Image Overviews, Tiling and Pyramids - Open Source Geospatial Tools: Applications in Earth Observation - page 92

Depending on the image file size, creating a pyramid can take some time. For the
Landsat 8 image of 7,801 by 7,571 pixels and 12 bands it takes close to 2min to
build a pyramid with four levels and tile size 256 256 pixels. 2 We set the name of

the attribute containing the tile name to location. The output data type is unsigned
integer ( UInt16 ).

×

gdal_retile.py -tileIndex index.shp -tileIndexField location -ot
→ UInt16 -targetDir pyramidDir -levels 4

→

LC82070232013160LGN00.tif

For each level, a directory is created. It contains a vector Shapefile index.shp that
provides an overview of the georeferencing information for the created raster tiles at
that level. The respective filenames are stored in an attribute tileIndexField ,
which has been named “location” in our example. With ogrinfo (see Sect. 2.3 ) , we
can retrieve basic information of the index vector. For instance, at the most detailed
level (level 0) we have:

ogrinfo -so index.shp index

INFO: Open of 'index.shp'
using driver 'ESRI Shapefile' successful.

Layer name: index
Geometry: Polygon
Feature Count: 930
Extent: (438585.000000, 5769885.000000) - (672615.000000,
→ 5997015.000000)
Layer SRS WKT:
PROJCS["WGS_1984_UTM_Zone_29N",
GEOGCS["GCS_WGS_1984",
DATUM["WGS_1984",
SPHEROID["WGS_84",6378137,298.257223563]],
PRIMEM["Greenwich",0],
UNIT["Degree",0.017453292519943295]],
PROJECTION["Transverse_Mercator"],
PARAMETER["latitude_of_origin",0],
PARAMETER["central_meridian",-9],
PARAMETER["scale_factor",0.9996],
PARAMETER["false_easting",500000],
PARAMETER["false_northing",0],
UNIT["Meter",1]]
location: String (254.0)

The vector Shapefile has one attribute (location) of type String. It contains the
filenames of the individual tiles. At this level the image has been divided in 30 by 31
(930 in total) tiles:

LC82070232013160LGN00_01_01.tif
LC82070232013160LGN00_01_02.tif
...
LC82070232013160LGN00_30_31.tif