Research On Comparative Analysis And Application Of Globeland30 And Vector Data

GlobeLand30 and vector data are two different types of data,which have their own advantages and disadvantages,and they adopt different ways to describe the real world,so they have some differences and similarities to some degree.With Globe Land30 compared with vector data,it can combine the advantages of the two kinds of data,complementing and validating each other.Therefore,it can not only assess the GlobeLand30 and improve its accuracy,but also it can enrich,produce and upgrade the vector data,which has important research significance and application value.This paper focuses on the comparative analysis and application of GlobeLand30 and vector data,and the main work and achievements accomplished are as follows.1.The research background and significance of the comparative analysis of GlobeLand30 and vector data was introduced,and the current research situation of global land cover mapping,vector data production and application,and the comparison between them was analyzed.Based on that,this paper pointed out the problems existing in the current research.2.The GlobeLand30 data products were introduced in detail,and the general characteristics of vector data and raster data and the surface characterization of regular features were analyzed.In addition,the comparative methods based on vector data and raster data were thoroughly discussed,which can provide theory and method for further research on guidance.3.The accuracy assessment method of GlobeLand30 was studied.Using the 1:100000 land data as the reference data,the overall elements of GlobeLand30 in Henan Province was evaluated and compared with GlobCover2009 and MCD12Q1.And the accuracy of GlobeLand30,such as water,residential land and forest land,was evaluated using the methods of spatial similarity,spatial sample and pixel error.4.The method of similarity measurement between Globe Land30 and vector data was studied.This method constructed the complex function to describe the shape by using the characteristics of the border of area entities such as multilevel chord length,arc-height and so on,then a comprehensive similarity measuring model was established by integrating the location,size,direction and shape of the area entities,which was applied to the identification and matching of GlobeLand30 and vector data,and the updating of water element layer was realized.5.The method of using vector data to verify and optimize GlobeLand30 was studied.Based on the data of vector topographic map,the artificial surface,area river and area vegetation elements of GlobeLand30 were validated.Some corresponding knowledge rules were summarized,and a new method based on vector data and multi-knowledge was put forward,which can be used to determine the classification target,and to optimize the correction to improve the classification accuracy of Globe Land30.6.The original system for comparative analysis of Globe Land30 and vector data was developed based on ArcGIS.Apart from some basic functions,it also included some modules of accuracy assessment,similarity measurement,validation and improvement,which can be used to verify the theory and method proposed in the paper.