如何堆叠 R 中范围几乎完全相同的栅格

Question

我正在尝试堆叠多波段栅格。 它们是通过 ESA Sentinel 1 预处理工具 SNAP 创建的。 每个 tif 文件都必须分层。

我加载了两个 rasterstacks 并尝试堆叠它们：

rs1 <- raster::stack("example/rs1.tif")
rs2 <- raster::stack("example/rs2.tif")
rsstack <- stack(rs1,rs2)

然后我收到以下错误消息：

compareRaster(x) 中的错误：不同程度

栅格堆栈的范围几乎相同：

> rs1
class      : RasterBrick 
dimensions : 2273, 2100, 4773300, 2  (nrow, ncol, ncell, nlayers)
resolution : 8.983153e-05, 8.983153e-05  (x, y)
extent     : 8.183134, 8.37178, 48.49076, 48.69495  (xmin, xmax, ymin, ymax)
crs        : +proj=longlat +datum=WGS84 +no_defs 
source     : memory
names      : S1A_IW_GRDH_1SDV_20180110T053421_20180110T053446_020088_0223E7_69E7_10.1, S1A_IW_GRDH_1SDV_20180110T053421_20180110T053446_020088_0223E7_69E7_10.2 
min values :                                                             1.729380e-07,                                                             1.077101e-06 
max values :                                                                 11.63158,                                                                109.76797 

> rs2
class      : RasterBrick 
dimensions : 2273, 2100, 4773300, 2  (nrow, ncol, ncell, nlayers)
resolution : 8.983153e-05, 8.983153e-05  (x, y)
extent     : 8.183171, 8.371817, 48.49071, 48.6949  (xmin, xmax, ymin, ymax)
crs        : +proj=longlat +datum=WGS84 +no_defs 
source     : memory
names      : S1A_IW_GRDH_1SDV_20180106T171534_20180106T171559_020037_02223E_CE2A_10.1, S1A_IW_GRDH_1SDV_20180106T171534_20180106T171559_020037_02223E_CE2A_10.2 
min values :                                                             8.244981e-08,                                                             5.691331e-06 
max values :                                                                 6.012002,                                                                64.965996 

我该怎么做才能将两者堆叠在一起？ 我怎样才能将范围调整到另一个？

我尝试了什么：我也有一个感兴趣的领域。 所以我尝试将两个堆栈裁剪到那个感兴趣的区域并尝试再次堆叠它们：

shp <- readOGR(dsn=path.expand(example/area.shp)))
shp <- sp::spTransform(shp, CRS(proj4string(rs[[1]]))) 
rs1 <- raster::crop(rs1,shp)
rs2 <- raster::crop(rs2,shp)
rsstack <- stack(rsstack,r2)

compareRaster(x) 中的错误：不同程度

shp
class       : SpatialPolygonsDataFrame 
features    : 16 
extent      : 8.183144, 8.371817, 48.49075, 48.69491  (xmin, xmax, ymin, ymax)
crs         : +proj=longlat +datum=WGS84 +no_defs 
variables   : 14
names       : fid,           Area,    BoundLen,         CentX,         CentY,       AreaIncI,  BoundNotIn,    CentXNotIn,    CentYNotIn,    PointInPol,    PointInPo1, CompactRat, CompactRa1, rast 
min values  :   1,     1440.64063,   168.67161, 3442255.57469, 5377418.66407,     1440.64063,   168.67161, 3442255.57469, 5377418.66407, 3442259.72286,  5377409.4208,    1.14669,    1.14669,    1 
max values  :  16, 76089100.06641, 89693.52095, 3451427.74745, 5393682.39749, 76858585.26953, 76662.07157, 3451427.74745, 5393682.39749, 3451416.96957, 5393687.75063,    2.57399,    2.90065,    1

有没有人有办法解决吗？

Answer 1

我可以看到您的栅格具有相同的坐标系（crs：+proj=longlat +datum=WGS84 +no_defs），只是范围不同。 首先，你可以裁剪它，然后你可以使用从raster package resample function 像

library(raster)

#Crop the raster
rs1 <- raster::crop(rs1,shp)
rs2 <- raster::crop(rs2,shp)

#Conversion of rasters into same extent
rs2_resampled <- resample(rs2, rs1, method='bilinear')

#Stack the rasters
rsstack <- stack(rs1,rs2_resampled)

Answer 2

我想这很容易实现。 咱们试试吧：

library(caret)

setwd("Your File Path")
# Open your reference Raster
Reference_Raster <- list.files(paste(getwd()), pattern = "Reference.tif$")
Reference_Raster <- raster(Reference_Raster)
# Open the other raster
Raster <- list.files(paste(getwd()), pattern = "Raster.tif$")
Raster <- raster(Raster)

# Now Let's Resample
Raster <- resample(Raster , Reference_Raster, method = "ngb")

# Let's Check the extent of both rasters, They are supposed to be the same.
extent(Raster)
extent(Reference_Raster)