观测数据采样化的遥感影像非监督分割

为了实现影像的自动化分割,提出一种利用非监督方式将观测数据采样化的遥感影像分割方法。该方法利用欧氏空间的概率分布建模采样数据和观测数据,并将其映射到黎曼空间,通过不断将观测数据转换为采样数据的方式实现影像的自动采样化。每次采样过程只需计算观测数据点到采样点的测地线距离,将距采样点测地线距离最小的观测数据转化为采样数据,以保证采样数据不断趋于该类数据的真实分割结果,同时使算法能够有效分割具有不同像素数的类别。将算法应用于模拟影像和真实遥感影像分割,对其分割结果以及传统基于统计、基于模糊的非监督算法和基于神经网络的监督算法相应分割结果定性定量的对比分析验证了该算法的有效性及可行性。     

深度融合网结合条件随机场的遥感图像语义分割

为了充分利用遥感图像中丰富的细节信息和上下文信息,提高图像语义分割精度,提出一种深度融合网结合条件随机场模型的遥感图像语义分割方法。方法在全卷积神经网络框架中增加反卷积融合结构,搭建深度融合DFN (Deep Fusion Networks)网络,通过深层网络自动获取多尺度特征,避免人工设计和选择特征,提高模型的泛化能力;同时借助反卷积融合结构,利用多尺度信息,将浅层细节信息和深层语义信息相融合,提高模型的处理精度。由全连接条件随机场引入空间上下文信息,更好地定位边界,得到最终的语义分割结果。在遥感图像数据集上的实验结果显示:(1)随着不同尺度细节信息的融入,结果的边缘轮廓越精确、接近标签图像;(2)增加了空间上下文信息后,语义分割结果边缘更细化、准确,精度更高。实验表明,该方法可以有效提高遥感图像语义分割的精度,改善结果的过平滑现象。     

Earth observation data based rapid flood-extent modelling for tsunami-devastated coastal areas

Earth observation (EO)-based mapping and analysis of natural hazards plays a critical role in various aspects of post-disaster aid management. Spatial very high-resolution Earth observation data provide important information for managing post-tsunami activities on devastated land and monitoring re-cultivation and reconstruction. The automatic and fast use of high-resolution EO data for rapid mapping is, however, complicated by high spectral variability in densely populated urban areas and unpredictable textural and spectral land-surface changes. The present paper presents the results of the SENDAI project, which developed an automatic post-tsunami flood-extent modelling concept using RapidEye multispectral satellite data and ASTER Global Digital Elevation Model Version 2 (GDEM V2) data of the eastern coast of Japan (captured after the Tohoku earthquake). In this paper, the authors developed both a bathtub-modelling approach and a cost-distance approach, and integrated the roughness parameters of different land-use types to increase the accuracy of flood-extent modelling. Overall, the accuracy of the developed models reached 87–92%, depending on the analysed test site. The flood-modelling approach was explained and results were compared with published approaches. We came to the conclusion that the cost-factor-based approach reaches accuracy comparable to published results from hydrological modelling. However the proposed cost-factor approach is based on a much simpler dataset, which is available globally.     

Combining spatiotemporal fusion and object-based image analysis for improving wetland mapping in complex and heterogeneous urban landscapes

Remote sensing has been proven promising in wetland mapping. However, conventional methods in a complex and heterogeneous urban landscape usually use mono temporal Landsat TM/ETM + images, which have great uncertainty due to the spectral similarity of different land covers, and pixel-based classifications may not meet the accuracy requirement. This paper proposes an approach that combines spatiotemporal fusion and object-based image analysis, using the spatial and temporal adaptive reflectance fusion model to generate a time series of Landsat 8 OLI images on critical dates of sedge swamp and paddy rice, and the time series of MODIS NDVI to calculate phenological parameters for identifying wetlands with an object-based method. The results of a case study indicate that different types of wetlands can be successfully identified, with 92.38%. The overall accuracy and 0.85 Kappa coefficient, and 85% and 90% for the user’s accuracies of sedge swamp and paddy respectively.     

Satellite image analysis using crowdsourcing data for collaborative mapping: current and opportunities

ABSTRACT

Researchers are continually finding new applications of satellite images because of the growing number of high-resolution images with wide spatial coverage. However, the cost of these images is sometimes high, and their temporal resolution is relatively coarse. Crowdsourcing is an increasingly common source of data that takes advantage of local stakeholder knowledge and that provides a higher frequency of data. The complementarity of these two data sources suggests there is great potential for mutually beneficial integration. Unfortunately, there are still important gaps in crowdsourced satellite image analysis by means of crowdsourcing in areas such as land cover classification and emergency management. In this paper, we summarize recent efforts, and discuss the challenges and prospects of satellite image analysis for geospatial applications using crowdsourcing. Crowdsourcing can be used to improve satellite image analysis and satellite images can be used to organize crowdsourced efforts for collaborative mapping.     

Data fusion and classifier ensemble techniques for vegetation mapping in the coastal Everglades

This study examined the applicability of data fusion and classifier ensemble techniques for vegetation mapping in the coastal Everglades. A framework was designed to combine these two techniques. In the framework, 20-m hyperspectral imagery collected from Airborne Visible/Infrared Imaging Spectrometer was first merged with 1-m Digital Orthophoto Quarter Quads using a proposed pixel/feature-level fusion strategy. The fused data set was then classified with an ensemble approach based on two contemporary machine learning algorithms: Random Forest and Support Vector Machine. The framework was applied to classify nine vegetation types in a portion of the coastal Everglades. An object-based vegetation map was produced with an overall accuracy of 90% and Kappa value of 0.86. Per-class classification accuracy varied from 61% for identifying buttonwood forest to 100% for identifying red mangrove scrub. The result shows that the framework is promising for automated vegetation mapping in the Everglades.     

A new approach for surface water change detection: Integration of pixel level image fusion and image classification techniques

Normally, to detect surface water changes, water features are extracted individually using multi-temporal satellite data, and then analyzed and compared to detect their changes. This study introduced a new approach for surface water change detection, which is based on integration of pixel level image fusion and image classification techniques. The proposed approach has the advantages of producing a pansharpened multispectral image, simultaneously highlighting the changed areas, as well as providing a high accuracy result. In doing so, various fusion techniques including Modified IHS, High Pass Filter, Gram Schmidt, and Wavelet-PC were investigated to merge the multi-temporal Landsat ETM+ 2000 and TM 2010 images to highlight the changes. The suitability of the resulting fused images for change detection was evaluated using edge detection, visual interpretation, and quantitative analysis methods. Subsequently, artificial neural network (ANN), support vector machine (SVM), and maximum likelihood (ML) classification techniques were applied to extract and map the highlighted changes. Furthermore, the applicability of the proposed approach for surface water change detection was evaluated in comparison with some common change detection methods including image differencing, principal components analysis, and post classification comparison. The results indicate that Lake Urmia lost about one third of its surface area in the period 2000–2010. The results illustrate the effectiveness of the proposed approach, especially Gram Schmidt-ANN and Gram Schmidt-SVM for surface water change detection.     

面向复杂山地的机载LiDAR与无人机影像融合应用

本文针对LiDAR点云与无人机影像数据特征的优缺点,利用LiDAR点云与无人机DOM影像融合,将影像数据光谱信息赋给LiDAR点云数据,使其不仅具备精准的空问结构信息,还能得到清晰的纹理信息.为验证融合数据应用的可行性与数据提取的准确性,对融合前后的点云数据进行地面点提取与DEM构建.试验表明:将无人机影像的光谱信息赋...     

Fusion of TerraSAR-x and Landsat ETM+ data for protected area mapping in Uganda

TerraSAR-X satellite acquires very high spatial resolution data with potential for detailed land cover mapping. A known problem with synthetic aperture radar (SAR) data is the lack of spectral information. Fusion of SAR and multispectral data provides opportunities for better image interpretation and information extraction. The aim of this study was to investigate the fusion between TerraSAR-X and Landsat ETM+ for protected area mapping using high pass filtering (HPF), principal component analysis with band substitution (PCA) and principal component with wavelet transform (WPCA). A total of thirteen land cover classes were identified for classification using a non-parametric C 4.5 decision tree classifier. Overall classification accuracies of 74.99%, 83.12% and 85.38% and kappa indices of 0.7220, 0.8100 and 0.8369 were obtained for HPF, PCA and WPCA fusion approaches respectively. These results indicate a high potential for a combined use of TerraSAR-X and Landsat ETM+ data for protected area mapping in Uganda.     

Land-use and vegetation-cover mapping of an indigenous land area in the state of Mato Grosso (Brazil) based on spectral linear mixing model,segmentation and region classification

A spectral linear-mixing model using Landsat ETM+ imagery was undertaken to estimate fraction images of green vegetation, soil and shade in an indigenous land area in the state of Mato Grosso in the central-western region of Brazil. The fraction images were used to classify different types of land use and vegetation cover. The fraction images were classified by the following two methods: (a) application of a segmentation based on the region-growing technique; and (b) grouping of the regions segmented using the per-region unsupervised classifier named ISOSEG. Adopting a 75% threshold, ISOSEG generated 44 clusters that were grouped into eight land-use and vegetation-cover classes. The mapping achieved an average accuracy of 83%, showing that the methodology is efficient in mapping areas of great land-use and vegetation-cover diversity, such as that found in the Brazilian cerrado (savanna).     

多源遥感影像融合在哈大齐土地利用分类中的应用

随着遥感技术的发展,同一区域的多源遥感影像数据越来越丰富。以哈大齐为例,利用ETM+和SPOT-5数据探讨不同遥感信息融合在土地利用过程中的处理方法,比较不同融合算法在土地分类中的差异,并进行定性和定量比较。为有关部门进行土地规划、管理提供科学依据有着十分重要的意义。     

小波的SAR和光学图像融合方法比较研究

本文通过光学图像和SAR图像的融合,对几种典型的小波融合方法进行了分析对比实验,并使用各种评价标准对它们的融合效果给出了评价。在此基础上,对光学图像和SAR图像融合效果的各种评价标准做出了讨论并提出自己的看法。     

面向林地信息提取的 Pleiades-1影像融合算法评价分析

以Pleiades-1影像为对象,以探究其面向林地信息提取的融合算法选择为视角,通过基于目视和定量特征分析、辅助面向对象分类分析的方法,为林业部门遥感影像大规模融合应用提供参考。研究认为Pansharp和GramSchmidt算法融合结果目视效果良好,各波段与原多光谱相关系数均高达0.8以上,清晰度与纹理增强明显。两种融合算法影像在不同林地层次信息提取能力各有优势,Pansharp融合结果在林地层次分类总精度可达86.55%,Gram-Schmidt融合结果则在森林类型层次具有最高的分类总精度78.76%。具体融合算法的选取需根据其应用的信息提取层次而定。     

Images difference of ASAR data for rice crop mapping in Fuzhou,China

This paper introduces ENVISAT ASAR data application on rice field mapping in the Fuzhou area, using multi-temporal ASAR dual polarization data acquired in 2005. The procedure for ASAR data processing here includes data calibration, image registration, speckle reduction and conversion of data format from amplitude to dB for backscatter. The backscatter of rice increases with the rice growing stages, which was much different from other land covers. Based on image difference techniques, 6 schemes were designed with ASAR different temporal and polarization data for rice field mapping. Difference images between images in the early period of rice crop and growing or ripening period, are more suitable for rice extraction than those difference images between different polarizations in the same date. The most accurate result of late rice extraction was achieved based on the difference of HH polarization data acquired in October and August. Therefore, for rice field mapping, the temporal information is more important than polarization information. The data during the early growing season of rice is very important for high accuracy rice mapping.     

GIS-based geomorphological mapping using remote sensing data and supplementary geoinformation: A case study of the Dhaka city area,Bangladesh

This study aims to prepare a detailed GIS-based geomorphological map accompanied with landfill sites of Dhaka city area which can be used for multipurpose functionality. Attainment of the geomorphological map is based upon interpretation of the oldest available aerial photographs (1:40,000) and contemporary topographic maps (1:8000) which reflect almost pre-urban ground of Dhaka. Randomly distributed 160 boreholes have been used to prepare representative soil profiles (RSP) to identify the near-surface lithology of the geomorphological units. The study reveals that 13 out of 18 low-lying geomorphic units, comprising 65% of the total area demand landfill practices for urban development. Landfill sites have been merged with urban growth on each low-lying geomorphic unit using a spatially enhanced fused image of IRS-1D PAN and ETM+ bands 5, 4 and 3, acquired February 2000 and 2002, respectively. We found that 43% area of the total low-lying geomorphic units experience fill practices so far. The fill sites have been differentiated into four classes based on their relative thickness. Integration of fill classes with geomorphological map shows the urban dynamics of Dhaka city area till 2002. Due to GIS integration, this map can be rapidly updated to demonstrate temporal modifications in urban ground. It can be used effectively in different geomorphological hazard mapping and urban land-use practices.     

监控视频与地理信息融合技术发展综述

监控视频在智慧城市、平安城市的安防、管理中发挥着重要作用。虽然视频数据具有实时、直观、丰富的特点,但与地理信息的耦合分析程度不高,无法有效利用视频本身的空间信息。同时,地理信息实时性差、更新慢,无法反映动态变化的场景。为了达到更好的效果,国内外很多学者、行业人员综合两者的优势,对监控视频与地理信息的融合开展研究。本文介绍了监控视频与地理信息融合的关键技术,分析了两者融合的研究现状及应用领域,最后结合遇到的瓶颈和困难,指出了今后可能的发展趋势。     

A new conception of image texture and remote sensing image segmentation based on Markov random field

The texture analysis is often discussed in image processing domain, but most methods are limited within gray-level image or color image, and the present conception of texture is defined mainly based on gray-level image of single band. One of the essential characters of remote sensing image is multidimensional or even high-dimensional, and the traditional texture conception cannot contain enough information for these. Therefore, it is necessary to pursuit a proper texture definition based on remote sensing i...     

基于航片颜色分割的土地利用更新调查技术

本文对基于航片的土地利用更新调查作业流程进行了分析,对其中的地物提取、地图纠正技术进行了探讨,提出了基于神经元网络和启发式学习规则的彩色地图分层的方法,用以提取彩色地形图中的特征地物以及进行颜色分割。还构造了地图自动纠正算法,它自动准确地提取图内的格网,以进行纠正。这些算法提高了土地利用更新作业效率。     

Identification and classification of geographically isolated wetlands in North Alabama using geographic object based image analysis (GeOBIA)

Due to recent Supreme Court rulings, there has been an increased interest in the isolated wetlands of the United States. These types of wetlands serve vital ecological roles such as water quality regulation and as a habitat of biological diversity. This study focuses specifically on mapping of geographically isolated wetlands, or those that are separated from traditional wetlands by a given spatial extent, using Geographic Object-Based Image Analysis (GeOBIA). GeOBIA is a type of remote sensing analysis that identifies objects and features in data-sets via automated methodologies. This type of analysis offers the opportunity to increase the efficiency of what has traditionally been a very labour intensive process of manual photo-interpretation. This analysis resulted in the delineation of 26,424 areas as geographically isolated wetlands. These results were assessed for accuracy through both manual inspection of aerial imagery and field verification which yielded accuracies of 83.7 and 87.7%, respectively.     

基于分离式金字塔模型的测绘立体影像数据检索系统设计与实现

针对立体影像数据查询检索的现状,找出其在测绘应用中的不足。首先搭建测绘影像数据检索系统的总体框架,其次提出对测绘立体影像数据进行基于分离式金字塔模型的瓦片处理并且对其创建混合网格索引(HGI),最后根据其不同的检索功能运用不同的检索算法,基于B/S结构,运用ArcSDE、Oracle相关组件、Tomcat相关网络技术,采用VS.NET2005(C#)、AE等开发实现多功能的测绘立体影像数据检索系统。