Remote sensing sea ice classification based on DenseNet and heterogeneous data fusion

Recent drastic reductions in the Arctic sea-ice cover have raised an interest in understanding the role of sea ice in the global system as well as pointed out a need to understand the physical processes that lead to such changes. Satellite remote-sensing data provide important information about remote ice areas, and Synthetic Aperture Radar (SAR) data have the advantages of penetration of the omnipresent cloud cover and of high spatial resolution. A challenge addressed in this paper is how to extract information on sea-ice types and sea-ice processes from SAR data. We introduce, validate and apply geostatistical and statistical approaches to automated classification of sea ice from SAR data, to be used as individual tools for mapping sea-ice properties and provinces or in combination. A key concept of the geostatistical classification method is the analysis of spatial surface structures and their anisotropies, more generally, of spatial surface roughness, at variable, intermediate-sized scales. The geostatistical approach utilizes vario parameters extracted from directional vario functions, the parameters can be mapped or combined into feature vectors for classification. The method is flexible with respect to window sizes and parameter types and detects anisotropies. In two applications to RADARSAT and ERS-2 SAR data from the area near Point Barrow, Alaska, it is demonstrated that vario-parameter maps may be utilized to distinguish regions of different sea-ice characteristics in the Beaufort Sea, the Chukchi Sea and in Elson Lagoon. In a third and a fourth case study the analysis is taken further by utilizing multi-parameter feature vectors as inputs for unsupervised and supervised statistical classification. Field measurements and high-resolution aerial observations serve as basis for validation of the geostatistical-statistical classification methods. A combination of supervised classification and vario-parameter mapping yields best results, correctly identifying several sea-ice provinces in the shore-fast ice and the pack ice. Notably, sea ice does not have to be static to be classifiable with respect to spatial structures. In consequence, the geostatistical-statistical classification may be applied to detect changes in ice dynamics, kinematics or environmental changes, such as increased melt ponding, increased snowfall or changes in the equilibrium line.

Sea ice is critical to map for safe and efficient maritime navigation, to mitigate ship trapping and capsizing. Sea ice is also vital to monitor to assess the state of the changing climate and a critical component in climate and weather models, reflecting sunlight towards space and acting as an insulating material between the ocean surface and the atmosphere.Professional sea ice analysts at national ice services map sea ice based on Synthetic Aperture Radar (SAR) images acquired by satellites, such as the Sentinel-1 (S1) satellite constellation. The ice analysts manually interpret the SAR images using their in-depth knowledge and experience to create sea ice charts with information on the sea ice conditions.A challenge for the S1 5.4 GHz SAR measurements is that the radar wave does not penetrate deep into the sea ice and is scattered/reflected by the surface. Therefore, the SAR images provide information primarily about the sea ice surface, useful for identifying and classify sea ice conditions. The charts describe, among others, the sea ice’s stage of development - the type of sea ice - an indicator of its thickness. The manual charting process apply sea ice classes, defined by the International Ice Charting Working Group (IICWG) on behalf of the World Meteorological Organization (WMO). Considerable uncertainties are associated with the ice classes that can vary from, e.g. 30-200cm or 70-120cm in thickness. Deep-learning models that produce stage-of-development information from S1 radar images exist but has the same inherent limitations of the sea ice charts in the model outputs.Current state-of-the-art sea ice thickness retrieval methods relies on altimeter satellites, such as the CryoSat-2 (CS2) satellite. The distance between the ocean and the sea ice is measured, known as the sea ice freeboard. For a Ku-band radar altimeter like CS2, it is assumed that the radar response penetrates the snow and returns from the sea ice surface. As the true penetration is unknown, and the radar wave propagation is delayed when the signal passes through snow, the measured quantity is known as the radar freeboard.The sea ice thickness can be estimated with an accuracy of 20-40% using the radar freeboard by calculating the sea ice's buoyancy based on snow and ice density estimates, and auxiliary snow depth information. However, CS2 only measures 1600m across the orbit and can thus only monitor sea ice thickness in the Arctic monthly - insufficient for many applications, such as maritime navigation, and leaves data record gaps. S1 SAR on the other hand, cover 400km in Extra Wide mode across the orbit with repeating coverage every week.Here, we present our preliminary results of circumventing the limitations of CS2 and S1 by training supervised deep-learning convolutional neural network (CNN) models to recognise sea ice textures in S1 SAR images and assign sea ice radar freeboard estimates acquired by CS2. This approach transfers information acquired by CS2 to S1, which we call Cryo2S1. A Cryo2S1 dataset is curated, containing several thousand collocated S1 SAR images and along-track CS2 measurements during 2020-2021.

The objectives of this research are to develop robust methods for segmentation of multitemporal synthetic aperture radar (SAR) and optical data and to investigate the fusion of multitemporal ENVISAT advanced synthetic aperture radar (ASAR) and Chinese HJ-1B multispectral data for detailed urban land-cover mapping. Eight-date multiangle ENVISAT ASAR images and one-date HJ-1B charge-coupled device image acquired over Beijing in 2009 are selected for this research. The edge-aware region growing and merging (EARGM) algorithm is developed for segmentation of SAR and optical data. Edge detection using a Sobel filter is applied on SAR and optical data individually, and a majority voting approach is used to integrate all edge images. The edges are then used in a segmentation process to ensure that segments do not grow over edges. The segmentation is influenced by minimum and maximum segment sizes as well as the two homogeneity criteria, namely, a measure of color and a measure of texture. The classification is performed using support vector machines. The results show that our EARGM algorithm produces better segmentation than eCognition, particularly for built-up classes and linear features. The best classification result (80%) is achieved using the fusion of eight-date ENVISAT ASAR and HJ-1B data. This represents 5%, 11%, and 14% improvements over eCognition, HJ-1B, and ASAR classifications, respectively. The second best classification is achieved using fusion of four-date ENVISAT ASAR and HJ-1B data (78%). The result indicates that fewer multitemporal SAR images can achieve similar classification accuracy if multitemporal multiangle dual-look-direction SAR data are carefully selected.

The aim of this article was to investigate the potential of polarimetric decomposition of Chinese Gaofen-3 (GF-3) C-band fully polarimetric synthetic aperture radar (PolSAR) data for Arctic sea ice classification during summer season. Five different polarimetric decomposition approaches, including the Cloude-Pottier decomposition (Cloude), the Freeman three-component decomposition (Freeman3), the Freeman three-component decomposition using the extended Bragg model (Freeman3X), the Yamaguchi three-component decomposition (Yamaguchi3), and the nonnegative eigenvalue decomposition (NNED) were analyzed using 35 scenes of GF-3 PolSAR data collected over the Fram Strait, Arctic from June 14–18, 2017. Polarimetric features extracted from these five methods were evaluated and utilized to train random forest classifiers to classify open water (calm water and rough water) and sea ice types (melted ice, unmelted ice, and deformed ice). The results show that NNED could ensure physically valid decomposed powers while the other three model-based decompositions had negative values. In terms of sea ice classification, NNED had the highest feature importance scores and achieved an overall accuracy and Kappa coefficient of about 86.18% and 0.82, respectively. Inclusion of radar incidence angle as a feature in the classifier could slightly improve the classification accuracy by about 3%. The influence of incidence angle on sea ice classification accuracy was also investigated and it was found that high incidence angles (39°–46°) were superior to low incidence angles (21°–27°) due to the overall higher accuracies.

In recent years, there has been an increased interest in using synthetic aperture radar (SAR) to detect and monitor sea ice in the Bohai Sea for protecting offshore exploration and supporting marine transport. Two important tasks are the classification of sea ice and the determination of sea ice thickness, which can be achieved by considering the specific scattering mechanisms of the different ice types. This paper describes a three-component scattering model to decompose polarimetric SAR (PolSAR) data of sea ice. The total backscatter is modeled as the incoherent summation of surface, double-bounce, volume, and residual components. The proposed model extends the volume scattering contribution of sea ice by considering transmission, extinction, and refraction effects. The model is validated using C-band Radarsat-2 quad-polarization data acquired over sea ice in the Bohai Sea. The results show that the proposed polarimetric decomposition approach helps to distinguish different ice types and offers a proxy for sea ice thickness.

The science plan for the Alaska SAR Facility (ASF) focuses on earth surface characteristics that are of interest within the overall concept of global change and that show significant regional, seasonal and interannual variations resulting in changes in the strength of their radar returns. The polar oceans, with the continuous motion and deformation of the pack ice and the changes in the surface state of the surrounding open seas, offer excellent opportunities for such research. Because such studies require both frequent and detailed analysis of Synthetic Aperture Radar (SAR) data, a Geophysical Processor System (GPS) has been developed to speed the extraction of useful geophysical information from SAR data sets. The system will initially produce three main types of products: (a) sets of ice motion vectors obtained by automated computer tracking of identifiable ice floes on sequential images, (b) the areal extent and location of several different ice types and open water and (c) a characterization of the wave state in ice-free regions as well as within the ice in the marginal ice zone at locations where significant wave penetration occurs. Details of these analysis procedures are described. Initially the GPS is planned to process 10 image pairs/day for ice motion, 20 images/day for ice type variations and 1 image/day for wave information, with a total estimated processing time of 13 hours. A variety of projects plan to utilize the SAR data stream in studies of ice, lead and polynya dynamics and thermodynamics. A common feature of these research programs will be attempts to provide, via the coupling of the SAR data with ice property and ice dynamics models, improved estimates of the heat and mass fluxes into both the atmosphere and the ocean as affected by the characteristics of the ice cover.Key words: SAR, radar, sea ice, image analysis, remote sensing

One of the problems associated with synthetic aperture radar (SAR) data analysis is the presence of random noise or speckle SAR data, being achromatic in nature, which offers very limited scope for the detection and delineation of various terrain features. ERS-1 SAR data for the coastal region of West Bengal, India were processed (a) to suppress the random noise using various filters, (b) to generate the intensity, hue and saturation (IHS) transform from temporal SAR data, and (c) to study the synergism of SAR data with optical sensor data. The results indicate that the Gamma MSP filter with a 5 5 pixel kernel size has been the most efficient in suppressing the noise and concurrently improving the image contrast. The IHS transform of temporal SAR data made it easier to discriminate between various wetland categories. This was also the case with hybrid image generated by the Indian Remote Sensing Satellite (IRS-1B) Linear Imaging and Self-scanning Sensor (LISS-II) data when compared to SAR data alone.

SAR and optical remote sensing: Assessment of complementarity and interoperability in the context of a large-scale operational forest monitoring system

Polar sea ice is one of the Earth’s climate components that has been significantly affected by the recent trend of global warming. While the sea ice area in the Arctic has been decreasing at a rate of about 4 % per decade, the multi-year ice (MYI), also called perennial ice, is decreasing at a faster rate of 10 %–15 % per decade. On the other hand, the sea ice area in the Antarctic region was slowly increasing at a rate of about 1.5 % per decade until 2014 and since then it has fluctuated without a clear trend. However, no data about ice type areas are available from that region, particularly of MYI. Due to differences in physical and crystalline structural properties of sea ice and snow between the two polar regions, it has become difficult to identify ice types in the Antarctic. Until recently, no method has existed to monitor the distribution and temporal development of Antarctic ice types, particularly MYI throughout the freezing season and on decadal time scales. In this study, we have adapted a method for retrieving Arctic sea ice types and partial concentrations using microwave satellite observations to fit the Antarctic sea ice conditions. The first circumpolar, long-term time series of Antarctic sea ice types; MYI, first-year ice and young ice is being established, so far covering years 2013–2019. Qualitative comparison with synthetic aperture radar data, with charts of the development stage of the sea ice, and with Antarctic polynya distribution data show that the retrieved ice types, in particular the MYI, are reasonable. Although there are still some shortcomings, the new retrieval for the first time allows insight into the evolution and dynamics of Antarctic sea ice types. The current time series can in principle be extended backwards to start in the year 2002 and can be continued with current and future sensors.

Sea ice disasters are already one of the most serious marine disasters in the Bohai Sea region of our country, which have seriously affected the coastal economic development and residents’ lives. Sea ice classification is an important part of sea ice detection. Hyperspectral imagery and multispectral imagery contain rich spectral information and spatial information and provide important data support for sea ice classification. At present, most sea ice classification methods mainly focus on shallow learning based on spectral features, and the good performance of the deep learning method in remote sensing image classification provides a new idea for sea ice classification. However, the level of deep learning is limited due to the influence of input size in sea ice image classification, and the deep features in the image cannot be fully mined, which affects the further improvement of sea ice classification accuracy. Therefore, this paper proposes an image classification method based on multilevel feature fusion using residual network. First, the PCA method is used to extract the first principal component of the original image, and the residual network is used to deepen the number of network layers. The FPN, PAN, and SPP modules increase the mining between layer and layer features and merge the features between different layers to further improve the accuracy of sea ice classification. In order to verify the effectiveness of the method in this paper, sea ice classification experiments were performed on the hyperspectral image of Bohai Bay in 2008 and the multispectral image of Bohai Bay in 2020. The experimental results show that compared with the algorithm with fewer layers of deep learning network, the method proposed in this paper utilizes the idea of residual network to deepen the number of network layers and carries out multilevel feature fusion through FPN, PAN, and SPP modules, which effectively solves the problem of insufficient deep feature extraction and obtains better classification performance.

Method for image classification considering Probability Density Function (PDF) based on simplified beta distributions is proposed. In this paper, image classification for Synthetic Aperture Radar (SAR) data is concerned. In particular, Probability Density Function (PDF) of SAR data is followed by not multivariate normal distribution but Chi-Square like distribution. It, however, is not always true that the PDF of SAR data is followed by Chi-Square distribution. Due to the mismatch between Chi-Square distribution and actual distribution, classification performance gets worth. In this paper, simplified beta distribution is assumed for the PDF of the SAR data. Furthermore, it is used to add texture information to the SAR data when the Maximum Likelihood classification is applied. In the paper, “Contrast” of texture feature is added to the SAR data. Through the experiments with real SAR data, it is found that matching error between real PDF and the proposed simplified beta distribution is smaller than the normal distribution. It is also found that applying the proposed distribution-adaptive maximum likelihood method using the simplified beta-distribution could achieve a classification accuracy improvement of 94.7% and 12.1%.

Fresh water is a scarce resource in the West-African Sahel region, seasonally influenced by droughts and floods. Particularly in terms of climate change, the importance of wetlands increases for flora, fauna, human population, agriculture, livestock and fishery. Hence, access to open water is a key factor. Long-term monitoring of water dynamics is of great importance, especially with regard to the spatio-temporal extend of wetlands and drylands. It can predict future trends and facilitate the development of adequate management strategies. Lake Tabalak, a Ramsar wetland of international importance, is one of the most significant ponds in Niger and a refuge for waterbirds. Nevertheless, human population growth increased the pressure on this ecosystem, which is now degrading for all uses. The main objective of the study is a long-term monitoring of the Lake Tabalak’s water dynamics to delineate permanent and seasonal water bodies, using weather- and daytime-independent multi-sensor and multi-temporal Synthetic Aperture Radar (SAR) data available for the study area. Data of the following sensors from 1993 until 2016 are used: Sentinel-1A, TerraSARX, ALOS PALSAR-1/2, Envisat ASAR, RADARSAT-1/2, and ERS-1/2. All SAR data are processed with the Multi-SAR-System, unifying the different characteristics of all above mentioned sensors in terms of geometric, radiometric and polarimetric resolution to a consistent format. The polarimetric representation in Kennaugh elements allows fusing single-polarized data acquired by older sensors with multi-polarized data acquired by current sensors. The TANH-normalization guarantees a consistent and therefore comparable description in a closed data range in terms of radiometry. The geometric aspect is solved by projecting all images to an earth-fixed coordinate system correcting the brightness by the help of the incidence angle. The elevation model used in the geocoding step is the novel global model produced by the TanDEM-X satellite mission. The advantage of the Multi-SAR-System is that it comprises ortho-rectification, radiometric enhancement, normalization and Kennaugh decomposition, independent from sensors, modes, polarizations or acquisition date of SAR data. In addition, optical satellite data can be included as well, to fill gaps where SAR data are missing due to the special normalization scheme. This kind of pre-processing is exclusively implemented at the Earth Observation Center of the German Aerospace Center in Oberpfaffenhofen, Germany. Therefore, the dynamic change of the open water of the Lake Tabalak could be classified over dry and rainy seasons and years, using different SAR data. The study provides a unique database and contributes to a better understanding of wetland systems in the Sahel region influenced by human pressure and climate change.

Despite recent research on the potential of dual- (DP) and full-polarimetry (FP) Synthetic Aperture Radar (SAR) data for crop mapping, the capability of compact polarimetry (CP) SAR data has not yet been thoroughly investigated. This is of particular concern, given the availability of such data from RADARSAT Constellation Mission (RCM) shortly. Previous studies have illustrated potential for accurate crop mapping using DP and FP SAR features, yet what contribution each feature makes to the model accuracy is not well investigated. Accordingly, this study examined the potential of the early- to mid-season (i.e., May to July) RADARSAT-2 SAR images for crop mapping in an agricultural region in Manitoba, Canada. Various classification scenarios were defined based on the extracted features from FP SAR data, as well as simulated DP and CP SAR data at two different noise floors. Both overall and individual class accuracies were compared for multi-temporal, multi-polarization SAR data using the pixel- and object-based random forest (RF) classification schemes. The late July C-band SAR observation was the most useful data for crop mapping, but the accuracy of single-date image classification was insufficient. Polarimetric decomposition features extracted from CP and FP SAR data produced relatively equal or slightly better classification accuracies compared to the SAR backscattering intensity features. The RF variable importance analysis revealed features that were sensitive to depolarization due to the volume scattering are the most important FP and CP SAR data. Synergistic use of all features resulted in a marginal improvement in overall classification accuracies, given that several extracted features were highly correlated. A reduction of highly correlated features based on integrating the Spearman correlation coefficient and the RF variable importance analyses boosted the accuracy of crop classification. In particular, overall accuracies of 88.23%, 82.12%, and 77.35% were achieved using the optimized features of FP, CP, and DP SAR data, respectively, using the object-based RF algorithm.

We exploit along-track offset estimation techniques applied to Optical and Synthetic Aperture Radar (SAR) data to constrain the co-seismic slip induced by the M w 7.5 earthquake occurred on September, 2018 in Sulawesi island, Indonesia. Such seismic event nucleated along the left-lateral NNW-SSE trending Palu-Koro strike-slip fault and strongly affected Palu Bay causing several effects such as tsunamis and soil liquefactions. Preliminary results show a deformation of more 1.5m along an approximately NS direction both on Optical and SAR data. Moreover, both techniques clearly highlight the surface rupture induced by the earthquake. Further analysis will be focused on accurate comparison and validation of the retrieved results.

Sea ice seriously endangers the safe navigation of ships at sea. Therefore, it is the key to accurately judge the type of sea ice around the ship. However, remote sensing technology will be restricted by external weather changes. Therefore, this paper presents a method based on statistical feature extraction and machine learning of ship main engine vibration signal, so as to realize the recognition of sea ice types. Firstly, according to the sea ice video information captured by the camera, the acceleration and vibration data collected by the vibration sensor are divided, and the wavelet threshold denoising algorithm is adopted to enhance the features. Then, based on the statistical characteristics of vibration signals, time domain and frequency domain features are extracted. Finally, Random Forest(RF) and Gradient Boosting Decision Tree(GBDT) machine learning algorithms are used to classify the extracted eigenvalues. The experimental results show that the accuracy of sea ice identification using ship main engine vibration signals is 79%. This method provides a reference for identifying sea ice types by ship ice-induced vibration, and is helpful for further research on sea ice classification.