The Study On The Automatical Information Extraction Method Of Water Erosion Desertification Based On The GF-1 Satellite

Water erosion desertification has become one of the world's most noticeable environmental problems.The macroscopic and timeliness of remote sensing satelite data is widely used in the field of extraction monitoring and evaluation of desertification information.However,there are still some deficiencies in previous studies.Firstly,low spatial resolution image data such as Landsat satelite are used as data sources to extract water erosion Desertified Area's refinement degree is not enough.Secondly,the automatic extraction of remote sensing information of water erosion desertification is less.Based on this,with the popularization of the domestic satel ite service,this paper selects the Yupao river Xingfu Vilage-Jinsha river entrance section at the junction of Chuxiong Yi Autonomous Prefecture,Dali Bai Autonomous Prefecture and Lijiang City,where the water erosion desertification is more developed,as the study area and chooses the satellite data of high division?GF-1?.The method of extracting automatic information of water erosion and desertification from GF-1 remote Sensing satelite data is studied in detail.This article has achieved innovative results as follows:1.Experiment optimizes the modified Soil Regulation Vegetation Index with derivative variance texture?Var_MSAVI?as the best method to extract vegetation cover information of water erosion desertification factor.Considering vegetation cover information as one of the important indexes of water erosion desertification,using six commonly used vegetation index models of RVI,NDVI,SAVI,MSAVI,EVI and GEVI,the comparative analysis of the derived texture vegetation index model based on the spatial information of the texture feature and the image element method is carried out respectively.Based on the objective evaluation of the isometric distribution method,the modified Soil Regulation Vegetation Index with derivative variance texture?Var_MSAVI?correction is about 0.5%-4%higher than other vegetation indices,and finally optimizes Var_MSAVI as the best method for extracting vegetation cover information in the study area.2.Water erosion desertification automatic extraction remote sensing index model consisting of slope,valey density,terrain relief,surface cutting depth,and vegetation coverage was constructed.The process of Water erosion desertification automatic extraction was optimized.?1?Systematic analysis of water erosion desertification impact factors can be reflected by remote sensing.In the past,three indicators of water erosion desertification,such as slope,valey density and vegetation coverage,are used to increase the surface cutting depth and topographic relief index.The automatic extraction of five index factors was explored,and a five-factor scale consisting of slope,valey density,terrain relief,surface cutting depth,and vegetation coverage suitable for the entrance of the Xingfu Vilage-Jinsha River entrance of the Yupao River was constructed.Water erosion desertification automatic extraction remote sensing index model:Y_j=?_i=1⁵w_ix_j?j=1,2,…,n?Among them,Y_j is the degree of water erosion desertification of the j unit,i is the impact factor of water erosion desertification,W_i is the weight of the i impact factor,X_j is the value of the i water erosion desertification impact factor in the j unit,n is The number of units.Based on the objective evaluation of confusion matrix and field verification the correct rate of field validation is 88.54%.?2?According to lots of experimental analysis,the method of water erosion desertification automatic extraction process of the entrance of the Xingfu Vilage-Jinsha River entrance of the Yupao River is summarized.The process of water erosion desertification automatic extraction,the optimization parameters of all factor which influence water erosion desertification are set as follows:slope analysis adopts 3D analyst tools of ArcGIS10.2;Valey density adopts hydrology module,it's unit parameter of automatically extracts is 1770m;Terrain relief used spatial analyst tools of ArcGIS10.2,19×19 is the best analysis window for terrain relief automatically extract;Valey density is applied to hydrology module and zonal statistics tools of ArcGIS10.2,the best scale unit for automatic extraction of valey density is a small watershed that takes the valley as the node of river network;vegetation coverage are used band math and texture tools of ENVI5.3.The modified Soil Regulation Vegetation Index with derivative variance texture?Var_MSAVI?is the best method whose step size is 1,angle is 135°,analysis window is 11×11.