Font Size: a A A

Scale Effect And Error Analysis Of Vector Data To Raster Conversion With Currentland Use Of Typical Landform

Posted on:2014-04-07Degree:MasterType:Thesis
Country:ChinaCandidate:J X WangFull Text:PDF
GTID:2250330401489382Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
Vector data and raster data are two kinds of basic data structure in GIS. Raster data has obvious advantages in many researches like land evaluation、eco-environment evaluation and soil erosion survey, etc. Therefore higher requirements of area accuracy are put forward after rasterization. Remote sensing image resolution is the same as scale of raster data. The same phenomenon or ground objects in different resolution images also have a scale effect. This study intended to explore under the same scale changes, what scale effect every land type has and analyze area error in different landforms.Rule of maximum area is used in this study. I choose four landforms which are comprehensive landform, plain, hill area and mountainous region from land type polygon of the second national land survey work in Xiangyuan Shanxi. Considering Remote sensing image resolution, I choose lm,5m,10m,15m,20m,25m,30m,40m,50m,60m,100m,250m,500m,1000m as raster scale. Then put four landforms above into rasterization. The main conclusion is made through researching scale effect and analyzing area error:(1) The area error accuracy of every land use type in different landforms is gradually increasing along with scale enlargement. When scale is less than60m, area error accuracy of every type is low. When scale is greater than60m, main types in every landform stand out, and others are ignored.(2) Changes in the same land use type of different landforms are different. Cultivated land in plain and comprehensive landform expands with scale enlargement. Forest and grassland in hill area become more obvious with scale enlargement. The area loss of construction land in mountainous region is the most serious.(3) In combination with remote sensing image resolution, the results show that every land use type in different landforms is very obvious, when resolution is less than100m.When resolution is greater than100m, others are replaced by main types in different landforms. The location and shape of polygon have big differences.(4) In order to ensure that absolute error accuracy of every land use type is lower than10percent, the scale which can be chose are respectively is less than250m,30m,50m,100m in comprehensive landform, plain, hill area and mountainous region.
Keywords/Search Tags:Vector Data, Raster Data, Area Error, Error Accuracy
PDF Full Text Request
Related items