一种中分辨率遥感影像道路的快速搜索算法

基于动态规划的思想,提出了一种用于中等分辨率遥感影像的道路搜索算法,利用线段代价最小规则进行道路线段跟踪,实验结果表明算法相对简单、快速有效、抗噪性能好。     

地表形变监测的改进相干目标法

如何从雷达干涉时间序列影像中获取更全面的相干目标集合,进行变形时间序列分析是当前研究的难点和热点。本文提出了改进的相干目标法,可获取更为全面且可信度高的相干目标集合,进而提高地表形变监测的时空分辨率和精度。根据雷达影像中同类型地物散射分布相近的特点,采用非参数同分布检验算法提取后向散射特性相近的同质点开展空间非局部滤波,提高干涉图的质量。与此同时,利用多尺度的极大似然条纹频率估计算法分离差分干涉图中的系统性相位,并基于同质点进行自适应相干性估计,获取相干性的平稳估计量,从而获取更多的相干点目标。利用20景TerraSAR-X条带模式时间序列影像,分别利用传统的及改进后的相干目标法对香港填海区域地表形变信息进行时序分析。对试验数据的分析结果表明,本文提出的改进方法在具有稀疏植被的填海区可有效增加相干目标点的提取,得到更为可信的沉降结果。     

�Ľ��¶ȵ�LiDAR������̬ѧ�˲��㷨

????????????????д????????ò???????????????????????????ε??????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????硢??????????????????Ч??????????????????????????????????????????С??     

Development of Geological Data Warehouse

Data warehouse (DW), a new technology invented in 1990s, is more useful for integrating and analyzing massive data than traditional database. Its application in geology field can be divided into 3 phrases: 1992-1996, commercial data warehouse (CDW) appeared; 1996-1999, geological data warehouse (GDW) appeared and the geologists or geographers realized the importance of DW and began the studies on it, but the practical DW still followed the framework of DB; 2000 to present, geological data warehouse grows, and the theory of geo-spatial data warehouse (GSDW) has been developed but the research in geological area is still deficient except that in geography. Although some developments of GDW have been made, its core still follows the CDW-organizing data by time and brings about 3 problems: difficult to integrate the geological data, for the data feature more space than time; hard to store the massive data in fifferent levels due to the same reason; hardly support the spatial analysis if the data are organized by time as CDW does. So the GDW should be redesigned by organizing data by scale in order to store mass data in different levels and synthesize the data in different granularities, and choosing space control points to replace the former time control points so as to integrate different types of data by the method of storing one type data as one layer and then to superpose the layers. In addition, data cube, a wide used technology in CDW, will be no use in GDW, for the causality among the geological data is not so obvious as commercial data, as the data are the mixed result of many complex rules, and their analysis needs the special geological methods and software; on the other hand, data cube for mass and complex geo-data will devour too much store space to be practical. On this point, the main purpose of GDW may be fit for data integration unlike CDW for data analysis.     

论地质项目管理中的几种关系

近年来湖南找矿突破前景乐观,如地质四一三队发现了含金刚石的橄榄钾镁煌斑岩体群。为扩大找矿突破成果,在地质项目管理中,需要理顺国家需要与地质可能、导向矿种与一般矿种、大型矿床与小型矿床、矿床新类型与传统类型以及勘查与开发等五种关系。     

湖南白兔潭矿区荷田区段构造特征与找煤前景

煤炭是湖南省的主要能源,经过多年的勘探开发,暴露区和浅部的煤炭资源已基本采完,因此,在隐伏区寻找新的煤炭资源已成为当务之急。隐伏区找煤主要指“三下”(老地层下、构造下和红层下)找煤,而“老地层下和构造下”找煤主要是指利用推覆构造理论寻找新的煤炭资源。为了深化对区域构造和煤田构造发育规律的认识,为本区煤炭资源勘探和矿井生产地质保障提供新思路、开拓新领域,以当代构造地质理论为指导,结合已有资料,对白兔潭矿区荷田区段的煤田构造进行了研究,揭示了煤田的构造格局,提出了该区可能存在的两种构造模式,对该区找煤前景做出了科学评价,指出了可供进一步勘探的有利区段。     

陕川丁家林-太阳坪金矿床黄铁矿标型特征

金矿床中黄铁矿的化学元素、晶体形态、粒度等标型特征与其形成的物理化学环境及介质条件密切相关.通过对陕川丁家林-太阳坪金矿区主要载金矿物黄铁矿在韧性剪切变形-构造分异热液期与脆性剪切变形-构造分异热液期化学元素含量对比,以及黄铁矿形态、粒度分析,研究黄铁矿与金矿物关系.丁家林-太阳坪金矿黄铁矿形态以{100}、{210}单晶及{100} {210}聚晶为其标型.黄铁矿在空间分布特征来看,深部成矿的可能性不大.应沿丁家林-太阳坪脆-韧性剪切带寻找金矿.     

基于二维图像特征的点云配准方法

为了提高低覆盖率点云的配准精度和收敛速度,提出了一种基于二维图像特征的点云配准方法。首先采用基于区域层次的点云配准算法实现粗配准;然后将三维点云转换成二维图像,再采用SURF算法提取二维图像的特征,并求解其匹配像素点对;最后根据二维匹配点获取相应的三维点云相关点,并计算刚体变换,由此实现点云的快速精确配准。试验结果表明,与迭代最近点（ICP）算法相比,该点云配准方法的配准精度和耗时分别提高了约20%和60%,是一种快速、高精度的点云配准算法。     

采集参数对音频大地电磁法二维非线性共轭梯度反演结果的影响研究

非线性共轭梯度(NLCG)算法广泛应用于音频大地电磁法(AMT)二维反演中,并且取得了良好效果.本文以Rodi(2001)提出的大地电磁法二维NLCG反演理论和算法为基础,开展AMT野外数据采集参数(测线长度、点距、频点数)对二维反演效果的影响研究.通过低阻理论模型的试验结果表明:野外数据采集参数的选择应该以探测目标体的规模、深度以及背景电阻率为依据,测线长度应大于目标体宽度和探测深度的两倍以上,异常体上方测点数及探测深度范围内的频点数至少大于三个.     

平面约束条件在LIDAR点云滤波中的应用

介绍了一种利用平面约束条件对LIDAR点云数据进行滤波的方法,利用每个数据点的邻域点拟合平面,根据平面约束条件和平面点分类方法得到地面点,最后利用地面点内插该区域的DTM.