GeoTIFF: Unterschied zwischen den Versionen

Aus Geoinformation HSR
Wechseln zu: Navigation, Suche
K (make each HowTo/Tip a subsection)
(add new Tipp "cloud optimized" GeoTIFF)
Zeile 14: Zeile 14:
 
   % gdal_translate -of VRT your.tif work.vrt
 
   % gdal_translate -of VRT your.tif work.vrt
 
   % gdal_translate work.vrt fixed_your.tif
 
   % gdal_translate work.vrt fixed_your.tif
 +
 +
=== "cloud optimized" GeoTIFF ===
 +
 +
==== General information ====
 +
* cogeo.org
 +
* https://trac.osgeo.org/gdal/wiki/CloudOptimizedGeoTIFF
 +
* [https://fosdem.org/2018/schedule/event/geo_gdal/ FOSDEM'18 talk "GDAL Tips and Tricks - GDAL installation, Python usage, and Cloud GeoTIFFs"] by Jeremy Mayeres, 00:09:10 to 00:13:18
 +
 +
==== creation: GeoTIFF to "cloud optimized GeoTIFF ====
 +
 +
(According to above talk, 00:11:14)
 +
 +
Create internal overviews (if not already created)
 +
 +
  gdaladdo -r average in.tif 2 4 8 16
 +
 +
Then
 +
 +
  gdal_translate in.tif out.tif \
 +
    -co TILED=YES \
 +
    -co COPY_SRC_OVERVIEWS=YES \
 +
    -co COMPRESS=LZW
 +
 +
Host result on server supporting HTTP 1.1 Range Requests (Byte Serving). (Amazon S3, Google Cloud Storage, Azure all support this.)
 +
 +
http://www.cogeo.org/developers-guide.html uses <code>-co COMPRESS=DEFLATE</code> instead of <code>-co COMPRESS=LZW</code>. According to https://trac.osgeo.org/gdal/wiki/CloudOptimizedGeoTIFF#Unspecifiedpoints, either can be used.

Version vom 6. Februar 2018, 13:03 Uhr

See also:

Tipps

How to convert Tiff to GeoTIFF

See GDAL.

How to remove the nodata values from a GeoTIFF?

I find many instructions on how to add the values, but none on how to remove them. => There isn't a direct way of accomplishing this. You could extract a work.vrt file, then manually edit it and remove the line(s) with the NoDataValue tags. Then convert back to geotiff:

 % gdal_translate -of VRT your.tif work.vrt
 % gdal_translate work.vrt fixed_your.tif

"cloud optimized" GeoTIFF

General information

creation: GeoTIFF to "cloud optimized GeoTIFF

(According to above talk, 00:11:14)

Create internal overviews (if not already created)

 gdaladdo -r average in.tif 2 4 8 16

Then

 gdal_translate in.tif out.tif \
   -co TILED=YES \
   -co COPY_SRC_OVERVIEWS=YES \
   -co COMPRESS=LZW

Host result on server supporting HTTP 1.1 Range Requests (Byte Serving). (Amazon S3, Google Cloud Storage, Azure all support this.)

http://www.cogeo.org/developers-guide.html uses -co COMPRESS=DEFLATE instead of -co COMPRESS=LZW. According to https://trac.osgeo.org/gdal/wiki/CloudOptimizedGeoTIFF#Unspecifiedpoints, either can be used.