Land cover classification by an artificial neural network with ancillary information

Abstract

Remote sensing is an important source of land cover data required by many GIS users. Land cover data are typically derived from remotely–sensed data through the application of a conventional statistical classification. Such classification techniques are not, however, always appropriate, particularly as they may make untenable assumptions about the data and their output is hard, comprising only the code of the most likely class of membership. Whilst some deviation from the assumptions may be tolerated and a fuzzy output may be derived, making more information on class membership properties available, alternative classification procedures are sometimes required. Artificial neural networks are an attractive alternative to the statistical classifiers and here one is used to derive a fuzzy classification output from a remotely–sensed data set that may be post–processed with ancillary data available in a GIS to increase the accuracy with which land cover may be mapped. With the aid ancillary information on soil type and prior knowledge of class occurrence the accuracy of an artificial neural network classification was increased by 29–93 to 77–37 per cent. An artificial neural network can therefore be used generate a fuzzy classification output that may be used with other data sets in a GIS, which may not have been available to the producer of the classification, to increase the accuracy with which land cover may be classified.     

Retrieval algorithm for atmospheric dust properties from Mars Global Surveyor Thermal Emission Spectrometer data during global dust storm 2001A

We present a new parameter retrieval algorithm for Mars Global Surveyor Thermal Emission Spectrometer data. The algorithm uses Newtonian first-order sensitivity functions of the infrared spectrum in response to variations in physical parameters to fit a model spectrum to the data at 499, 1099, and 1301 cm⁻¹. The algorithm iteratively fits the model spectrum to data to simultaneously retrieve dust extinction optical depth, effective radius, and surface temperature. There are several sources of uncertainty in the results. The assumed dust vertical distribution can introduce errors in retrieved optical depth of a few tens of percent. The assumed dust optical constants can introduce errors in both optical depth and effective radius, although the systematic nature of these errors will not affect retrieval of trends in these parameters. The algorithm does not include the spectral signature of water ice, and hence data needs to be filtered against this parameter before the algorithm is applied. The algorithm also needs sufficient dust spectral signature, and hence surface-to-atmosphere temperature contrast, to successfully retrieve the parameters. After the application of data filters the algorithm is both relatively accurate and very fast, successfully retrieving parameters, as well as meaningful parameter variability and trends from tens of thousands of individual spectra on a global scale (Elteto, A., Toon, O.B. [2010]. Icarus, this issue). Our results for optical depth compare well with TES archive values when corrected by the single scattering albedo. Our results are on average 1–4 K higher in surface temperatures from the TES archive values, with greater differences at higher optical depths. Our retrieval of dust effective radii compare well with the retrievals of Wolff and Clancy (Wolff, M.J., Clancy, R.T. [2003]. J. Geophys. Res. 108 (E9), 5097) for the corresponding data selections from the same orbits.     

铁路缓和曲线的等间隔放样方法

介绍了铁路缓和曲线设置的目的、作用、基本原则和重要意义，阐述了等间隔放样缓和曲线的基本原理，给出了等间隔放样缓和曲线放样要素采集的基本方法，叙述了缓和曲线等间隔放样的工作过程。     

2002年秋季山东省干旱遥感监测分析

利用极轨气象卫星遥感资料对2002年山东省秋季干旱进行监测和服务,其监测方法主要采用热惯量法。为了提高土壤水分模式的计算精度,将全省分为4个区域,对利用卫星资料反演的地表温度日较差进行植被指数等环境因素订正。在地理信息系统支持下,比较准确地计算出农田受旱面积,效果良好。     

AutoCAD与其他常用软件的数据转换

本文就AutoCAD与生产应用中常用的ARCGIS、ARC／INFO、MicroStation、3DS等软件系统之间的数据转换作一些浅谈。     

土地利用动态管理系统研发中的若干问题研究

土地利用管理是土地管理的核心内容之一。土地利用动态变化加剧了土地管理的压力，建设具有辅助决策功能的土地利用动态信息系统是大势所趋。本文在阐述建设此系统的重要意义基础上，讨论了系统研发中几个关键问题，即“动态”管理的实现、空间数据和属性数据集成管理和决策的初步实现等。本文提出的解决方法，对于类似土地信息系统的建设具有普遍的指导意义。     

煤矿地理信息系统数据模型的研究

地下煤矿开采所面对的空间目标具有灰色、动态变化的特征。本文在分析煤矿空间信息特点的基础上,提出了实用于煤矿的地理信息系统数据模型。这些数据模型包括全要素的结构化不规则三角网（ＴＩＮ）与ＧＩＳ一体化数据模型以及网状模型。     

教师在数字化测图教学实践中的主导作用

介绍在数字化测图实践中,各关键技术环节上教师的主导作用,总结了实践教学的经验和体会。     

互联网时代GIS设计的新观念──以数据为中心的GIS

本文讨论了以数据为中心的GIS设计新概念。指出互联网造就了巨大的信息资源,GIS的数据源从自供自给到从网上获取是一种观念的巨大转变。进而探讨了WebGIS在网上数据服务,网上数据互操作,网上数据采掘和网上数据管理等方面的新进展和面临的新挑战。     

基于GIS城市给排水管网信息系统数据结构的研究与实现

本文研究了城市给排水管网信息系统的数据结构及其建立方法,以及在微机上利用C语言进行城市给排水信息系统的设计,实现了部分空间分析功能。