Multisensor approach to automated classification of sea ice image data

A multisensor data fusion algorithm based on a multilayer neural network is presented for sea ice classification in the winter period. The algorithm uses European Remote Sensing (ERS), RADARSAT synthetic aperture radar (SAR), and low-resolution television camera images and image texture features. Based on a set of in situ observations made at the Kara Sea, a neural network is trained, and its structure is optimized using a pruning method. The performance of the algorithm with different combinations of input features (sensors) is assessed and compared with the performance of a linear discriminant analysis (LDA)-based algorithm. We show that for both algorithms a substantial improvement can be gained by fusion of the three different types of data (91.2% for the neural network) as compared with single-source ERS (66.0%) and RADARSAT (70.7%) SAR image classification. Incorporation of texture increases classification accuracy. This positive effect of texture becomes weaker with increasing number of sensors (from 8.4 to 6.4 percent points for the use of two and three sensors, respectively). In view of the short training time and smaller number of adjustable parameters, this result suggests that semiparametric classification methods can be considered as a good alternative to the neural networks and traditional parametric statistical classifiers applied for the sea ice classification.     

Statistical and structural approaches to texture

In this survey we review the image processing literature on the various approaches and models investigators have used for texture. These include statistical approaches of autocorrelation function, optical transforms, digital transforms, textural edgeness, structural element, gray tone cooccurrence, run lengths, and autoregressive models. We discuss and generalize some structural approaches to texture based on more complex primitives than gray tone. We conclude with some structural-statistical generalizations which apply the statistical techniques to the structural primitives.     

Estimating Gaussian Markov random field parameters in anonstationary framework: application to remote sensing imaging

In this paper, we tackle the problem of estimating textural parameters. We do not consider the problem of texture synthesis, but the problem of extracting textural features for tasks such as image segmentation. We take into account nonstationarities occurring in the local mean. We focus on Gaussian Markov random fields for which two estimation methods are proposed, and applied in a nonstationary framework. The first one consists of extracting conditional probabilities and performing a least square approximation. This method is applied to a nonstationary framework, dealing with the piecewise constant local mean. This framework is adapted to practical tasks when discriminating several textures on a single image. The blurring effect affecting edges between two different textures is thus reduced. The second proposed method is based on renormalization theory. Statistics involved only concern variances of Gaussian laws, leading to Cramer-Rao estimators. This method is thus especially robust with respect to the size of sampling. Moreover, nonstationarities of the local mean do not affect results. We then demonstrate that the estimated parameters allow texture discrimination for remote sensing data. The first proposed estimation method is applied to extract urban areas from SPOT images. Since discontinuities of the local mean are taken into account, we obtain an accurate urban areas delineation. Finally, we apply the renormalization based on method to segment ice in polar regions from AVHRR data.     

Wavelet-Based SAR Image Despeckling and Information Extraction, Using Particle Filter

This paper proposes a new-wavelet-based synthetic aperture radar (SAR) image despeckling algorithm using the sequential Monte Carlo method. A model-based Bayesian approach is proposed. This paper presents two methods for SAR image despeckling. The first method, called WGGPF, models a prior with Generalized Gaussian (GG) probability density function (pdf) and the second method, called WGMPF, models prior with a Generalized Gaussian Markov random field (GGMRF). The likelihood pdf is modeled using a Gaussian pdf. The GGMRF model is used because it enables texture parameter estimation. The prior is modeled using GG pdf, when texture parameters are not needed. A particle filter is used for drawing particles from the prior for different shape parameters of GG pdf. When the GGMRF prior is used, the particles are drawn from prior in order to estimate noise-free wavelet coefficients and for those coefficients the texture parameter is changed in order to obtain the best textural parameters. The texture parameters are changed for a predefined set of shape parameters of GGMRF. The particles with the highest weights represents the final noise-free estimate with corresponding textural parameters. The despeckling algorithms are compared with the state-of-the-art methods using synthetic and real SAR data. The experimental results show that the proposed despeckling algorithms efficiently remove noise and proposed methods are comparable with the state-of-the-art methods regarding objective measurements. The proposed WGMPF preserves textures of the real, high-resolution SAR images well.     

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.     

SAR image statistics related to atmospheric drag over sea ice

The possibility of using synthetic aperture radar (SAR) data to distinguish sea-ice regions with different atmospheric drag is explored. Both the amplitude of the radar return and statistics derived from SAR image data are examined. Roughness statistics data from several pack-ice areas are used in a backscatter model to predict the return from surfaces with measured drag coefficients. The results suggest that the scattering coefficient for typical radar wavelengths is insensitive to the roughness elements responsible for the observed drag coefficient variations over pack ice free of major ridges. For marginal ice zones, where ice concentration and floe deformation contribute to atmospheric drag, a simple model for the atmospheric boundary layer is used to provide qualitative relationships between drag coefficient and regional ice properties (ice concentration, floe size distribution, floe edge density) derivable from SAR data. Simple algorithms to produce maps of ice concentration and edge density are outlined and applied to 23.5-cm SAR digital image data     

Sea ice classification using SAR backscatter statistics

Sea ice classification accuracy using standard statistics and higher order texture statistics generated from grey-level co-occurrence (GLC) matrices were compared for synthetic aperture radar (SAR) data collected during the Marginal Ice Zone Experiment (MIZEX) in April 1987. Standard stepwise discriminate analysis was used to identify the statistics modes useful for discrimination. Range was the most effective statistic, correctly classifying the ice types 75% of the time. Overall, the standard statistics (mean, variance, range, etc.) outperformed the texture statistics (87% accuracy vs. 75% accuracy). Given the added difficulty and computational cost of generating texture statistics, this result suggests that standard statistics should be used for sea ice classification. Odden and multiyear ice categories were the most difficult to statistically separate for these data     

Efficient Texture Analysis of SAR Imagery

We address the problem of efficiency in texture analysis for synthetic aperture radar (SAR) imagery. Motivated by the statistical occupancy model, we introduce the notion of patch reoccurrences. Using the reoccurrences, we propose the use of approximate textural features in analysis of SAR images. We describe how the proposed approximate features can be extracted for two popular texture analysis methods—the gray-level cooccurrence matrix and Gabor wavelets. Results on image texture classification show that the proposed method can provide an improved efficiency in the analysis of SAR imagery, without introducing any significant degradation in the classification results.     

Surface roughness characterizations of sea ice and ice sheets: case studies with MISR data

This work is an examination of potential uses of multiangular remote sensing imagery for mapping and characterizing sea ice and ice sheet surfaces based on surface roughness properties. We use data from the Multi-angle Imaging SpectroRadiometer (MISR) to demonstrate that ice sheet and sea ice surfaces have characteristic angular signatures and that these angular signatures may be used in much the same way as spectral signatures are used in multispectral classification. Three case studies are examined: sea ice in the Beaufort Sea off the north coast of Alaska, the Jakobshavn Glacier on the western edge of the Greenland ice sheet, and a region in Antarctica south of McMurdo station containing glaciers and blue-ice areas. The MISR sea ice image appears to delineate different first-year ice types and, to some extent, the transition from first-year to multiyear ice. The MISR image shows good agreement with sea ice types that are evident in concurrent synthetic aperture radar (SAR) imagery and ice analysis charts from the National Ice Center. Over the Jakobshavn Glacier, surface roughness data from airborne laser altimeter transects correlate well with MISR-derived estimates of surface roughness. In Antarctica, ablation-related blue-ice areas, which are difficult to distinguish from bare ice exposed by crevasses, are easily detected using multiangular data.     

一种基于集成学习和特征融合的遥感影像分类新方法

针对多源遥感数据分类的需要,提出了一种基于全极化SAR影像、极化相干矩阵特征、光学遥感影像光谱和纹理的多种特征融合和多分类器集成的遥感影像分类新方法.对全极化PALSAR数据进行预处理和极化相干矩阵特征提取,利用灰度共生矩阵计算光学和SAR影像的对比度、逆差距、二阶距、差异性等纹理特征参数,并与光谱特征结合,形成6种组合策略.利用集成学习方法对随机森林分类器、子空间分类器、最小距离分类器、支持向量机分类器、反向传播神经网络分类器等分类器进行组合,对不同组合策略的遥感影像特征集进行分类.结果表明提出的基于多种特征和多分类器集成的新方法很好地利用了主被动遥感数据在不同地表景观类型提取上的潜力,综合了多种算法的优势,能够有效地提高总体精度和各类别的分类精度.     

Automated delineation of dry and melt snow zones in Antarctica using active and passive microwave observations from space

This paper presents the algorithms and analysis results for delineating snow zones using active and passive microwave satellite remote sensing data. With a high-resolution Radarsat synthetic aperture radar (SAR) image mosaic, dry snow zones, percolation zones, wet snow zones, and blue ice patches for the Antarctic continent have been successfully delineated. A competing region growing and merging algorithm is used to initially segment the SAR images into a series of homogeneous regions. Based on the backscatter characteristics and texture property, these image regions are classified into different snow zones. The higher level of knowledge about the areal size of and adjacency relationship between snow zones is incorporated into the algorithms to correct classification errors caused by the SAR image noise and relief-induced radiometric distortions. Mathematical morphology operations and a line-tracing algorithm are designed to extract a vector line representation of snow-zone boundaries. With the daily passive microwave Special Sensor Microwave/Imager (SSM/I) data, dry and melt snow zones were derived using a multiscale wavelet-transform-based method. The analysis results respectively derived from Radarsat SAR and SSM/I data were compared and correlated. The complementary nature and comparative advantages of frequently repeated passive microwave data and spatially detailed radar imagery for detecting and characterizing snow zones were demonstrated.     

基于能量分解的影像纹理多尺度分析

按照地维Ｃａｂｏｒ函数的特点和视觉机制，提出了用来捕捉纹理基元的纹理检测器函数。基于纹理检测器和扩展的小波变换，提出了基于能量分解的影像纹理多尺度分析方法，并按照神经动力学的测抑制和端点抑制理论，实现了对多尺度纹理特征的融合，这一多尺度分析方法直接将影像纹理能量在时间－尺度空间分解，包含了相位信息，避免了基于线性变换多尺度分解引起的能量与相位分离，为纹理分析提供了一个层次性的框架，有效提高了纹理的     

A SAR for Real-Time Ice Reconnaissance

Prior research has shown that open water first-year ice and multi-year ice can be distinguished on an image obtained by a Synthetic-aperture radar (SAR). A small lightweight SAR, STAR-1, has been built that employs this capability for support of engineering operations for oil exploration in the Arctic. The SAR is a fully focused side-looking system capable of mapping either side of the aircraft. It has two modes, wide swath, which covers 45 km, and high resolution, which covers 22 km. The azimuth resolution is 6 m for both modes, with a range resolution of 12 m in the wide swath and 6 meters in the high resolution mode. The instrument is installed in a Cessna 441 Conquest capable of flying at high altitudes with minimum fuel consumption. An image is produced in real time by a digital image formation processing system aboard the aircraft. This image is transmitted to a ground station via a data link where a hardcopy is formed on heat-developed film.     

Radar Image Processing for Rock-Type Discrimination

Radar images have unique radiometric and geometric characteristics which present unique problems and opportunities for geological application. This paper reviews preprocessing and analytical techniques found useful or promising for applications of radar images to geologic problems such as rock-type discrimination. The use of coherent monochromatic illumination in radar images results in image speckle noise which interferes with characterization of the imaged surface. Median value filtering of the radar images removes speckle with minimal edge effects and resolution degradation. Variations in radar scene illumination due to uncompensated sensor platform motions or antenna pattern effects can be somewhat corrected for by mean and variance equalization in a direction perpendicular to the resulting image gradient. Registration of radar images to a map base and compensation of terrain induced image distortion can be accomplished by registration to digital elevation models and knowledge of imaging geometry. Analysis of SEASAT images with coregistered LANDSAT images indicates that the radar data can make a significant contribution to rock-type discrimination, especially if textural measures are incorporated. The sensitivity of radar backscatter to local slopes makes radar images an excellent medium from which to extract textural measures. Three techniques for extraction of the textural data inherent in the radar images are presented. Computation of image tone variance over various areas can numerically encode image texture. Hue-saturation-intensity split spectrum processing displays low-frequency variations in color while preserving high-frequency detail.     

Sea Ice Deformation State From Synthetic Aperture Radar Imagery—Part II: Effects of Spatial Resolution and Noise Level

C- and L-band airborne synthetic aperture radar (SAR) imagery acquired at like- and cross-polarizations over sea ice under winter conditions is examined with the objective to study the discrimination between level ice and ice deformation features. High-resolution low-noise data were analyzed in the first paper. In this second paper, the main topics are the effects of spatial resolution and signal-to-noise ratio. Airborne high-resolution SAR scenes are used to generate a sequence of images with increasingly coarser spatial resolution from 5 to 25 m, keeping the number of looks constant. The signal-to-noise ratio is varied between typical noise levels for airborne imagery and satellite data. Areal fraction of deformed ice and average deformation distance are determined for each image product. At L-band, the retrieved values of the areal fraction get larger as the image resolution is degraded. The areal fraction at C-band remains constant. The retrieved average distance between deformation features increases both at C- and L-bands as the image resolution gets coarser. The influence of noise becomes noticeable if its level is equal or larger than the average intensity backscattered from the level ice. The retrieval of deformation parameters using simulated images that resemble ERS-2 SAR, Envisat ASAR, and ALOS PALSAR data products is discussed. Basic differences between real and simulated ERS-2 SAR images are analyzed.      

一种SAR与可见光图像压缩感知融合增强方法

针对机载多传感器成像战场态势感知的问题，提出了一种合成孔径雷达(Synthetic Aperture Radar，SAR)与可见光图像压缩感知融合增强方法。该方法首先对SAR与可见光图像分别进行压缩感知测量，得到压缩测量值，然后通过基于局部权值的融合方法实现对压缩测量值的融合，再利用有序度最优分割法提取SAR图像的强散射目标，最后对融合测量值重建得到初步融合图像，初步融合图像通过目标对比度增强得到最终融合图像。对多组图像进行了仿真分析，视觉及数值结果表明该方法能显著增强融合图像的目标对比度，提升了图像纹理清晰度，较大程度降低了图像融合过程中的数据计算量。     

小卫星分布式SAR系统的信噪比研究

小卫星分布式合成孔径雷达(SAR)系统可以提高成像分辨率,检测地面运动目标和进行数字高程测量。但由于小卫星接收天线面积的减小使其输入信噪比相对于主星大大降低,这将严重影响系统的性能。该文以JERS-1为主星,分析了小卫星分布式SAR系统的信噪比以及信噪比降低对成像、测高、检测动目标等的影响。分析表明此系统的信噪比将不能满足系统对信噪比的最低需求。最后讨论了用相干或非相干处理来改善系统输出信噪比的方法。     

一种基于多重分形的SAR图像边缘检测方法

分形维数只能刻画那些具有理想的自相似性的分形体,现实中的许多纹理并不满足这一条件,因此单一的分形维数并不足以描述和刻画SAR图像的纹理,多重分形维数更适合于描述图像的纹理.通过计算原始SAR图像离散点数据的奇异性指数,然后对应每一点奇异性指数计算全局多重分形奇异谱,根据判决准则区分边缘和纹理可以实现SAR图像的边缘检测,实验结果表明,基于多重分形特征的边缘检测算法能够检测到许多局部细节,同时又避免出现不重要的细节,突出了主要的边缘信息,很好地区分出SAR图像的纹理和边缘.     

Information-theoretic heterogeneity measurement for SAR imagery

A heterogeneity feature, calculable from synthetic aperture radar (SAR) images on a per-pixel basis, but relying on global image statistics, is defined and discussed. Starting from the multiplicative speckle and texture models relating the amount of texture and speckle to the local mean and variance at every pixel, such a feature is rigorously derived from Shannon's information theory as the conditional information of local standard deviation to local mean. Thanks to robust statistical estimation, it is very little sensitive to the noise affecting SAR data, and thus capable of capturing subtle variations of texture whenever they are embedded in a heavy speckle. Experimental results carried out on two SAR images with different degrees of noisiness demonstrate that the proposed feature is likely to be useful for a variety of automated segmentation and classification tasks.