Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Master → Chine → 2007 → Remote Sensing Dynamic Analyse of Desertification Evolution in Qinghai Hu

Chinese Geology University (Beijing) 2007

Remote Sensing Dynamic Analyse of Desertification Evolution in Qinghai Hu

Zhou Ying Jie

Titre : Remote Sensing Dynamic Analyse of Desertification Evolution in Qinghai Hu

Auteur : Zhou Ying Jie

Grade : Master’s Theses 2007

Université : Chinese Geology University (Beijing)

Land Desertification is a worldwide environmental disasters, is of great significance to study the distribution of desertification status and dynamic trend of desertification control. Remote sensing technology for its large amount of access to information, a wide range of short cycle, true and accurate data for remote sensing of land desertification investigation. The application of remote sensing technology, the classification of desertification survey : computer automatic classification the screen visually interpreted two ways to hook painted classified. In this paper, the Qinghai Lake area, for example, the three land desertification remote sensing survey of the study area. And select the test area, desertification computer automatic classification of research : application of TM data inversion three remote sensing monitoring indicators established Desertification monitoring index system, computer automatic classification of the test area. Image spectral information enhancement methods, research of this paper include : application of a contrast expansion or adjustment, color enhancement, principal component analysis of the test data TM data processing, and analysis of a variety of methods to enhance the effects of desertification. The results expand or adjust the contrast, false color composite, IHS transform certain prominent desertification in the test area, the main component analysis method applied in the test area, the effect is not good. Desertification in the Qinghai Lake area, the application of Landsat Landsat data (MSS, TM, ETM) were three land desertification information extraction, access to the study area three vector results in Fig. The typical spectral characteristics of surface features, and interpret the basic elements and build a remote sensing image interpretation signs of desertification of the study area. Qinghai Lake region three of land desertification situation is analyzed in the past 30 years, the Qinghai Lake area with mild desert of the area in general have greatly reduced, but moderate desertification area, heavy degree of desertification area in the year by year increase The analysis of the natural and human interaction is the reason for the phenomenon of desertification in the region. On the basis of previous studies of desertification indicators, combined with the use of remote sensing data source, select three remote sensing indicators, adjust Vegetation Index (MSAVI) that vegetation coverage (FVC) and improved soil and albedo (Albedo) The inversion of the pilot area of ??remote sensing monitoring indicators. Desertification monitoring index system, decision tree classifier for the classification of computer-based remote sensing monitoring indicators in the test area. Desertification monitoring index system classification accuracy is not high, the test zone is divided into two parts classification : First, the application of multiple indicators for monitoring desertification classification, extracted the part of a higher classification accuracy, precision extracted area, using a different decision tree re-classification results of the classification accuracy has been much higher than have been desertification classification map of the test area. VirtualGIS module of the application ERDAS image The software’s study area desertification three-dimensional stereoscopic display analysis, and application of three-dimensional image can be more precise type of desertification conducted interpreted and can desert the study area to resolve translation carried out interpretation, correct, improve desertification Decoded accuracy.

Mots clés : Desertification Remote sensing technology Information Extraction Image processing Desertification indicators

Présentation (Dissertationtopic)

Page publiée le 8 septembre 2014, mise à jour le 8 novembre 2017