规则格网DEM地形综合方法分析

针对当前丰富多样的DEM地形综合模型,分类阐述其综合原理、关键技术及优缺点,以黄土丘陵1：1万DEM为基础,对其地形综合效果进行对比分析。研究结果表明：现有的DEM地形综合模型各有优缺点;不同综合模型综合后的地形,高程数值随着综合尺度的增加,存在显著差异,而且不同综合模型对流域边界的综合处理也存在显著差异,没有形成客观有效的流域边界描述;当前困扰DEM地形综合的关键问题,是缺乏不同尺度下地形综合质量的客观评价标准,致使现有综合模型地形综合结果和DEM格网分辨率之间没有建立本应存在的内在联系,有待进一步深入研究。     

引潮位的Doodson规格化展开及其精度评定

基于引潮位Doodson规格化展开的基本原理以及ELP/MPP02月球历表、Newcomb太阳历表的结构,设计了一个用以代表三角函数的数据结构;并由此自定义了三角函数的乘法与加法算法,将月球、太阳引潮位分别展开至5阶、3阶,振幅绝对值截断阈值为10-7,在展开过程中对"伪波"进行"滤波"处理后,最终得到包含4 686项展开式的引潮位展开表(其中振幅绝对值超过10-6的展开式有2 441项)。以德国BFO(Black Forest Observatory)测站为例,基于DE431历表,通过天球参考系变换计算得到1950—2050年间时间间隔为1 h的法向引潮力基准序列BFDE431;并根据各个引潮位展开表计算得到相应的法向引潮力序列;通过与BFDE431基准序列求差,得到各个差值序列的统计值。计算结果表明,文中给出的展开表对应的差值序列数值在±58×10~(-11)m·s~(-2)(nGal)以内,均方差为12.5×10~(-11)m·s~(-2),与XI89展开表的精度相当。但由于没有考虑行星及地球扁率的影响,仍未达到HW95和RATGP95展开表的精度水平。     

双混沌映射的土家织锦图案数字加密算法

结合混沌序列易生成性、对初始条件敏感性、可完全重现性等特点,给出了一种基于混沌映射的土家织锦图案数字加密算法。首先对原始图案进行离散小波变换(DWT),变换的矩阵奇行与Logistic映射产生的二值序列异或,偶行与Hybrid映射产生的二值序列异或,然后对产生的新矩阵进行DWT逆变换,即得到加密的织锦图案。实验结果表明,本文算法具有较高的安全性和较好的加密效果。     

一种基于伪随机序列和DCT的遥感影像水印算法

首先,对原始影像进行分块,通过伪随机序列选择要嵌入的影像块;然后,采用量化的方法在每一块的DCT低频系数中嵌入水印信息;最后,通过精度约束,对嵌入水印信息后的遥感影像进行误差控制。由于采用了量化的方法嵌入水印信息,其提取过程不需要原始影像的参与,实现了盲检测。实验结果表明,该算法在满足水印算法鲁棒性的同时,也能达到对遥感影像近乎零修改的目标。     

Application of Spatial Database in Quantitative Analysis of Litho-Paleogeography—A Case Study of a Middle Ordovician Sequence Interval in the Ordos Basin

In the study of sequence stratigraphy and litho-paleogeography,quantitative analysis,precise calculation and detailed comparison of tremendous geological data,such as field profiles,logging records and seismic curves from different areas,are the basic requirements.In order to obtain a more reliable and precise result,this paper presents a novel method that combines spatial database analysis with the single-factor mapping technology to establish sequence stratigraphical succession and to map the Ordo-vician litho-paleogeography of the Ordos Basin,one of the largest oil-gas bearing basins in North China Platform.By using this method,all of the related basic geological data can be quantitatively analyzed and effectively managed.Various attributes of the ba-sic stratigraphic units and their characters,such as sequence thickness,penecontemporaneous dolostone content,shallow water parget content,and terrigenous material content,can be fully utilized statistically in facies analysis and in mapping.Based on this analysis,this paper has be exerted single-factor isopachous mapping quantitatively for each of the Ordovician sequences in the ba-sin,and finally synthesized multiple factors to reconstruct the litho-paleogeography for each of the sequence intervals.The study shows that the proposed method is quite effective and has a much higher resolution in recognizing litho-paleogeographic subunits compared with traditional ways.For example,in one of the Middle Ordovician sequence intervals(SQ19 in the Lower Majiagou Formation) of the Ordos Basin,the authors have successfully developed a mathematical formula to divide the distribution of various facies units substantially,such as old lands,submarine uplifts,supratidal zones,intertidal zones and subtidal zones.     

软多边形地图要素弯曲识别模型及其应用研究

以边界线构成点的坐标单调性变化分析为基础,提出了地图上软多边形要素弯曲识别模型,给定了弯曲的凸凹性定义,建立了弯曲的多种形态量化指标,规范了弯曲邻接关系的形式及确定方法。研究了模型在软多边形地图要素图形综合简化中的应用,形成了简化这类要素图形的三种基本手段。     

动态灰色模型在变形预测中的应用

本文在引入序列算子成功构造灰色序列的基础上,详细讨论了灰色预测模型的基本内容和两种动态灰色预测模型的实现,并成功将其应用于广州地铁变形监测的数据预报。实践及理论证明,新陈代谢灰色预测模型由于实时加入系统的最新信息,提高灰区间的白度,预测精度最高;灰数递补模型通过引入灰数,一定程度上也提高了预测的精度。新陈代谢灰色预测模型实践证明预测精度可达到一级。     

SAS/ETS在变形监测数据处理中的应用

在综合考虑数学模型的预测精度与编程实现的可操作性基础上,提出了利用模块对变形监测数据进行时间序列分析,并概括介绍了该模块的3个主要命令使用方法,解决了时间序列分析在监测数据处理中实现起来非常繁琐的问题。最后通过与环境下动态灰色模型综合比较,证明了时间序列分析在变形监测数据处理中操作简单,模拟精度和预测精度都较高。     

BOOK REVIEW

Eolian landforms as sandy dunes and dune heaps were mapped by the author in 1979–1980 in a zone where vegetation had been destroyed by a forest fire on 30 June 1976, in an area to the east of Elverum, south-east Norway. The mapped eolian landforms are discussed together with information from 40 samples of the eolian sediment. Winds, mostly from the north and north-west, were responsible for the development of these eolian landforms, utilizing the fluvioglacial sediments accumulated as a sandur during the retreat of the inland ice approximately 10,000 years ago. The winds were active during a short period of time, only a century or so before the area became vegetated.