地性线的山地区域的卫星影像几何精纠正

提出了一种山地区域基于DEM地性线的控制纠正新方法,该方法以数字地形模型DEM为无几何变形的控制基准纠正卫星影像。阐述了提取沟谷、山脊、山峰和凹地区域的地性线的原理和算法,给出了山地区域基于地性线进行卫星图像几何精纠正实施步骤,进一步讨论了地性线提取、控制点采集存在的问题,以及解决问题的途径。实验结果表明,对于山地区域,地性线的空间数量数倍于水系、道路等常规地图层;地性线来源于DEM,其空间稳定性和可靠性更高,可以用于山地区域的卫星影像的严格控制纠正。用该方法进行几何纠正处理,几何误差能控制在一个像元的水平上。     

非线性多维数据可视化分类预测方法

地理信息分类的传统线性算法具有正向直接判定的快速优势,但局限于对已知数据进行线性的判别划分,而非线性未知信息的分类预测同样是GIS技术的重要内容。人工神经网络算法为一些非线性知识的发现提供了可能。本文在通用的GIS格式数据基础上,采用L-M算法进行分类,通过分类结果来预测未知信息。开发出可视化的GIS数据神经网络分类预测软件模块。并以美国各镇人口为样例数据进行测试,分类预测结果显示该算法具有可行性及系统具有实用性。     

网络环境下全球影像的高效传输与调度策略

本文针对网络环境下实现全球影像数据的快速可视化所涉及的数据组织和传输调度策略等关键技术进行了探讨。在计算机屏幕分辨率限制的基础上,对全球数据的分块方法进行了优化,并提出了智能控制的多线程网络传输技术和多级缓存的数据流机制,最后通过实验表明,该技术方案是切实可行的,能够平稳快速地对全球范围内TB级影像数据进行浏览,满足了C/S模式下大规模影像数据实时漫游的要求。     

地形三维场景实时变换及可视化方法研究

地形数据的三维可视化技术一直是3DGIS领域一项重要的研究技术。从地形实时渲染的需求出发,提出大地坐标系下的多分辨率地形组织模型,并对可视化过程中的坐标空间变换方法进行研究,提出一种适合大区域地形可视化的坐标空间变换方案,使可视化的三维地形场景尽可能真实反映其自然的分布形态,使影像纹理、地物与地形之间进行精确的映射与匹配。     

地形因素对InSAR影像配准的影响及分析

常规的InSAR影像配准方法主要是根据相干系数最大的原则进行配准,在短基线和高相干区域可以取得很好的效果。但该方法未考虑到地形因素的影响,在地形起伏较大的区域容易导致误匹配。为避免该问题,通过考虑地形因素进行InSAR影像配准来提高配准的精度,对地形因素在InSAR影像配准时的影响进行研究,通过比较分析常规配准方法与考虑地形因素配准方法的结果,表明利用地形信息辅助配准能够部分提高配准的精度,尤其是在方位向和失相干区域。     

Effects of Interpolation Errors on the Analysis of DEMs

A suite of methods to interpolate a digital elevation model from a ground survey was evaluated with respect to precision and ability to maintain the shape of the original height data. This shape reliability was evaluated by comparing the spatial patterns of secondary terrain parameters derived from the interpolated elevation data. The best interpolation method for this study area was found to be a spline interpolation, which is somewhat contradictory to findings in the literature. The error and uncertainty found in the results for terrain analysis and modelling tools is important and sometimes distressingly high, even for some frequently used local or context operations on altitude. Positional operations, in which the output is determined more by the position in the topographic structure, seem to give more reliable results. Therefore, the results obtained by terrain analysis and spatial modelling need careful interpretation. © 1997 John Wiley & Sons, Ltd.     

Atmospheric deposition to watersheds in complex terrain

Single collection stations for wet or bulk deposition are generally inadequate to describe atmospheric inputs to watersheds in complex terrain. Atmospheric deposition is delivered by wet, dry and cloud deposition processes, and these processes are controlled by a wide range of landscape features, including canopy type and structure, topographic exposure, elevation and slope orientation. As a result, there can be a very high degree of spatial variability within a watershed, and a single sampling point, especially at low elevation, is unlikely to be representative. Atmospheric inputs at the watershed scale can be calculated from the whole watershed mass balance if the outputs and within-watershed sources and sinks are known with sufficient accuracy. Alternatively, indices of atmospheric deposition such as Pb accumulation in the forest floor and SO²⁻₄ flux in throughfall can be used to characterize patterns of total deposition, and these indices can be used to model deposition to the entire watershed based on known landscape features such as elevation and canopy type. © 1997 John Wiley & Sons, Ltd.     

Hydrological runoff modelling by the use of remote sensing data with reference to the 1993–1994 and 1995 floods in the river Rhine catchment

In hydrological modelling of runoff processes, including water balance, various input data and parameters can be acquired or estimated by the use of remote sensing (RS) techniques.The acquisition and use of synoptic RS areal information rather than traditional point information is an important issue in hydrology. Hydrological models allow runoff/water balance in catchments to be calculated and flow routing within flow channels to be done. For runoff and water balance computations land use, soil moisture, detection of snow and ice, digital terrain models (DTM), as well as hydrometeorological information and discharge are important. For flow routing, water level information, geometric–topographic information such as cross-sections for normal and flood conditions, coefficient of roughness and velocity of flow and its cross-sectional distribution are required. In addition, water level information (lower and upper level) is needed for shipping and for design purposes. In the German part of the River Rhine catchment, several focus areas in the December 1993–January 1994 and January 1995 floods were covered with RS data [ERS-1 and airborne SAR, both C-band VV, passive microwave (18·7, 36·5, 89 GHz), TIR, UV, aerial photographs (b/w PAN, b/w NIR)], giving a good opportunity for a comparison of methods. Evaluation is still continuing. The importance of soil saturation for flood generation and, therefore, for flood monitoring, was shown on this occasion. The use of ERS SAR data for soil moisture estimation is currently being investigated by the Federal Institute of Hydrology. Also, the need for emergency schemes for data acquisition and easy, quick and affordable RS data dissemination was demonstrated. The assimilation of RS data with GIS information such as DTMs, including relevant topographic features like dams, which is omitted in currently available raster digital elevation models, is promising. RS altimetry techniques can be a step towards high resolution DTMs for hydrological purposes. Ground truth reference data are still needed. © 1997 John Wiley & Sons, Ltd.     

3维GIS在国土信息可视化中的运用研究

在国土资源信息管理中,为了更好掌握现状,对3维可视化的要求已经成为很多国土信息系统的高端要求,本文论述了3维GIS的数据结构原理、3维空间数据库的建设,并对一个具体工程进行了介绍。     

全球降水计划IMERG和GSMaP反演降水在四川地区的精度评估

IMERG和GSMaP是全球降水计划（GPM）时代最主要的高分辨率降水产品。为研究其在中国四川地区的适用性,以中国气象局提供的自动气象站融合降水数据为参考基准,采用6种统计指数分析了IMERG（IMERG_Uncal, IMERG_Cal）和GSMaP（GSMaP_MVK, GSMaP_Gauge）系列产品在四川的误差特征。结果表明：① 在日和小时尺度上,GSMaP系列产品均高估地面降水观测,GSMaP_MVK高估最显著,校正产品GSMaP_Gauge的相关系数（CC）、相对偏差（BIAS）和均方根误差（RMSE）较GSMaP_MVK均有较大提高,尤其对川西高原降水的高估现象改善明显,而IMERG_Uncal存在低估川西高原降水、轻微高估四川盆地降水的问题,校正产品IMERG_Cal一定程度上降低了对川西高原降水的低估现象,但整体精度（CC, RMSE）提高不明显。② IMERG系列产品对降水事件的探测准确性更好,GSMaP_Gauge虽然在四川表现出较高的命中率（POD）,但存在较多的误报降水,在盆地和四川南部各产品均表现出较高的POD和关键成功指数（CSI）以及低误报率（FAR）,而四川西北部表现最差,尤其是在无自动站分布地区。③ 4套降水产品中,IMERG_Cal表现出最好的探测强降水和弱降水的能力,具有一定的监测极端降水的潜力。总体上,IMERG和GSMaP在盆地的反演精度优于高原山区,校正产品精度优于纯卫星产品,不同地形地区精度差异明显,表明对卫星降水产品进行不同地形误差订正仍是未来降水反演工作的重点和难点。