共查询到20条相似文献，搜索用时 15 毫秒
1.
ABSTRACTTerrain feature detection is a fundamental task in terrain analysis and landscape scene interpretation. Discovering where a specific feature (i.e. sand dune, crater, etc.) is located and how it evolves over time is essential for understanding landform processes and their impacts on the environment, ecosystem, and human population. Traditional induction-based approaches are challenged by their inefficiency for generalizing diverse and complex terrain features as well as their performance for scalable processing of the massive geospatial data available. This paper presents a new deep learning (DL) approach to support automatic detection of terrain features from remotely sensed images. The novelty of this work lies in: (1) a terrain feature database containing 12,000 remotely sensed images (1,000 original images and 11,000 derived images from data augmentation) that supports data-driven model training and new discovery; (2) a DL-based object detection network empowered by ensemble learning and deep and deeper convolutional neural networks to achieve high-accuracy object detection; and (3) fine-tuning the model’s characteristics and behaviors to identify the best combination of hyperparameters and other network factors. The introduction of DL into geospatial applications is expected to contribute significantly to intelligent terrain analysis, landscape scene interpretation, and the maturation of spatial data science. 相似文献
2.
Automatic extraction of lineament information from satellite images using digital elevation data 总被引：1，自引：0，他引：1
Using conventional visual interpretation in lineament analysis presents two main problems. The first is subjectivity, introduced because of the bias of various interpreters. The second problem is that lineaments detected from satellite images are constrained by the direction of the illumination source. Since lineament identification mainly involves recognition of diagnostic morphological features, the use of digital elevation can contribute significant information about these features. Further, in generating images using digital elevation data, the direction of illumination can easily be controlled. Thus, the use of digital elevation data offers the possibility of revealing features not apparent in regular satellite images.We discuss a sequential line detection method for extraction of linear features from digital elevation data. In this method, raw elevation data is used for generating shaded relief images using the Lambertian reflection model, wherein the illumination direction is controlled by the user. The Directional Segment Detection Algorithm (DSDA) is used for detecting linear topographic features in user-defined trends. Locational information about these linear features is stored in the computer as coordinate pairs amenable to editing and subsequent analysis. Finally, three-dimensional terrain models are generated by combining the digital elevation data and satellite images. The experiments were carried out using digital elevation data of southwest Japan and Landsat MSS data. 相似文献
3.
基于多方位DEM地形晕渲的黄土地貌正负地形提取 总被引：5，自引：0，他引：5
以陕北绥德县韭园沟流域5 m分辨率DEM数据为基础,在数字地形分析、多元统计和数据挖掘方法的基础上,提出利用多方位DEM地形晕渲、坡度等多元指标,以主成分分析消除多重共线性和约减维数,并以Logis-tic回归模型提取黄土高原正、负地形的方法。研究结果表明:模型提取精度为82.1%,Kappa统计量为0.629;模型在6个不同流域测试样本上正、负地形的平均精度分别为77.6%,84.9%,加权平均精度为81.3%,模型具有较好的一致性和泛化能力,正、负地形提取效果良好。 相似文献
4.
Jingsong Wu Kevin Amaratunga 《International journal of geographical information science》2013,27(3):273-289
GIS applications have recently begun to emerge on the Internet. The management of three-dimensional geographic datasets in this distributed environment poses a particularly challenging problem, which highlights the need for a good data representation. This paper presents a new multiresolution data representation: the Wavelet Triangulated Irregular Network (WTIN). Compared to the traditional cell-based Digital Elevation Model (DEM) format and the Triangulated Irregular Network (TIN) format, it is more compact and suitable for scalable distributed GIS services. This format is based on the second-generation wavelet theory and is specially designed for geographical height field data. The modified Butterfly scheme is used for constructing the wavelet transform. For every point in the geographic surface, only a single wavelet coefficient is used, which makes the final data representation very efficient and easy to compress. Because the transform used in the data representation is a linear filter operation, the computational efficiency is better than other multiresolution data formats for terrain surfaces. Results from numerical experiments on real data are given to demonstrate that the proposed data representation can be efficiently implemented. The results show that the proposed WTIN data format can provide multiresolution data sets, which achieve significant compression while preserving geographical features. The quality is found to be quite acceptable for geographical terrain representation. 相似文献
5.
B. Yang W. Shi Q. Li 《International journal of geographical information science》2013,27(10):1019-1038
Multi‐resolution terrain models are an efficient approach to improve the speed of three‐dimensional (3D) visualizations, especially for terrain visualization in Geographical Information Systems (GIS). As a further development to existing algorithms and models, a new model is proposed for the construction of multi‐resolution terrain models in a 3D GIS. The new model represents multi‐resolution terrains using two major methods for terrain representation: Triangulated Irregular Network (TIN) and regular grid (Grid). In this paper, first, the concepts and formal definitions of the new model are presented. Second, the methodology for constructing multi‐resolution terrain models based on the new model is proposed. Third, the error of multi‐resolution terrain models is analysed, and a set of rules is proposed to retain the important features (e.g. boundaries of man‐made objects) within the multi‐resolution terrain models. Finally, several experiments are undertaken to test the performance of the new model. The experimental results demonstrate that the new model can be applied to construct multi‐resolution terrain models with good performance in terms of time cost and maintenance of the important features. Furthermore, a comparison with previous algorithms/models shows that the speed of rendering for 3D walking/flying through has been greatly improved by applying the new model. 相似文献
6.
Çetin Koca 《International journal of geographical information science》2014,28(9):1821-1847
Terrain rendering is a crucial part of many real-time applications. The easiest way to process and visualize terrain data in real time is to constrain the terrain model in several ways. This decreases the amount of data to be processed and the amount of processing power needed, but at the cost of expressivity and the ability to create complex terrains. The most popular terrain representation is a regular 2D grid, where the vertices are displaced in a third dimension by a displacement map, called a heightmap. This is the simplest way to represent terrain, and although it allows fast processing, it cannot model terrains with volumetric features. Volumetric approaches sample the 3D space by subdividing it into a 3D grid and represent the terrain as occupied voxels. They can represent volumetric features but they require computationally intensive algorithms for rendering, and their memory requirements are high. We propose a novel representation that combines the voxel and heightmap approaches, and is expressive enough to allow creating terrains with caves, overhangs, cliffs, and arches, and efficient enough to allow terrain editing, deformations, and rendering in real time. 相似文献
7.
8.
Mengyun Zhou Jianya Gong 《International journal of geographical information science》2016,30(11):2208-2229
Polygonal vector data are important for representing countries, lakes, residential settlements, and other polygonal features. The proper representation of polygonal vector data is the basis of efficient rendering and picking and quick access and display of the analysis results based on polygons (e.g., 3D overlaying and surface area measurement in mountainous areas) in a virtual globe. However, polygonal vector data are displayed using texture-based or boundary-based approaches in most existing virtual globes. The texture-based approach cannot easily support interactive operations (e.g., picking) and spatial analysis (e.g., adjacency analysis and spatial measurement). The boundary-based approach treats the holes as independent features; however, it is difficult to recognize which boundaries constitute a polygon. Further research is needed on how to better organize the polygons to support efficient rendering, picking, and analysis in a virtual globe. In this article, we propose two methods to drape interior filled 2D polygons onto a multi-resolution 3D terrain. Both proposed methods combine polygon clipping and polygon triangulation. The difference between the two methods is in the way holes are eliminated. Method 1 recursively subdivides a terrain triangle until the child-triangles contain no holes; every resulting clipped polygon, which is then triangulated, contains no holes. Method 2 directly clips a polygon against a terrain triangle and creates bridge edges to transform the resulting polygons with holes to degenerate polygons that are further triangulated. The experimental results demonstrate that both proposed methods can efficiently process polygons with holes resulting in appropriate numbers of triangles. The processed interior-filled polygons remain close to the terrain surface in a virtual globe. Both proposed methods support real-time rendering of polygonal vector data in a virtual globe. 相似文献
9.
不同分辨率DEM提取地面坡度的不确定性模拟:以在黄土高原的试验为例 总被引：1，自引：0，他引：1
Slope is one of the crucial terrain variables in spatial analysis and land use planning,especially in the Loess Plateau area of China which is suffering from serious soil erosion. DEM based slope extracting method has been widely accepted and applied in practice. However slope accuracy derived from this method usually does not match with its popularity. A quantitative simulation to slope data uncertainty is important not only theoretically but also necessarily to applications. This paper focuses on how resolution and terrain complexity impact on the accuracy of mean slope extracted from DEMs of different resolutions in the Loess Plateau of China. Six typical geomorphologic areas are selected as test areas, representing different terrain types from smooth to rough. Their DEMs are produced from digitizing contours of 1:10,000 scale topographic maps. Field survey results show that 5 m should be the most suitable grid size for representing slope in the Loess Plateau area. Comparative and math-simulation methodology was employed for data processing and analysis. A linear correlativity between mean slope and DEM resolution was found at all test areas,but their regression coefficients related closely with the terrain complexity of the test areas. If taking stream channel density to represent terrain complexity, mean slope error could be regressed against DEM resolution (X) and stream channel density (S) at 8 resolution levels and expressed as (0.0015S2+0.031S-0.0325)X-0.0045S2-0.155S+0.1625, with a R2 value of over 0.98. Practical tests also show an effective result of this model in applications. The new development methodology applied in this study should be helpful to similar researches in spatial data uncertainty investigation. 相似文献
10.
Loess terrain segmentation from digital elevation models based on the region growth method 总被引：1，自引：0，他引：1
Regional automatic segmentation – automatic terrain segmentation according to terrain features – is significant for modern geographical analysis. We propose a new approach of terrain region segmentation based on the region growth method. This method features actual runoff nodes as seed points. The corresponding growth threshold is defined based on statistical analysis of quantitative indexes of topographic features. Terrain segmentation of some regions is completed using the growth threshold. The corresponding edge boundaries of different terrain regions are extracted by image processing. Thus, the automatic segmentation of the terrain region is realized by the edge boundary. The application of the method to a typical Chinese loess landform area and automatic segmentation of three types of terrain regions – gully interfluve land, gully slope land, and gully groove land – are achieved by analyzing characteristics of the curvature structure of surface profiles. Segmentation results, compared with results of visual interpretation from a high-precision digital orthophoto map, show an average accuracy of segmentation of 93.51%. Topographic factor features of segmentation results are statistically analyzed. This study presents an effective and practical approach for segmenting terrain regions. This approach may be incorporated into the theory and method system of digital terrain analysis. 相似文献
11.
TANGGuoan ZHAOMudan LITianwen LIUYongmei ZHANGTing 《地理学报(英文版)》2003,13(4):387-394
Slope is one of the crucial terrain variables in spatial analysis and land use planning, especially in the Loess Plateau area of China which is suffering from serious soil erosion. DEM based slope extracting method has been widely accepted and applied in practice. However slope accuracy derived from this method usually does not match with its popularity. A quantitative simulation to slope data uncertainty is important not only theoretically but also necessarily to applications. This paper focuses on how resolution and terrain complexity impact on the accuracy of mean slope extracted from DEMs of different resolutions in the Loess Plateau of China. Six typical geomorphologic areas are selected as test areas, representing different terrain types from smooth to rough. Their DEMs are produced from digitizing contours of 1:10,000 scale topographic maps. Field survey results show that 5 m should be the most suitable grid size for representing slope in the Loess Plateau area. Comparative and math-simulation methodology was employed for data processing and analysis. A linear correlativity between mean slope and DEM resolution was found at all test areas, but their regression coefficients related closely with the terrain complexity of the test areas. If taking stream channel density to represent terrain complexity, mean slope error could be regressed against DEM resolution (X) and stream channel density (S) at 8 resolution levels and expressed as(0.0015S2 0.031S-0.0325)X-0.0045S2-0.155S 0.1625, with a R2 value of over 0.98. Practical tests also show an effective result of this model in applications. The new development methodology applied in this study should be helpful to similar researches in spatial data uncertainty investigation. 相似文献
12.
基于Kriging的地形高程插值 总被引：6，自引：0，他引：6
将地形高程作为区域化变量,根据普通Kriging法由散乱的高程点进行地形高程插值,并采用Matlab软件开发专门的程序,实现研究区高程插值计算与结果可视化分析。以广州市南沙区10 km2范围内的200个高程点数据为例,分别运用球面模型、指数模型和高斯理论变差函数模型进行10 m×10 m格网插值,借助Matlab可视化分析插值结果及其精度,表明采用指数模型效果最好。 相似文献
13.
Slope is one of the crucial terrain variables in spatial analysis and land use planning, especially in the Loess Plateau area of China which is suffering from serious soil erosion. DEM based slope extracting method has been widely accepted and applied in practice. However slope accuracy derived from this method usually does not match with its popularity. A quantitative simulation to slope data uncertainty is important not only theoretically but also necessarily to applications. This paper focuses on how resolution and terrain complexity impact on the accuracy of mean slope extracted from DEMs of different resolutions in the Loess Plateau of China. Six typical geomorphologic areas are selected as test areas, representing different terrain types from smooth to rough. Their DEMs are produced from digitizing contours of 1:10,000 scale topographic maps. Field survey results show that 5 m should be the most suitable grid size for representing slope in the Loess Plateau area. Comparative and math-simulation methodology was employed for data processing and analysis. A linear correlativity between mean slope and DEM resolution was found at all test areas, but their regression coefficients related closely with the terrain complexity of the test areas. If taking stream channel density to represent terrain complexity, mean slope error could be regressed against DEM resolution (X) and stream channel density (S) at 8 resolution levels and expressed as (0.0015S2+0.031S-0.0325)X-0.0045S2-0.155S+0.1625, with a R2 value of over 0.98. Practical tests also show an effective result of this model in applications. The new development methodology applied in this study should be helpful to similar researches in spatial data uncertainty investigation. 相似文献
14.
A digital elevation model (DEM), which is used to represent a terrain surface, is normally constructed by applying an interpolation method on given sample elevation points. Interpolation methods can be classified into two classes: linear methods, which have a low time cost and are suitable for terrains where there is little change in elevation, and nonlinear methods, which normally consume comparatively more time and are more suitable for terrains where there are frequent changes in elevation. A hybrid interpolation method, which involves both a linear method and a nonlinear method of interpolation, is proposed in this paper. The proposed method aims to integrate the advantages of both linear and nonlinear interpolation methods for the refinement of regular grid DEM. Here, the bilinear is identified as the linear method, and the bi‐cubic is taken to be the nonlinear interpolation method. The hybrid method is an integration of a linear model and nonlinear interpolation model with a parameter that defines the weights for each of the models. The parameter is dependent on the complexity of the terrain, for which a DEM is to be interpolated. The experimental results in this study demonstrate that the hybrid method is effective for interpolating DEMs for various types of terrain. 相似文献
15.
DEM提取黄土高原地面坡度的不确定性 总被引：72，自引：0，他引：72
选择陕北黄土高原6个典型地貌类型区为试验样区,采用野外实测及高精度的1:1万比例尺DEM为基准数据,研究栅格分辨率及地形粗糙度对DEM所提取地面平均坡度精度的影响。结果显示,对于1:1万比例尺DEM,5 m是保证该地区地形描述精度的理想分辨率尺度;多要素逐步回归模拟的方法进一步揭示了DEM所提取的地面平均坡度误差E与栅格分辨率X以及地形起伏的代表性因子-沟壑密度S之间存在的量化关系为E = (0.0015S2+0.031S-0.0325)X-0.0045S2-0.155S+0.1625,该结果也为确定适用的DEM分辨率提供了理论依据。 相似文献
16.
Quanhua Dong 《International journal of geographical information science》2018,32(2):302-323
Geometric buffers are important for spatial analysis in many applications of geographic information systems (GISs), such as environmental measurement and management, human health, urban planning, etc. Geometric buffer generation algorithms are well studied in the Euclidean space where the buffer distance is measured by Euclidean metrics; however, very few algorithms are available for generating geometric buffers on the terrain surface in a virtual globe where the buffer distance is measured by geodesic metrics. This paper proposes a tile-based method for geodesic buffer generation according to the characteristics of a virtual globe. It extends the vector tile model (VTM) to organize terrain and vector data, and the XYH algorithm is improved to build geodesic distance fields for terrain meshes. Based on the data organization and the improved XYH algorithm, a geodesic buffer is generated via three main steps: selecting and assembling tiles, updating geodesic distance fields and tracing the boundaries of buffer zones. This method is implemented with multi-scale terrain and vector data, and the experimental results show that it is valid and exact and can be applied in practical applications. 相似文献
17.
建设三维城市模型(3DCM)首先需要建立场景所需的海量地形地物空间数据库,从而有效组织三维实体数据来满足大范围高精度场景实时绘制的要求.该文提出一种以三维空间实体为单位的基于关系数据库(RDBMS)的3DCM的几何与纹理数据一体化管理方法.采用Direct3D 9.O三维开发平台、C#编程语言实现了3DCM空间数据库模型管理系统.实验证明:该系统在效率、安全、网络多用户访问方面都优于传统基于文件的管理模式,满足了大范围3DCM海量模型数据管理与应用需求. 相似文献
18.
面向地貌学本源的数字地形分析研究进展与展望 总被引：5，自引：2，他引：3
数字地形分析（DTA）是地理信息科学（GIS）研究的热点。但是,当前基于数字高程模型（DEM）的数字地形分析在地貌学研究中存在重形态轻机理、重现象轻过程、重地上轻地下等问题,急需从单一的地貌形态分析,迈向面向成因、过程与机理等地貌学本源问题的研究转变。据此,本文系统梳理了面向地貌学本源的数字地形分析的相关研究现状,并从地貌学本源认识、地貌形态建模、地形因子提取、以及其他地形分析方法等研究进行了系统的回顾、梳理与分析。研究表明,基于DEM的数字地形分析虽具有地貌特征分析的潜力与优势,但是,数字地形分析存在数据表达与分析模式上的先天缺陷,亟待通过基础理论与关键技术的突破,实现理论与方法的创新发展,实现从“坡面”走向“区域”,从“形态”走向“过程”,从“地形”走向“地貌”。而当今地球系统科学的研究发展态势也到了数字地形分析研究从重视地貌形态走向揭示地貌学本源的关键阶段。因此,本文从DEM数据模型增值、地形因子及其地形空间关系、以及宏观地形分析等侧面展望了面向地貌学本源的数字地形分析研究。当今基于DEM的数字地形分析研究,正像当前的GIS是否能够真正支撑地理学发展一样,已经处于一个非常关键的十字路口。面向地貌学本源的数字地形分析研究思路可望成为地理信息科学领域理论与方法创新的一次重要探索与实践。 相似文献
19.
基于山地灾害动态过程仿真需要的考虑,并针对该动态过程仿真时渲染数据量过大所造成的实时性较差的问题,提出了利用改进的三角形二叉树LOD算法实现地形三维建模与可视化的方法。算法对山地地形数据进行了分层和分块的预处理,用三角形二叉树表示地形网格,并结合视点和局部地形的粗糙程度,动态的载入所需的地形块和释放无用的地形块,使得内存中的地形数据维持在一定范围内。实验结果表明;在对地形渲染不失真的前提下,本方法能够有效地提高地形绘制的效率,可应用到大规模山地地形的三维建模与可视化中去,为整个山地灾害的动态过程仿真奠定了良好的基础。 相似文献
20.
E.G. Paredes M. Amor M. Bóo J.D. Bruguera J. Döllner 《International journal of geographical information science》2016,30(6):1095-1116
Hybrid terrain models combine large regular data sets and high-resolution irregular meshes [triangulated irregular network (TIN)] for topographically and morphologically complex terrain features such as man-made microstructures or cliffs. In this paper, a new method to generate and visualize this kind of 3D hybrid terrain models is presented. This method can integrate geographic data sets from multiple sources without a remeshing process to combine the heterogeneous data of the different models. At the same time, the original data sets are preserved without modification, and, thus, TIN meshes can be easily edited and replaced, among other features. Specifically, our approach is based on the utilization of the external edges of convexified TINs as the fundamental primitive to tessellate the space between both types of meshes. Our proposal is eminently parallel, requires only a minimal preprocessing phase, and minimizes the storage requirements when compared with the previous proposals. 相似文献