1.将独立的栅格波段合成为一张图像

#将独立的栅格波段合成一张图像
import os
import numpy as np
from osgeo import gdal

os.chdir(r'E:\桌面文件保存路径\gdal\osgeopy-data\osgeopy-data\Washington')
band1_fn = r'E:\桌面文件保存路径\gdal\osgeopy-data\osgeopy-data\Landsat\Washington\p047r027_7t20000730_z10_nn10.tif'
band2_fn = r'E:\桌面文件保存路径\gdal\osgeopy-data\osgeopy-data\Landsat\Washington\p047r027_7t20000730_z10_nn20.tif'
band3_fn = r'E:\桌面文件保存路径\gdal\osgeopy-data\osgeopy-data\Landsat\Washington\p047r027_7t20000730_z10_nn30.tif'

in_ds = gdal.Open(band1_fn)
in_band = in_ds.GetRasterBand(1)
gtiff_driver = gdal.GetDriverByName('GTiff')
out_ds = gtiff_driver.Create('nat_color.tif', in_band.XSize, in_band.YSize, 3, in_band.DataType)
out_ds.SetProjection(in_ds.GetProjection())
print(in_ds.GetGeoTransform())
out_ds.SetGeoTransform(in_ds.GetGeoTransform())

in_data = in_band.ReadAsArray()
out_band = out_ds.GetRasterBand(3)
out_band.WriteArray(in_data)

in_ds = gdal.Open(band2_fn)
out_band = out_ds.GetRasterBand(2)
out_band.WriteArray(in_ds.ReadAsArray())
#在数据集上调用ReadAsArray时，如果数据集有多个波段，则得到一个三维数组
#如果数据集是单波段，则返回二维数组。GetRasterBand的作用只是返回要查询的波段

in_ds = gdal.Open(band3_fn)
out_band = out_ds.GetRasterBand(1)
out_band.WriteArray(in_ds.ReadAsArray())

out_ds.FlushCache()
#确保数据写入磁盘，刷新缓存
for i in range(1, 4):
    out_ds.GetRasterBand(i).ComputeStatistics(False)
#计算每个波段的统计数据
out_ds.BuildOverviews('average', [2,4,8,16,32])
#创建概视图
del out_ds
#释放数据集

2.band.ReadAsArray(xoff,yoff,win_xsize,win_ysize,buf_xsize,buf_ysize,buf_obj)

#xoff是列起点、win_xsize是读取的列数、buf_xsize是输出数组的列数，如果和前者不同会重采样、buf_obj存放数组类型
#数据类型转换
data = np.empty((3, 6), dtype=float)
band.ReadAsArrray(14000,6000,6,3, buf_obj=data)

data = band.ReadAsArrray(1400,600,6,3).astype(float)

 

3.分块处理栅格

import os
import numpy as np
from osgeo import gdal

os.chdir(r'E:\桌面文件保存路径\gdal\osgeopy-data\osgeopy-data\Landsat\Washington')

in_ds = gdal.Open('')
in_band = in_ds.GetRasterBand(1)
xsize = in_band.XSize
ysize = in_band.YSize

block_xsize, block_ysize = in_band.GetBlockSize()
nodata = in_band.GetNoDataValue()  #获取nodata

out_ds = in_ds.GetDriver().Create('dem_feet.tif', xsize, ysize, 1, in_band.DataType)
out_ds.SetProjection(in_ds.GetProjection())
out_ds.SetGeoTransform(in_ds.GetGeoTransform())
out_band = out_ds.GetRasterBand(1)

for x in range(0, xsize, block_xsize):
    if x+block_xsize < xsize:
        cols = block_xsize
    else:
        cols = xsize-x
    for y in range(0, ysize, block_ysize):
        if y+block_ysize < ysize:
            rows = block_ysize
        else:
            rows = ysize-y
        data = in_band.ReadAsArray(x, y, cols, rows)
        data = np.where(data == nodata, nodata, data*3.28084)
        out_band.WriteArray(data, x, y)

out_band.FlushCache()
out_band.SetNoDataValue(nodata)  #将nodata排除在外
out_band.ComputeStatistics(False)
out_ds.BuildOverviews('average', [2, 4, 8, 16, 32])
del out_ds

4.GeoTransform

#使用现实世界的坐标
#geotransform
#0、3 原点x\y坐标  现实坐标，一般是投影坐标
#1、5 像素的宽度和高度（高度负值）
#2、4 x\y旋转
gt = ds.GetGeoTransform()  #正变换，从图像坐标到现实坐标
inv_gt = gdal.InvGeoTransform(gt)  #逆变换

offsets = gdal.ApplyGeoTransfrom(inv_gt, 465200, 5296000)
#ApplyGeoTransform返回浮点数，若要传递给ReadAsArray需要转为整数
xoff, yoff = map(int, offsets)  #python内置map函数，对数据做映射
value = band.ReadAsArrray(xoff, yoff, 1, 1)[0, 0]

data = band.ReadAsArray()
xoff, yoff = map(int, gdal.ApplyGeoTransform(inv_gt, 465200, 5296000))
value = data[yoff, xoff]  #numpy数组[行，列]，与gdal不同

5.保存图片的子集

#提取并保存图片的子集
import os
from osgeo import gdal

vashon_ulx, vashon_uly = 532000, 5262600
vashon_lrx, vashon_lry = 548500, 5241500

os.chdir(r'E:\桌面文件保存路径\gdal\osgeopy-data\osgeopy-data\Landsat\Washington')
in_ds = gdal.Open('nat_color.tif')
in_gt = in_ds.GetGeoTransform()

inv_gt = gdal.InvGeoTransform(in_gt)
if gdal.VersionInfo()[0] == '1':
    if inv_gt[0] == 1:
        inv_gt = inv_gt[1]
    else:
        raise RuntimeError('Inverse geotransform failed')
elif inv_gt is None:
    raise RuntimeError('Inverse geotransform failed')

offsets_ul = gdal.ApplyGeoTransform(inv_gt, vashon_ulx, vashon_uly)
offsets_lr = gdal.ApplyGeoTransform(inv_gt, vashon_lrx, vashon_lry)
off_ulx, off_uly = map(int, offsets_ul)   #取整得到行列值
off_lrx, off_lry = map(int, offsets_lr)

rows = off_lry - off_uly
columns = off_lrx - off_ulx

gtiff_driver = gdal.GetDriverByName('GTiff')
out_ds = gtiff_driver.Create('vashon2.tif', columns, rows, 3)
out_ds.SetProjection(in_ds.GetProjection())
subset_ulx, subset_uly = gdal.ApplyGeoTransform(in_gt, off_ulx, off_uly)  #这里作正变换是防止原左上角坐标落入像素中间的某个地方
print(subset_ulx)  #计算出来的531995.25\5262624.75是原左上角坐标，正变换后不会落入像素中间。
print(subset_uly)
out_gt = list(in_gt)  #元组转列表，可修改
out_gt[0] = subset_ulx
out_gt[3] = subset_uly
out_ds.SetGeoTransform(out_gt)

for i in range(1, 4):
    in_band = in_ds.GetRasterBand(i)
    out_band = out_ds.GetRasterBand(i)
    data = in_band.ReadAsArray(off_ulx, off_uly, columns, rows)
    out_band.WriteArray(data)

out_ds.FlushCache()
del out_ds

6.图像重采样

import os
import numpy as np
from osgeo import gdal

os.chdir(r'')
in_ds = gdal.Open(r'')
in_band = in_ds.GetRasterBand(1)
out_rows = in_band.YSize*2
out_columns = in_band.XSize*2

gtiff_driver = gdal.GetDriverByName('GTiff')
out_ds = gtiff_driver.Create('band_resampled.tif', out_columns, out_rows)

out_ds.SetProjection(in_ds.GetProjection())
geotransform = list(in_ds.GetGeoTransform())
geotransform[1] /= 2
geotransform[5] /= 2
out_ds.SetGeoTransform(geotransform)

data = in_band.ReadAsArray(buf_xsize=out_columns, buf_ysize=out_rows)
out_band = out_ds.GetRasterBand(1)
out_band.WriteArray(data)

out_band.FlushCache()
out_band.ComputeStatistics(False)
out_ds.BuildOverviews('average', [2, 4, 8, 16, 32, 64])
del out_ds

#重采样为更大的元素
data = np.empty((2, 3), np.int)  #2行3列
band.ReadAsArray(1400, 6000, 6, 4, buf_obj=data)

 7.ReadRaster\WriteRaster读取存储为字节序列

 

#ReadRaster\WriteRaster读取存储为字节序列
ds.ReadRaster(1400, 6000, 2, 2, band_list=[1])
ds.WriteRaster(1400, 6000, 6, 4, data, band_list=[1])

#字节序列ReadRaster
import os
import numpy as np
from osgeo import gdal

in_ds = gdal.Open(r'nat_color.tif')
out_rows = int(in_ds.RasterYSize/2)  #ds直接读取行列数
out_columns = int(in_ds.RasterXSize/2)
num_bands = in_ds.RasterCount  #ds读取波段数目

gtiff_driver = gdal.GetDriverByName('GTiff')
out_ds = gtiff_driver.Create('nat_color_resampled.tif', out_columns, out_rows, num_bands)

out_ds.SetProjection(in_ds.GetProjection())
geotransform = list(in_ds.GetGeoTransform())
geotransform[1] *= 2
geotransform[5] *= 2
out_ds.SetGeoTransform(geotransform)

data = in_ds.ReadRaster(buf_xsize=out_columns, buf_ysize=out_rows)
#利用较小的缓冲来读写数据，重采样
out_ds.WriteRaster(0, 0, out_columns, out_rows, data)
#为输出数据源写入数据
out_ds.FlushCache()
for i in range(num_bands):
    out_ds.GetRasterBand(i+1).ComputeStatistics(False)
out_ds.BuildOverviews('average', [2, 4, 8, 16])
del out_ds

8.子数据集HDF（modis）

 

#子数据集HDF，GDAL的版本需要支持HDF
from osgeo import gdal

ds = gdal.Open(r'‪E:\桌面文件保存路径\gdal\osgeopy-data\osgeopy-data\Modis\MYD13Q1.A2014313.h20v11.005.2014330092746.hdf')
subdatasets = ds.GetSubDatasets()  #返回子数据集列表
print('Number of subdatasets:{}.format(len(subdatasets))')
for sd in subdatasets:
    print('Name:{0}\nDescription:{1}\n'.format(*sd))
ndvi_ds = gdal.Open(subdatasets[0][0])  #第一个[0]读取第一个数据集，第二个则读取当前数据集的名称

 

9.xml保存影像（driver.CreateCopy）

ds = gdal.Open('listing9—6.xml')
gdal.GetDriverByName('PNG').CreateCopy('liberty.png', ds)
#driver.CreateCopy('', ds)用来复制ds