python 矢量数据转栅格数据

Posted 2020-11-15 ninic

from osgeo import gdal,osr,ogr
#定义投影
sr = osr.SpatialReference(‘LOCAL_CS["arbitrary"]‘)
#在内存中创建一个shape文件的图层，含有两个多边形
source_ds = ogr.GetDriverByName(‘Memory‘).CreateDataSource( ‘shapefile‘ )
source_lyr = source_ds.CreateLayer(‘poly‘, srs=sr, geom_type=ogr.wkbPolygon )
source_lyr.CreateField(ogr.FieldDefn(‘TCODE‘,ogr.OFTReal))
wkt_geom = [‘POLYGON((1020 1030 40,1020 1045 30,1050 1045 20,1050 1030 35,1020 1030 40))‘,
 ‘POLYGON((1010 1046 85,1015 1055 35,1055 1060 26,1054 1048 35,1010 1046 85))‘]
#栅格值
celsius_field_values = [50,200]
for i in range(len(wkt_geom)):
    feat = ogr.Feature(source_lyr.GetLayerDefn())
    feat.SetGeometryDirectly(ogr.Geometry(wkt=wkt_geom[i]))
    feat.SetField(‘TCODE‘, celsius_field_values[i])
    source_lyr.CreateFeature(feat)
#在内存中，创建一个 100*100 大小的1波段的空白图像
#‘’代表不往磁盘上写的话，文件名可以是空
target_ds = gdal.GetDriverByName(‘MEM‘).Create(‘‘, 100, 100, 1, gdal.GDT_Byte )
target_ds.SetGeoTransform( (1000,1,0,1100,0,-1) )
target_ds.SetProjection( sr.ExportToWkt())
#调用栅格化函数。RasterizeLayer函数有四个参数，分别有栅格对象，波段，矢量对象，TCODE的属性值将为栅格值
err = gdal.RasterizeLayer( target_ds, [1], source_lyr,options= ["ATTRIBUTE=TCODE"])
#将内存中的图像，存储到硬盘文件上
gdal.GetDriverByName(‘GTiff‘).CreateCopy(‘rasterized_poly.tif‘, target_ds)
del target_ds
del source_ds