Automated tracking of ice floes: a stochastic approach

A method for tracking ice floes in sequential satellite imagery is presented. The approach, based on probability distributions, directly incorporates the spatial information about feature locations into the estimation procedure. Given an image taken at time t₀, a probability model is used to determine how features in the image will appear at time t₁, and the probability distribution is used to identify features common to both images. The use of a probability model provides a means to measure the goodness of fit of the resulting matches. The features used are the outlines of sea ice floes observed in SAR (synthetic aperture radar) images, although the method can be applied to any set of features. The floe outlines are found using an erosion-propagation algorithm which combines erosion from mathematical morphology with local propagation of information about floe edges     

Least average residual algorithm (LARA) for tracking the motion ofArctic sea ice

Suppose we have two images ℐ and 𝒥 of a collection of ice floes taken at times t₀ and t₁, respectively. Given an ice floe I in ℐ, the authors develop a new automated tracking algorithm to track the position of I in image 𝒥. The proposed least average residual algorithm (LARA) performs a constrained search for matching ice floes by determining an appropriate search space for each floe to be tracked. LARA takes a stepwise approach with suitable decision rules at various stages. It takes into account various distinguishing characteristics such as: i) the geometry of the ice field; ii) size of an ice floe; and iii) geometry of the floe. LARA also attempts to detect broken floes and amalgamated floes     

SAR图像的自动分割方法研究

由于存在相干斑噪声的影响,给SAR图像分割造成很大的困难,该文提出了一种SAR图像的自动分割方法。首先在特征提取阶段,通过计算小波能量提取纹理信息,用邻域统计量提取灰度信息,用保边缘平均灰度提取边缘信息,以确保边缘准确。然后提出一种改进的完全无监督的聚类算法进行图像分割,该算法可以自动确定分割的类型数目。由于该方法充分考虑了SAR图像的纹理、灰度和边缘信息,因而极大地提高了其最终分割性能。实验结果证明了该方法的有效性。     

Sea Ice Tracking by Nested Correlations

Spatial differences in sea ice displacement affect ice stress, ice production, and the mass balance of the ice cover. Our concepts about the spatial structure of this field have been undernourished because of a paucity of data with high spatial detail and because of the tedium of extracting such measurements from images manually. A method is described that measures displacements from synthetic aperture radar digital imagery with fine spatial resolution, and does so fully automatically. Many small areas of ice common to two images are identified by correlating the two images. The strategy is to acquire a crude displacement field first from highly averaged images, and to refine this field with images of successively higher resolution. The median discrepancy between automatically and manually measured displacements is three pixels (0.075 km). The algorithm operates successfully on compact ice with large floes and modest rotation rates; we believe it will prove applicable to most of the arctic ice cover throughout the year.     

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.      

Sea ice monitoring by L-band SAR: an assessment based on literature and comparisons of JERS-1 and ERS-1 imagery

Spaceborne single-polarization C-band synthetic aperture radar (SAR) imagery is widely used to gather information about the state of the sea ice cover in the polar regions. C-band is regarded as a reasonable choice for all-season monitoring capabilities. For specific mapping tasks, however, other frequency bands can be more suitable. In the first part of this paper, the summary of a literature study dealing with the utilization of L-band SAR imagery for sea ice monitoring is presented. Investigations reveal that if deformation features such as ice ridges, rubble fields, and brash ice are to be mapped, L-band radar is superior in a number of cases. The second part of this paper addresses the comparison of JERS-1 and ERS-1 SAR images that were acquired over sea ice east of Svalbard and along the east coast of Greenland. The effects of the different frequencies, polarizations, and incidence angles of the two SAR systems are discussed. It is demonstrated that the images of both sensors complement one another in the analysis of ice conditions, resulting in a more detailed view of the sea ice cover state.     

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.     

基于区域MRF和贝叶斯置信传播的SAR图像分割

 本文通过定义新的势函数,将贝叶斯置信传播算法和区域MRF模型有效结合,提出了一种SAR图像分割算法.考虑到SAR图像丰富的纹理信息,该算法对分水岭分割后的过分割区域提取纹理特征,在得到的区域邻接图上构建MRF模型,并加入区域灰度均值和方差作为区域特征,利用FCM聚类的初分割结果定义区域的关联势函数,并将区域特征引入到置信传播算法中,定义了新的交互势函数.该算法充分利用了SAR图像空间的背景信息,所定义的新的交互势函数能在促进分割结果区域一致性的同时较好保护边缘.实验结果表明,相对于其他MRF模型分割算法,本文算法能取得更好的分割效果.     

Application of neural networks for sea ice classification inpolarimetric SAR images

Several automatic methods have been developed to classify sea ice types from fully polarimetric synthetic aperture radar (SAR) images, and these techniques are generally grouped into supervised and unsupervised approaches. In previous work, supervised methods have been shown to yield higher accuracy than unsupervised techniques, but suffer from the need for human interaction to determine classes and training regions. In contrast, unsupervised methods determine classes automatically, but generally show limited ability to accurately divide terrain into natural classes. In this paper, a new classification technique is applied to determine sea ice types in polarimetric and multifrequency SAR images, utilizing an unsupervised neural network to provide automatic classification, and employing an iterative algorithm to improve the performance. The learning vector quantization (LVQ) is first applied to the unsupervised classification of SAR images, and the results are compared with those of a conventional technique, the migrating means method. Results show that LVQ outperforms the migrating means method, but performance is still poor. An iterative algorithm is then applied where the SAR image is reclassified using the maximum likelihood (ML) classifier. It is shown that this algorithm converges, and significantly improves classification accuracy. The new algorithm successfully identifies first-year and multiyear sea ice regions in the images at three frequencies. The results show that L- and P-band images have similar characteristics, while the C-band image is substantially different. Classification based on single features is also carried out using LVQ and the iterative ML method. It is found that the fully polarimetric classification provides a higher accuracy than those based on a single feature. The significance of multilook classification is demonstrated by comparing the results obtained using four-look and single-look classifications     

Analysis of glaciers and geomorphology on Svalbard usingmultitemporal ERS-1 SAR images

The all-weather capability makes synthetic aperture radar (SAR) interesting with respect to glaciological studies in remote Arctic areas. The aim of this project was to investigate how the European Remote Sensing Satellite (ERS) SAR backscatter from certain glaciological and geomorphological structures varied with time and find out at which time of the year a SAR acquisition will give the best result when observing certain features. Five ERS-1 SAR images, one from the winter and four from the summer, were acquired over the northwestern part of Svalbard, Norway. Ground measurements and observations were made at the same time as one of the summer SAR acquisitions. The ground data as well as meteorological recordings were used to analyze the SAR backscatter changes in the multitemporal data set. A zonation on the glaciers representing snow, firn/superimposed ice, and glacier ice was detected in the winter image only. The equilibrium line could also be derived from this SAR image. Information on drainage patterns, distribution of wet and dry snow areas, and the occurrence of crevasses was derived from the summer images. Changes in snow cover distribution was identified in the summer images. The form and position of ice-cored moraines, fossil beach ridges, and river channels were identified in all images. Early summer images showed the highest potential for identification of landforms. Small-scale landforms, such as patterned ground and tundra polygons, could not be identified     

基于ICA和FCM的多时相SAR图像变化检测

针对多时相合成孔径雷达(Synthetic Aperture Radar,SAR)图像提出了一种非监督的变化检测方法,首先利用独立成分分析(Independent Component Analysis,ICA)抑制相干斑噪声和减少混合像元,提高图像的分类性能.然后对独立分量图像构造差值图像进行非监督的变化检测,最后采用加入空间邻域信息的模糊C-均值聚类(Fuzzy C-Means)方法把差异图像分为变化类和非变化类,以此克服需要选择统计模型的影响.实验结果证实了该算法能有效抑制了变化检测中的虚警.     

基于CFAR级联的SAR图像舰船目标检测算法

SAR图像舰船目标检测在军事监视和海洋环境监管等方面有着重要的意义。针对SAR图像的特点,提出了一种基于全局CFAR检测与局部CFAR检测级联的舰船目标检测算法。在全局CFAR检测中,通过海杂波特性拟合优选海杂波统计模型,以较高的虚警率筛选潜在的目标点;在局部CFAR检测中,以潜在目标点的连通区域为单位,通过检测窗口的选取、背景像素的确定和海杂波拟合等步骤以后,以较低的虚警率确定目标。最后,通过条件扩张算法和目标像素聚类完善船只细节。实验结果表明,文中算法在保证良好的检测性能的同时,具有检测效率高、舰船细节完整等优点,为舰船目标鉴别和信息提取提供了良好的保障,更加符合实际应用需求。     

一种基于像素差值编码度量不确定SAR图像相似性的新方法

在SAR图像目标识别、图像匹配等应用中,定义合理可靠的相似度尤为重要。该文在分析现有SAR图像相似度基础上,提出一种基于像素差值编码的相似度准则。首先将SAR图像按照相邻像素灰度差异生成编码图像,然后以编码图像之间的一致性作为相似度。该文从理论上证明了该相似度对SAR图像中相干斑噪声、部分遮挡和模糊等因素的鲁棒性和适应性,还讨论了将该准则应用于SAR图像匹配时,如何针对不确定性,给出一定置信水平下所有合理的匹配位置。理论和实验结果表明该文提出的相似度准则对SAR图像上相干斑噪声、部分遮挡以及模糊不敏感,能有效应用于不确定SAR图像的匹配。     

基于稀疏表示和道路辅助的单幅SAR图像运动目标检测方法

本文提出一种利用单幅SAR(Synthetic Aperture Radar)图像实现运动目标检测的方法.首先提出一种基于压缩感知的SAR图像道路检测算法:根据SAR图像中道路的特点,使用模糊C均值方法将图像进行模糊分类,获得大致的道路区域,然后利用Hough变换域的稀疏性,用压缩感知精确定位图像中的道路信息.其次利用图像稀疏表示的方法对运动目标进行检测:不同速度运动目标的散焦量和距离单元跨越不同,由此生成样本图像,继而构造超完备字典.将待测图像分块,并计算子图像在字典下的稀疏系数,检测并匹配出运动目标的速度参数.最后,结合已检测出的道路辅助信息,消除多普勒模糊影响,剔除虚假的运动目标,并对运动目标速度参数进行校正.实验结果证明了所提方法的有效性.     

Automated Matching of Pairs of SIR-B Images for Elevation Mapping

During the SIR-B mission in October 1984, a significant number of overlapping synthetic aperture radar (SAR) images of various ground areas was collected. This has offered the first opportunity to perform stereo analyses on images from space that cover large ground areas to determine elevation information. This paper presents the preliminary results of an investigation to obtain elevation data from stereo pairs of SIR-B images. First, the accuracy with which elevation information can be derived from SIR-B image pairs is evaluated theoretically. It is shown that elevation accuracy is a function of the slant range resolution, the incidence angles with which the stereo pair is obtained, the accuracies in spacecraft state estimation, and determination of corresponding pixels in the stereo pair. Next, a hierarchical method is developed to match the corresponding pixels. This method involves iterative removal of local distortions and correlations of pairs of local neighborhoods in the two images. Since it is necessary to perform the matching at every pixel in the image, it is very computationally intensive. Therefore, it has been implemented on the Massively Parallel Processor (MPP) at the Goddard Space Flight Center (GSFC). The MPP's speed permits two iterations of this technique to operate on a pair of 512 × 512 images within 7 s. Results of applying this algorithm to SIR-B images of Mount Shasta, CA, are shown. The matching algorithm performs well in regions of the image with significant features.     

Methods for digital analysis of AVHRR sea ice images

Algorithms for obtaining information on sea ice conditions from Advanced Very High Resolution Radiometer (AVHRR) digital imagery are presented. These algorithms comprise three steps: discrimination of cloud-covered areas from ice-covered and open water areas, estimation of ice concentration, and measurement of the sizes of ice floes and open water areas (polynyas) within the ice field. For ice-water-cloud discrimination, a dynamic threshold technique is applied to six combinations of the five spectral channels of the AVHRR. This results in an image that is useful for ice analyses. To derive ice concentration estimates, a simple linear relationship between pixel intensity values and amount of ice cover is applied to the visible channel reflectance data. The method for obtaining size frequency distributions is based on a size transform of binary realizations of the ice concentration result and is equally applicable to measuring polynyas and ice floes     

一种海洋涡旋SAR图像仿真方法

海洋涡旋对海洋热循环起着关键作用,是海洋科学研究中的一个重要分支。合成孔径雷达(Synthetic Aperture Radar, SAR)为海洋涡旋的观测和研究提供了大量的图像数据,但是涡旋在SAR成像时会受到各种海洋环境因素的影响,难以解译涡旋SAR图像特征。仿真SAR图像可以用于研究涡旋的特征,但是目前极少有关于涡旋SAR图像仿真方法的研究。为了更好地解译SAR图像中的涡旋特征,该文提出了一种涡旋SAR图像仿真方法。首先,基于流体力学中典型的Burgers-Rott涡旋模型,建立涡旋2维表面流场;然后,利用SAR海洋成像仿真模型,仿真给定涡旋2维流场、海面风场以及雷达系统参数下的涡旋SAR图像。该文针对气旋式涡旋与反气旋式涡旋进行了仿真实验,并建立了仿真涡旋SAR图像的相似度评价标准。实验结果表明,仿真的涡旋SAR图像与真实星载涡旋SAR图像能够较好地吻合,验证了方法的有效性。      

免疫克隆SAR图像分割算法

由于存在相干斑噪声的影响,使得常规的图像分割技术应用于SAR图像时,效果往往较差。该文提出一种新人工免疫系统SAR图像分割算法,算法首先提取每幅图像的纹理特征,并对其进行实数编码,然后通过借鉴生物学免疫系统的抗体克隆选择机理,构造适合于图像分割的克隆算子,以较快的收敛速度优化目标函数。实验结果表明,新算法是一种有效的SAR 图像区域分割方法。     

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