An evaluation of Guided Regularized Random Forest for classification and regression tasks in remote sensing

New Earth observation missions and technologies are delivering large amounts of data. Processing this data requires developing and evaluating novel dimensionality reduction approaches to identify the most informative features for classification and regression tasks. Here we present an exhaustive evaluation of Guided Regularized Random Forest (GRRF), a feature selection method based on Random Forest. GRRF does not require fixing a priori the number of features to be selected or setting a threshold of the feature importance. Moreover, the use of regularization ensures that features selected by GRRF are non-redundant and representative. Our experiments based on various kinds of remote sensing images, show that GRRF selected features provides similar results to those obtained when using all the available features. However, the comparison between GRRF and standard random forest features shows substantial differences: in classification, the mean overall accuracy increases by almost 6% and, in regression, the decrease in RMSE almost reaches 2%. These results demonstrate the potential of GRRF for remote sensing image classification and regression. Especially in the context of increasingly large geodatabases that challenge the application of traditional methods.     

地壳演化与成矿作用——以川滇地洼系“四层楼”铜矿床序列为例

提出川滇地洼系“四层楼”铜矿床序列的形成与陆壳演化的成生联系,是与本区陆壳由前地槽—地槽—地台—地洼演化各阶段与之相匹配的成矿作用的产物.与此同时,并总结了本区“四层楼”铜矿床序列的成矿作用具有明显的继承性、新生性、旋回性及层控性四大特点和多因复成矿床的成矿模式.     

测井资料在碳酸盐岩储层评价中的应用

以塔河油田碳酸盐岩储集层基本特征为依据，利用常规和能谱测井资料信息比值法和重叠法等技术．致力于储集层缝、孔、洞测井响应特征敏感性评价研究，从中提取8项储层敏感性参数，并对其进行综合评判，确定出划分储层类别的标准，经过50口井的测试资料验证，采用多参数综合评判来划分储层类别是可行的，对储层的评价效果较好，使常规测井资料得到了更深入、广泛的应用。     

滇西潞西上芒岗卡林型金矿

上芒岗金矿赋存于二叠系沙子坡组碳酸岩和侏罗系勐戛组下段砂岩、灰岩中,明显受上芒岗断裂控制,与美国模式卡林型金矿极为相似,具有相同的物质来源和成因模式。上芒岗地区具有广阔的深部找矿前景。     

上饶市2000年6月22～23日连续暴雨过程浅析

对上饶市2000年6月22～23日连续暴雨的形势场以及卫星云图进行了分析,得出了一些有益的结论。     

山东龙口大顶金矿地质特征及找矿方向

重点介绍大顶金矿区的地质特征及区内主要矿脉体(36号矿脉群)特征，同时，对控矿因素作了一定的分析，初步总结了36号矿脉群在双顶矿段北东缘的成矿规律，指明了本矿区的找矿标志及下一步的找矿方向。     

Karst database development in Minnesota: design and data assembly

The Karst Feature Database (KFD) of Minnesota is a relational GIS-based Database Management System (DBMS). Previous karst feature datasets used inconsistent attributes to describe karst features in different areas of Minnesota. Existing metadata were modified and standardized to represent a comprehensive metadata for all the karst features in Minnesota. Microsoft Access 2000 and ArcView 3.2 were used to develop this working database. Existing county and sub-county karst feature datasets have been assembled into the KFD, which is capable of visualizing and analyzing the entire data set. By November 17 2002, 11,682 karst features were stored in the KFD of Minnesota. Data tables are stored in a Microsoft Access 2000 DBMS and linked to corresponding ArcView applications. The current KFD of Minnesota has been moved from a Windows NT server to a Windows 2000 Citrix server accessible to researchers and planners through networked interfaces.     

腾冲-梁河地区锡铜铅锌稀有多金属成矿模式

腾冲-梁河地区Sn、Cu、Pb、Zn等元素成矿规律明显,找矿标志清楚。由东往西划分为四个次级成矿带,各具有不同矿产组合或典型矿床。根据成矿地质条件、矿产分布特征、控矿因素和找矿标志等,归纳5种成矿模式。初步建立该区矿床找矿模型,为今后矿产预测奠定基础。     

从黄铁矿特征论380地区金鸡岭岩体外带铀矿床成矿机理

本文以铀矿区广泛发育的黄铁矿为研究对象，通过成因矿物学方法探讨金鸡岭岩体外带铀矿床的成矿机理；论证了矿质来源于岩体的围岩，与花岗岩关系不密切；岩体侵位所引起的围岩热变质作用，以及由此而产生的变质热液，是导致成矿物质重新活化、迁移和富集成矿的根本原因。     

基于北京一号卫星影像的建筑阴影测量建筑高度的应用研究

在分析遥感影像建筑物阴影与实际高度关系的基础上,阐述了依据影像阴影估算城市建筑物高度的原理和方法,设计了一种基于影像建筑物阴影特征快速提取高程信息的技术方法。在使用北京一号影像数据的试验中,70%的有阴影楼房测量误差在4米以内。