Font Size: a A A

Study On Spatial Structure Evolvement Of The Oasis Town Based On High-resolution Remote Sensing Image

Posted on:2019-09-21Degree:MasterType:Thesis
Country:ChinaCandidate:L LiuFull Text:PDF
GTID:2370330545952250Subject:Photogrammetry and Remote Sensing
Abstract/Summary:PDF Full Text Request
In the arid and semi-arid regions of northwest China,the development of oasis cities is vitally important.However,a large number of irrational economic activities make the ecological environment of some oasis cities seriously damaged,furthermore,the ecological environment is keep deteriorating in the early stage of development in Xinjiang.In view of this,the research on the evolution of the spatial pattern of Xinjiang oasis is of great significance to the ecological stability and sustainable development of cities.This paper takes the 125th Regiments of the seventh Agricultural Division of Xinjiang as the research area,and extracts the remote sensing images of Quickbird(2005)and Spot(2015)respectively,by combining remote sensing and GIS related techniques with the theory and algorithm of landscape ecology.The characteristics of spatial pattern evolution are comprehensively analyzed,and the CLUE-S model is used to predict the spatial pattern of 2025.The main contents and results show below:(1)This paper completes the work of high-resolution remote sensing image preprocessing firstly,then uses RAMS and ESP tools synthetically to seek and determine the best segmentation scales of different land use types.Finally,the segmentation of image is realized by using the method of multi-scale segmentation,which is based on the FNEA algorithm.The results show that the segmentation effect is ideal.(2)This paper realized the selection and analysis of object characteristics,and optimized the feature space,which is based on the SEaTH algorithm,then extracted information from high-resolution remote sensing image using support vector machine(SVM)classifier.It shows that the overall classification accuracy and Kappa coefficient of 2005 are 0.86 and 0.83,while the overall classification accuracy and Kappa coefficient of 2015 are 0.9 and 0.87.(3)This paper analyzes the spatial pattern evolution of the research area from the perspective of land use dynamics,land types change,change ratio and transformation direction.The results show that during the period of 2005-2015,the main landscape types and spatial patterns of the 125th Regiments are cultivated land and undeveloped land,and the cultivated area keeps the largest.The undeveloped land changes most greatly,followed by the cultivated land.It shows that the work of reunion field and the land salinization are effective.Some arable land and undeveloped land are transformed into urban and rural construction land,and the increase of urban and rural construction land are considerably obvious,which reflects the process of the urbanization of the 125th Regiments.(4)The evaluation system of landscape pattern is constructed by integrating various landscape indexes on the basis of landscape ecology,and the change of landscape pattern is analyzed at the class level and landscape level respectively.During these ten years,the fragmentation degree of the group field increased and the diversity of the landscape keeps decreasing,while the dominance of the dominant landscape type increases,and the effect of human interference is obvious.(5)This paper takes 2005 and 2015 as the first year of the simulation respectivelyand selects 10 terrain and socioeconomic driving factors,uses the CLUE-S model to simulate the land use and spatial pattern distribution in 2015 and 2025 respectively.Model validation is tested and evaluation by comparing the simulation of land use and the actual land use status map in 2015.The results prove that the kappa coefficient reachs 83.78%,and the simulation result is ideal.It indicates that CLUE-S model has good applicability in the study of land use and spatial pattern evolution.
Keywords/Search Tags:Oasis City, High-resolution Remote Sensing Image, Multiresolution Segmentation, Spatial Structure Evolvement, Land Use Prediction
PDF Full Text Request
Related items