN classification show that the average correspondences with the expert analysis amount to 85 $\%$ , 83 $\%$ , and 80 $\%$ for LFYI, DFYI, and MYI, respectively. The Bayesian pixel-based method can provide a higher resolution in the classified image and, therefore, better capability to identify leads compared to the NN method. Both methods may be effectively used in the Central Arctic where MYI is predominant.

Two-dimensional wavelet variance estimation with application to sea ice SAR images

The surface of Arctic sea ice presents complex patterns of cracks and ridges that change with the seasons according to the external forces acting on the ice and the internal stresses within the ice. We propose a new statistical tool for analysis of these patterns based on a two-dimensional Maximal Overlap Discrete Wavelet Transform (MODWT) of Synthetic Aperture Radar (SAR) images, which can be used to track how ice conditions change over the course of the year. Here we give details on an extended pyramid algorithm that efficiently computes the MODWT coefficients for all combinations of vertical and horizontal scales. We show how to use these coefficients to form mean- and median-based wavelet variance estimates along with confidence intervals for the true unknown variances. We demonstrate the usefulness of the statistical tool on images acquired by the SAR sensor onboard RADARSAT, but the tool is of potential use in other geoscience applications and in other areas (e.g. medical imaging). We provide a Matlab implementation of this tool but also give sufficient details so that it can be encoded in other languages.

Evaluating SAR Sea Ice Image Segmentation Using Edge-Preserving Region-Based MRFs

This paper presents a new approach to sea ice segmentation in synthetic aperture radar (SAR) intensity images by combining an edge-preserving region (EPR)-based representation with region-level MRF models. To construct the EPR-based representation of a SAR image, edge strength is measured using instantaneous coefficient of variation (ICOV) upon which the watershed algorithm is applied to partition the image into primitive regions. In addition, two new metrics for quantitative assessment of region characteristics (region accuracy and region redundancy) are defined and used for parameter estimation in the ICOV extraction process towards desired region characteristics. In combination with a region-level MRF, the EPR-based representation facilitates the segmentation process by largely reducing the search space of optimization process and improving parameter estimation of feature model, leading to considerable computational savings and less probability of false segmentation. The proposed segmentation method has been evaluated using a synthetic sea ice image corrupted with varying levels of speckle noise as well as real SAR sea ice images. Relative to the existing region-level MRF-based methods, testing results have demonstrated that our proposed method substantially improves the segmentation accuracy at high speckle noise and achieves on average 29% reduction of computational time.

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.

SAR Sea-Ice Image Analysis Based on Iterative Region Growing Using Semantics

Synthetic aperture radar (SAR) has been intensively used for sea-ice monitoring in polar regions. A computer-assisted analysis of SAR sea-ice imagery is extremely difficult due to numerous imaging parameters and environmental factors. This paper presents a system which, with some limited information provided, is able to perform an automated segmentation and classification for the SAR sea-ice imagery. In the system, both the segmentation and classification processes are based on a Markov random-field theory and are formulated in a joint manner under the Bayesian framework. Solutions to the formulation are obtained by a region-growing technique which keeps refining the segmentation and producing semantic class labels at the same time in an iterative manner. The algorithm is a general-segmentation approach named iterative region growing using semantics, which, in this paper, is dedicated to the problem of classifying the operational SAR sea-ice imagery provided by the Canadian Ice Service (CIS). The classified image results have been validated by the CIS personnel, and the resulting classifications are quite successful using the same algorithm applied to diverse data sets.

Filament Preserving Model (FPM) Segmentation Applied to SAR Sea-Ice Imagery

Modeling spatial context constraints using a Markov random field (MRF) has been widely used in the segmentation of noisy images. Its applicability to synthetic aperture radar (SAR) sea-ice segmentation has also been demonstrated recently. However, most existing MRF models are not capable of preserving filaments, specifically leads and ridges for SAR sea ice, which are valuable for ship navigation applications and necessary for identifying certain ice types. In this paper, a new statistical context model is proposed that, within the same scene, can simultaneously preserve narrow elongated features while producing similar smooth segmentation results comparable to typical MRF-based approaches. Tested on one synthetic image and two SAR sea-ice scenes, this filament preserving model substantially improves classification accuracies when compared to standard Gaussian mixture and MRF-based segmentation algorithms

Dependence between standard deviation and measurement length for C-band backscattering signatures of the Baltic Sea ice

This paper studies whether the standard deviation (std) of the Baltic Sea ice backscattering coefficient (sigmadeg) depends on the length of measurement (l). For many kinds of surfaces, especially for a fractal one, this is the case. The study was conducted using one-dimensional C-band helicopter-borne scatterometer data and ENVISAT synthetic aperture radar (SAR) images. The results with both data sets indicate mostly a strong linear dependence between ln(l) and ln(std(sigmadeg)) up to a distance of at least a few kilometers. Based on the analysis of empirical and simulated data (fractal and nonfractal profiles), it seems that sea ice sigmadeg as a function of l is not completely described either by fractional Brownian motion or by a process with a single-scale autocorrelation function. Neither can the values of sigmadeg be regarded as samples from only one probability distribution. The regression coefficients describing the dependency of ln(l) versus ln(std(sigmadeg)) do not discriminate various ice types better than just mean and std of sigmadeg. However, the use of regression coefficients instead of mean and std is preferred due to their scale-invariant comparability with the results of other studies. The dependence of std(sigmadeg) on l should also be taken generally into account in the data analysis, e.g., when constructing classifiers for sea ice SAR data

Operational map-guided classification of SAR sea ice imagery

This paper presents a map-guided sea ice classification system built to work in parallel with the Canadian Ice Service (CIS) operations to produce pixel-based ice maps that complement actual maps produced by CIS. The system uses the CIS maps as input to guide classification by providing information on the number of ice types and their final label for specific regions. Segmentation is based on a modified adaptive Markov random field (MRF) model that uses synthetic aperture radar (SAR) intensities and texture features as input. The ice type labeling is performed automatically by gathering evidences based on a priori information on one or two classes and deducing the other labels iteratively by comparing distributions of segments. Three methods for comparing the segment distributions (Fisher criterion, Mahalanobis distance, and Kolmogorov-Smirnov test) were implemented and compared. The system is fully described with special attention to the labeling procedure. Examples are presented in the form of two CIS SAR-based ice maps from the Gulf of Saint Lawrence region and one example from the Beaufort Sea. The results indicate that when the segmentation is good, the labeling attains best results (between 71% and 89%) based on evaluation by a sea ice analyst. Some problems remain to be assessed which are primarily attributable to discrepancies in the information provided by the egg code and what is actually visible in the SAR image. Subscale information on floe size and shape available to human analysts, but not in this classification system, also appear to be a critical information for separating some ice types.

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.

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.

Comparing Cooccurrence Probabilities and Markov Random Fields for Texture Analysis of SAR Sea Ice Imagery

This paper compares the discrimination ability of two texture analysis methods: Markov random fields (MRFs) and gray-level cooccurrence probabilities (GLCPs). There exists limited published research comparing different texture methods, especially with regard to segmenting remotely sensed imagery. The role of window size in texture feature consistency and separability as well as the role in handling of multiple textures within a window are investigated. Necessary testing is performed on samples of synthetic (MRF generated), Brodatz, and synthetic aperture radar (SAR) sea ice imagery. GLCPs are demonstrated to have improved discrimination ability relative to MRFs with decreasing window size, which is important when performing image segmentation. On the other hand, GLCPs are more sensitive to texture boundary confusion than MRFs given their respective segmentation procedures.