{"id":1771,"date":"2018-07-02T12:50:01","date_gmt":"2018-07-02T10:50:01","guid":{"rendered":"https:\/\/blogs.fu-berlin.de\/reseda\/?page_id=1771"},"modified":"2018-09-25T11:09:37","modified_gmt":"2018-09-25T09:09:37","slug":"landsat-8-bulk-process","status":"publish","type":"page","link":"https:\/\/blogs.fu-berlin.de\/reseda\/landsat-8-bulk-process\/","title":{"rendered":"Landsat 8 Bulk Preprocessing"},"content":{"rendered":"

Imagine you have 50 Landsat 8 records. Of course, it would be very tedious to modify and start the script 50 times. That’s why there is now an automated solution for any number of data products!<\/p>\n

Again, the prerequisite is that you have the L8 Level-2 datasets downloaded to a folder (“\/media\/sf_exchange\/landsatdata” in our example), in which only all of your Landsat 8 scenes are stored – no other files! Of course you should replace the file path according to your storage location and create a character variable:<\/p>\n

\r\npathToFiles <- "\/media\/sf_exchange\/landsatdata"\r\n\r\ndir(pathToFiles)\r\n## [1] "LC081920232017083001T1-SC20180613163601.tar.gz"\r\n## [2] "LC081930232017060201T1-SC20180613160412.tar.gz"\r\n<\/pre>\n
You can check the files inside pathToFiles <\/span> with the dir()<\/span> function. It should list all your Landsat 8 files. If that is not the case, make sure you did not mess up the file path name.
\nWe can write all the products that exist in the folder in a vector products<\/span> using the list.files()<\/span> function:<\/p>\n
\r\nproducts <- list.files(pathToFiles, full.names = TRUE)\r\n\r\nproducts\r\n## [1] "\/media\/sf_exchange\/landsatdata\/LC081920232017083001T1-SC20180613163601.tar.gz"\r\n## [2] "\/media\/sf_exchange\/landsatdata\/LC081930232017060201T1-SC20180613160412.tar.gz"\r\n<\/pre>\n
To go through all the steps we saw in the previous section above for each of the scenes in products<\/span>, we use a for-loop. Conceptually, the for-loop does the following:<\/p>\n
\r\nfor (i in products) {\r\n  print(i)\r\n  print(\"do all the preprocessing stuff\")\r\n}\r\n## [1] \"\/media\/sf_exchange\/landsatdata\/LC081920232017083001T1-SC20180613163601.tar.gz\"\r\n## [1] \"do all the preprocessing stuff\"\r\n## [1] \"\/media\/sf_exchange\/landsatdata\/LC081930232017060201T1-SC20180613160412.tar.gz\"\r\n## [1] \"do all the preprocessing stuff\"\r\n<\/pre>\n
For all landsat scenes i<\/span> in products<\/span>, it will do all the preprocessing stuff. The variable, or iterator, i<\/span> is just a placeholder, which is sequentially occupied with the file names of the Landsat products.
\nThus, you can just imagine any filename (as a string) every time you see the iterator i.
\nThe remaining steps are then identical to those described above.<\/p>\n
Here is the complete code. Just adjust your pathToFiles <\/span> and run it in R-Studio to preprocess all of your Landsat 8 Level-2 files!<\/p>\n
<\/a><\/p>\n
\r\nlibrary(raster)\r\n\r\npathToFiles <- "\/media\/sf_exchange\/landsatdata"\r\n\r\nproducts <- list.files(pathToFiles, full.names = TRUE)\r\n\r\nfor (i in products) {\r\n  print( paste0("processing: ", i) )\r\n  \r\n  productname <- substr(i, 1, nchar(i) - 7) \r\n\r\n  untar(i, exdir = productname)\r\n  \r\n  productfiles <- list.files(productname, full.names = TRUE)\r\n  \r\n  bands <- c(grep('band1', productfiles, value=TRUE),\r\n             grep('band2', productfiles, value=TRUE),\r\n             grep('band3', productfiles, value=TRUE),\r\n             grep('band4', productfiles, value=TRUE),\r\n             grep('band5', productfiles, value=TRUE),\r\n             grep('band6', productfiles, value=TRUE),\r\n             grep('band7', productfiles, value=TRUE)\r\n            ) \r\n  \r\n  writeRaster(stack(bands), \r\n              paste0(productname, ".tif"), \r\n              format = "GTiff"\r\n              )\r\n  \r\n  unlink(productname, recursive=TRUE)\r\n\r\n  makeOVR <- paste0("gdaladdo -ro ", productname, ".tif 2 4 8 16")\r\n  system( makeOVR )\r\n  }\r\n<\/pre>\n
What does the script for Level-1 data look like? Get an answer and check your knowledge during the following tasks!<\/p>\n

\n