Fusion of multisensor multitemporal satellite data for land cover mapping

Efficient integration of remote sensing information with different temporal, spectral and spatial resolutions is important for accurate land cover mapping. A new temporal fusion classification (TFC) model is presented for land cover classification, based on statistical fusion of multitemporal satellite images. In the proposed model, the temporal dependence of multitemporal images is taken into account by estimating transition probabilities from the change pattern of a vegetation dynamics indicator (VDI). Extension of this model is applicable to Synthetic Aperture Radar (SAR) images and integration of multisensor multitemporal satellite images, concerning both temporal attributes and reliability of multiple data sources. The feasibility of the new method is verified using multitemporal Landsat Thematic Mapper (TM) and ERS SAR satellite images, and experimental results show improved performance over conventional methods.     

Extraction of bridges over water from IKONOS panchromatic data

Compared to remote sensing images of medium or low spatial resolution, high‐resolution remote sensing images can provide observation data containing more detailed information for georesearch. Accordingly, an important issue for current computer and geoscience experts is to develop useful methods or technology to extract information from these high‐resolution satellite images. As part of a series of research into object extraction, this paper focuses mainly on the extraction of bridges over water from high‐resolution panchromatic satellite images. Since bridges over water are obviously adjacent to water in remote sensing images, this paper proposes a practical knowledge‐based bridge extraction method for remote sensing images of high spatial resolution. The steps involved are: water extraction based on Gauss Markov Random Field (GMRF)‐Support Vector Machine (SVM) classification methods which use a SVM to classify the image based on textural features expressed by a GMRF; image thinning and removal of fragmented lines; main trunk detection by width; vectorization; and feature expression. Finally, tests are described for two pieces of panchromatic IKONOS satellite images with a 1 m resolution. The experimental results show that the proposed method is suitable for images with a single‐peak histogram (contrast between water and land is sharp) or a multi‐peak histogram (greyscale value of water is close to that of land).     

Information extraction from remotely sensed data.

Digital image processing is now widely available for users of remotely sensed data. Although such processing offers many new opportunities for the user (or analyst) it also makes heavy demands on the acquisition of new skills, if the data are to yield useful information efficiently. In deciding on the best approach for image classification the user faces a bewildering array of choices, many of which have been poorly evaluated. It is clear, however, that the use of both internal and external contextual information can be of great value in improving classification performance. The ultimate use of information extracted from remote sensing data is strongly affected by its compatability with other geographic data planes. Problems in achieving such compatibility in the framework of automated geographical information systems are discussed. The success of image analysis and classification methods is highly dependent on the relationships between the abilities of sensing systems themselves and the character of the phenomena being studied. This is illustrated by reference to the capabilities of future high resolution satellite systems.     

Applying soft computing for remote sensing data composite algorithms

Remote sensing of the earth’s surface using satellite-mounted sensor data is one of the most important methods for global environmental monitoring today. However, when using satellite sensor data, clouds in the atmosphere can interfere with the remote sensing, and specific land points may not be correctly monitored on any given day. In order to overcome this problem, a common alternative is to use multiple day composite data. Multiple day composite data use several consecutive days’ remote sensing data, and choose the most accurate data within the temporal dataset for the same land point. This allows the creation of a more complete dataset by patching together data which have had no cloud interference during a specified time period in order to create a clearer, more usable dataset. In this article, we propose the application of soft computing, namely fuzzy logic, in order to select the clearest data in the temporal interval to use for the composite data. Moderate resolution remote sensing data of areas in Japan were used for the evaluation, and the results were compared with previous composite methods.     

Feature selection and land cover classification of a MODIS-like data set for a semiarid environment

The advent of the Earth Observing System (EOS), and the Moderateresolution Imaging Spectroradiometer (MODIS) in particular, will usher in a new era of global remote sensing by providing very large data volumes for interpretation and processing. Since many data streams will contain correlated data, feature selection is an important practical problem for such activities as classification of global land cover based on spectral, temporal, spatial and directional data. Treebased classification methods offer a suite of promising approaches to extraction of meaningful features from large measurement spaces. This research develops a tree-based model that performs feature selection on a satellite database containing information on land covers in a semiarid region in Cochise County, Arizona. In addition, we test the abilities of several classifiers to correctly label land cover using this reduced set of inputs under various sampling schemes. Results from this analysis indicate that decision trees can reduce a high-dimension dataset to a manageable set of inputs that retain most of the information of the original database, while remaining largely insensitive to choice of sampling strategy, and that Fuzzy ARTMAP, a type of artificial neural network classifier, achieves highest accuracy in comparison to maximum-likelihood or decision-tree classifiers.     

容差粗糙集卫星遥感数据分类

结合粗糙集理论和遥感数据中地物光谱特征空间分布信息,提出了一种基于光谱特征邻域的容差粗糙集分类方法,用来处理卫星遥感数据分类中的不确定性问题。利用北京地区Landsat-5 TM数据进行分类试验,对算法分类过程进行讨论及其分类结果进行验证分析;结果表明：文中方法在可理解性和稳定性上体现出比较好的性质,能够有效处理卫星遥感数据分类中存在的不确定性因素,在具有复杂光谱特征地物分类方面具有发展潜力。     

MODIS图象的云检测及分析

云一直是遥感图象处理、图象分析的一大障碍.为了解决这一问题,试图探讨利用中分辨率成像光谱仪MODIS检测云的方法,该方法充分考虑到MODIS数据具有36个光谱通道,特别是红外波段细分的特点,先是基于云的波谱特性采用多光谱综合法、红外差值法及指数法来对MODIS图象上的云点进行检测,鉴于这些方法有一定的局限性,因而还运用了一种基于空间结构分析和神经网络的云自动检测算法;最后将各种方法的云检测结果进行相互映证和对照分析,结果表明,这些方法检测到的云互相吻合,说明利用MODIS图象可成功地检测云点像元.这不仅为云的去除奠定了良好基础,而且也可以提高图象识别、图象分类及图象反演的精度.     

Incorporating nocturnal UAV side-view images with VIIRS data for accurate population estimation: a test at the urban administrative district scale

ABSTRACT

In contrast to daytime remote sensing used for observing the Earth, night-time light remote sensing with satellites primarily assesses human activity using urban parameters such as building lights or lighted highways to help determine population density and other habitation characteristics. One limitation to conventional night-time remote sensing is that light emitted from high-rise buildings, for example, is not easily detected because of optical geometry as satellite sensors are generally pointed in only a downward direction. Furthermore, satellite sensors often receive weak optical signals because of streetlights reflected from the Earth’s surface. As a result, accurate information on night-time human activity cannot be gathered from existing satellite remote-sensing methods. To address this, a new method for night-time remote setting is presented. Specifically, an unmanned aerial vehicle (UAV) is used to capture panoramic images of night-time light and processed to reveal side-view light spot information from urban buildings. This dataset was used to predict population density alone, and with the Visible Infrared Imaging Radiometer Suite (VIIRS) data by simple multiple linear regression. The results confirm that nocturnal UAV side-view data or VIIRS data alone can be used to estimate population density, while the combination of the two significantly increases the accuracy of population density estimation compared against estimating population density using nocturnal UAV side-view data or VIIRS data alone. This outcome suggests that multi-angular night-time remote-sensing data sources increase the accuracy of urban population density estimation. One reason for this may be that the side-view night-time data and orthophoto data infer urban population density from different agent variables: building occupancy is a proxy of side-view night-time data, while density of illuminated road network is that of orthophoto data.     

Elevation modelling from satellite visible and infrared (VIR) data

Since the early emergence of Earth observation satellites, researchers have investigated different methods of extracting three-dimensional information using satellite data. Apart from a few early stereo images by hand-held photographs acquired during the Gemini and Apollo missions, the first experiments to extract three-dimensional data using stereo viewing from space began with the Earth Terrain Camera flown onboard SkyLab in 1973/74. Since this time, various analogue or digital sensors in the visible spectrum have flown to provide researchers and geoscientists with spatial data to extract and interpret three-dimensional information of the Earth's surface. Although clinometry techniques can be applied with the optical sensor images, stereo viewing of images was and still is the most common method used by the mapping, photogrammetry and remote sensing communities for elevation modelling. The paper will review clinometry and stereoscopy and their applicability to the different satellite sensors (space photographs and scanners). Their performances to extract absolute or relative elevation from various research and commercial organizations are addressed. The respective advantages, difficulties and constraints of the sensors are discussed, as well as the methods and the technologies used for extracting elevation data in an operational context.     

Land-cover classification of partly missing data using support vector machines

Land-cover classification based on multi-temporal satellite images for scenarios where parts of the data are missing due to, for example, clouds, snow or sensor failure has received little attention in the remote-sensing literature. The goal of this article is to introduce support vector machine (SVM) methods capable of handling missing data in land-cover classification. The novelty of this article consists of combining the powerful SVM regularization framework with a recent statistical theory of missing data, resulting in a new method where an SVM is trained for each missing data pattern, and a given incomplete test vector is classified by selecting the corresponding SVM model. The SVM classifiers are evaluated on Landsat Enhanced Thematic Mapper Plus (ETM?+?) images covering a scene of Norwegian mountain vegetation. The results show that the proposed SVM-based classifier improves the classification accuracy by 5–10% compared with single image classification. The proposed SVM classifier also outperforms recent non-parametric k-nearest neighbours (k-NN) and Parzen window density-based classifiers for incomplete data by about 3%. Moreover, since the resulting SVM classifier may easily be implemented using existing SVM libraries, we consider the new method to be an attractive choice for classification of incomplete data in remote sensing.     

Analytical algorithms for lake water TSM estimation for retrospective analyses of TM and SPOT sensor data

Suspended matter in inland waters is related to total primary production and fluxes of heavy metals and micropollutants such as PCBs. Synoptic information on suspended matter cannot be obtained from an in situ monitoring network since suspended matter is a spatially inhomogeneous parameter. This problem can be solved by the integrated use of remote sensing data, in situ data and water quality models. To enable retrospective model and remote sensing data comparison of suspended matter concentration and distribution, a methodology is required for processing satellite images that is independent of in situ measurements. Analytical optical modelling, based on knowledge of the in situ inherent optical properties, leads to reliable multi-temporal algorithms for estimating suspended matter concentration in lakes for the data from the SPOT and Landsat TM sensors. This methodology allows multi-temporal, multi-site and multi-instrument comparison of TSM maps derived from satellite imagery. This means that satellite sensor data can now become an independent measurement tool for water management authorities. The remote sensing maps showed that large gradients in TSM were observed for the various lakes as well as temporal changes of these spatial gradients. In situ point samples are shown to be not representative for suspended matter in the lakes.     

Image classification of a northern peatland complex using spectral and plant community data

Ordination and cluster analysis are two common methods used by plant ecologists to organize species abundance data into discrete “associations”. When applied together, they offer useful information about the relationships among species and the ecological processes occurring within a community. Remote sensing provides surrogate data for characterizing the spatial distribution of ecological classes based on the assumption of characteristic reflectance of species and species associations. Currently, there exists a need to establish and clarify the link between theories and practices of classification by ecologists and remote sensing scientists. In this study, high spatial resolution Compact Airborne Spectrographic Imager (CASI) reflectance data were examined and compared to plant community data for a peatland complex in northern Manitoba, Canada. The goal of this research was to explore the relationship between classification of species cover and community data and reflectance values. Ordination and cluster analysis techniques were used in conjunction with spectral separability measures to organize clusters of community-based data that were suitable for classification of CASI reflectance data, while still maintaining their ecological significance. Results demonstrated that two-way indicator species analysis (TWINSPAN) clusters did not correspond well to spectral reflectance and gave the lowest classification results of the methods investigated. The highest classification accuracies were achieved with ecological classes defined by combining the information obtained from a suite of analysis techniques (i.e., TWINSPAN, correspondence analysis (CA), and signature separability analysis), albeit not statistically superior to the classification obtained from the signature separability analysis alone.     

Measurement of uncertainty by the entropy: application to the classification of MSS data

Uncertainty is imposed simultaneously with multispectral data acquisition in remote sensing. It grows and propagates in processing, transmitting and classification processes. This uncertainty affects the extracted information quality. Usually, the classification performance is evaluated by criteria such as the accuracy and reliability. These criteria can not show the exact quality and certainty of the classification results. Unlike the correctness, no special criterion has been propounded for evaluation of the certainty and uncertainty of the classification results. Some criteria such as RMSE, which are used for this purpose, are sensitive to error variations instead of uncertainty variations. This study proposes the entropy, as a special criterion for visualizing and evaluating the uncertainty of the results. This paper follows the uncertainty problem in multispectral data classification process. In addition to entropy, several uncertainty criteria are introduced and applied in order to evaluate the classification performance.     

基于ZY3数据的南水北调西线生态遥感调查研究

国产ZY3高分辨率卫星数据在国土资源管理方面已经有较广泛的应用,通过对ZY3数据的处理和信息自动提取,可获取南水北调西线工程沿线的土地利用及植被群落分布状况。通过本次遥感解译流程,克服卫星时相、区域地貌差异等问题,形成适用于ZY3卫星遥感影像生态遥感解译的流程,与传统的生态遥感解译方法相比,不仅成本低,而且工期短,可为大区域性生态遥感解译提供依据。     

星载数据加密容错研究

在空间辐射环境中,单粒子反转效应(SEU)会导致星载系统存储器逻辑位发生翻转,且无法单纯依赖硬件措施完全消除,又由于卫星通信加密设备在大多数加密模式下具有错误扩散特性,星载数据加密设备的SEU软故障会导致批量数据不可用.针对星载数据加密过程的SEU影响问题,设计了基于奇偶校验码的星载数据加密过程检错算法和基于海明码的星载数据加密过程纠错算法,该容错方案可以有效降低SEU对星载数据加密过程的影响,提高星载数据加密的可靠性.通过大量图像数据仿真实验结果表明,提出的容错方案对星载数据加密过程可靠性的提高率与位出错概率成反比,有较好的空间适应性.     

Automatic land cover analysis for Tenerife by supervised classification using remotely sensed data

Automatic land cover classification from satellite images is an important topic in many remote sensing applications. In this paper, we consider three different statistical approaches to tackle this problem: two of them, namely the well-known maximum likelihood classification (ML) and the support vector machine (SVM), are noncontextual methods. The third one, iterated conditional modes (ICM), exploits spatial context by using a Markov random field. We apply these methods to Landsat 5 Thematic Mapper (TM) data from Tenerife, the largest of the Canary Islands. Due to the size and the strong relief of the island, ground truth data could be collected only sparsely by examination of test areas for previously defined land cover classes.We show that after application of an unsupervised clustering method to identify subclasses, all classification algorithms give satisfactory results (with statistical overall accuracy of about 90%) if the model parameters are selected appropriately. Although being superior to ML theoretically, both SVM and ICM have to be used carefully: ICM is able to improve ML, but when applied for too many iterations, spatially small sample areas are smoothed away, leading to statistically slightly worse classification results. SVM yields better statistical results than ML, but when investigated visually, the classification result is not completely satisfying. This is due to the fact that no a priori information on the frequency of occurrence of a class was used in this context, which helps ML to limit the unlikely classes.     

基于频带特征融合的GL-CNN遥感图像场景分类

高分辨率卫星遥感图像场景信息的分类对影像分析和解译具有重要意义,传统的高分辨卫星遥感图像场景分类方法主要依赖于人工提取的中、低层特征且不能很好的利用图像丰富的场景信息,针对这一问题,提出一种基于频带特征融合与GL-CNN(Guided Learning Convolutional Neural Network,指导学习卷积神经网络)的分类方法。首先通过NSWT（Non-Subsampled Wavelet Transform,非下采样小波变换）提取出图像的高低频子带,将高频子带进行频带特征融合得到融合高频子带,然后联合频谱角向能量分布曲线的平稳区间分析实现融合高频子带与低频子带的样本融合,最后指导卷积神经网络自动提取图像的高低频子带包含的高层特征来实现场景分类。通过对UCM_LandUse 21类数据进行试验表明,本文方法的分类正确率达到94.52%,相比以往算法有显著提高。     

基于频带特征融合的GL-CNN遥感图像场景分类

高分辨率卫星遥感图像场景信息的分类对影像分析和解译具有重要意义，传统的高分辨卫星遥感图像场景分类方法主要依赖于人工提取的中、低层特征且不能很好的利用图像丰富的场景信息，针对这一问题，提出一种基于频带特征融合与GL-CNN(Guided Learning Convolutional Neural Network，指导学习卷积神经网络)的分类方法。首先通过NSWT（Non-Subsampled Wavelet Transform，非下采样小波变换）提取出图像的高低频子带，将高频子带进行频带特征融合得到融合高频子带，然后联合频谱角向能量分布曲线的平稳区间分析实现融合高频子带与低频子带的样本融合，最后指导卷积神经网络自动提取图像的高低频子带包含的高层特征来实现场景分类。通过对UCM_LandUse 21类数据进行试验表明，本文方法的分类正确率达到94.52%，相比以往算法有显著提高。     

高性能大气校正算法中遥感数据切分策略研究

高分辨率卫星遥感数据在地物识别等方面具有明显优势,然而其定量化应用中需要精确的大气校正,该过程通常相当耗时。分别研究了大气校正算法串行处理方法及基于通用计算机集群系统的并行处理过程。通过对2012年7月我国华北地区的环境卫星CCD数据进行大气校正,并分析了串、并行过程各个步骤运行时间,表明了对大气校正并行处理的高可行性。针对并行过程中负载不均衡和通讯频繁等问题,设计了基于卫星像元特征的数据切分策略,并对不同并行算法进行了性能分析,表明了本文反演结果的可靠性,以及提出的切分策略能达到更高的加速比。