1

Recently, I am trying to export some Landsat-8 surface reflectance data from Google Earth Engine, and the values of these data are around thousands, which can also verified by example code provided by Google Earth Engine (min:0, max:3000 for visualization):

enter image description here

However, it is not very convenient for Google Earth Engine to export a huge size of data, so I found that Google Cloud Storage has public bucket (cloud.google.com/storage/docs/public-datasets/landsat) for Landsat-8 data according to the suggestion from @spatialthoughts.

When I downloaded some Landsat-8 data from Cloud collections and open them in ENVI, I found DN values of these data are about 10 times bigger than those data from Google Earth Engine. Plus, I also found FAQ of Google Earth Engine mentioned that:

enter image description here

Therefore, I am wondering is there any preprocess procedure for Landsat-8 data in Google Earth Engine? I really need many Landsat-8 full images exactly same as Google Earth Engine.

enter image description here enter image description here

1

The visualisation that you see in the code is so that it can better be represented on the map. However the data structure is still the same in both cases.

Here is the structure of 'B4'

data_type: signed int16
type: PixelType
max: 32767
min: -32768
precision: int
  • In fact, the data_type of Landsat-8 data downloaded from Google Cloud Storage is unsigned int 16 (I found this by open images by ENVI), so maybe they are different? – Mr_C Sep 2 '19 at 3:36
  • I also export a Landsat-8 patch from Google Earth Engine and find the image with the same ID in Google Cloud Storage. When I compare these two images DN values in ENVI, they are different and it seems they exist a non-linear relation. – Mr_C Sep 2 '19 at 11:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.