python 用geopandas，rasterio，gdal完成arcgis分区统计功能（zonal）

前面已经有一篇实现分区统计功能的文章了（https://blog.csdn.net/qq_20373723/article/details/111350741），但是核心计算部分是用的numpy，比较难实现majority这个属性，这里用了geopandas，rasterio，gdal地理处理模块组合实现了自动分区统计到表，希望对大家有所帮助
python 3.7.3，geopandas 0.8.2 ，GDAL 3.2.1，rasterio 1.2.0，其中的gdal和rasterio是严格对应的，如果不对应，rasterio是无法安装的。
结果为证：

代码：

import geopandas as pd
import rasterio
from rasterio.plot import show
from rasterstats import zonal_stats
from osgeo import gdal, ogr

ras_path = './fixed_0.tif'
shp_path = './v2.shp'
stats_list = ['min', 'max', 'mean', 'median', 'majority']

ras_driver = rasterio.open(ras_path)
array = ras_driver.read(1)
affine = ras_driver.transform
shp_driver = pd.read_file(shp_path)
zs = zonal_stats(shp_path, array, affine=affine, stats = stats_list)

driver = ogr.GetDriverByName('ESRI Shapefile')
layer_source = driver.Open(shp_path,1)
lyr = layer_source.GetLayer()
defn = lyr.GetLayerDefn()

featureCount = defn.GetFieldCount()
exists_fields = []
for i in range(featureCount):
	field = defn.GetFieldDefn(i)
	field_name = field.GetNameRef()
	exists_fields.append(field_name)

for ele in stats_list:
	if ele in exists_fields:
		pass
	else:
		# cls_name = ogr.FieldDefn(k, ogr.OFTString)
		cls_name = ogr.FieldDefn(ele, ogr.OFTReal)
		# cls_name.SetWidth(64)
		lyr.CreateField(cls_name)

count = 0
feature = lyr.GetNextFeature()
while feature is not None:
	for i in range(featureCount):
		field = defn.GetFieldDefn(i)
		field_name = field.GetNameRef()
		if field_name in stats_list:
			feature.SetField(field_name, zs[count][field_name])
			lyr.SetFeature(feature)
		else:
			pass
	count+=1
	feature = lyr.GetNextFeature()

参考链接：https://pythonhosted.org/rasterstats/manual.html#vector-data-sources

上面的代码可能会遇到异常，第一次运行属性表没有改变，下面的代码应该不会有问题

import time
import geopandas as pd
import rasterio
from rasterio.plot import show
from rasterstats import zonal_stats
from osgeo import gdal, ogr

start = time.time()
ras_path = './cq_test.tif'
shp_path = './cq_test.shp'
# stats_list = ['min', 'max', 'mean', 'median', 'majority']
stats_list = ['majority']

ras_driver = rasterio.open(ras_path)
array = ras_driver.read(1)
affine = ras_driver.transform
shp_driver = pd.read_file(shp_path)
zs = zonal_stats(shp_path, array, affine=affine, stats = stats_list)

driver = ogr.GetDriverByName('ESRI Shapefile')
layer_source = driver.Open(shp_path,1)
lyr = layer_source.GetLayer()
defn = lyr.GetLayerDefn()

featureCount = defn.GetFieldCount()
exists_fields = []
for i in range(featureCount):
	field = defn.GetFieldDefn(i)
	field_name = field.GetNameRef()
	exists_fields.append(field_name)

for ele in stats_list:
	if ele in exists_fields:
		pass
	else:
		# cls_name = ogr.FieldDefn(k, ogr.OFTString)
		cls_name = ogr.FieldDefn(ele, ogr.OFTReal)
		# cls_name.SetWidth(64)
		lyr.CreateField(cls_name)

driver = None

driver = ogr.GetDriverByName('ESRI Shapefile')
layer_source = driver.Open(shp_path,1)
lyr = layer_source.GetLayer()
defn = lyr.GetLayerDefn()

featureCount = defn.GetFieldCount()

count = 0
feature = lyr.GetNextFeature()
while feature is not None:
	for i in range(featureCount):
		field = defn.GetFieldDefn(i)
		field_name = field.GetNameRef()
		if field_name in stats_list:
			feature.SetField(field_name, zs[count][field_name])
			lyr.SetFeature(feature)
		else:
			pass
	count+=1
	feature = lyr.GetNextFeature()

end = time.time()
print((end-start)/3600.0)