[R-sig-Geo] Chapter 4 in ASDAR book GeoTiff question

Barry Rowlingson b.rowlingson at lancaster.ac.uk
Mon Jul 1 10:01:48 CEST 2013


On Sat, Jun 29, 2013 at 11:11 PM, Hodgess, Erin <HodgessE at uhd.edu> wrote:
> Dear GEO-phytes
>
> I am working through the ASDAR book (which is TOTALLY amazing), and I have a question about something in Chapter 4, please:  We are looking at a GeoTiff file, and get the following warning message at the end.
>
>> GDALinfo("70042108.tif")
> rows        1200
> columns     1320
> bands       1
> lower left origin.x        174.2
> lower left origin.y        -37.5
> res.x       0.0008333333
> res.y       0.0008333333
> ysign       -1
> oblique.x   0
> oblique.y   0
> driver      GTiff
> projection  +proj=longlat +datum=WGS84 +no_defs
> file        70042108.tif
> apparent band summary:
>    GDType hasNoDataValue NoDataValue blockSize1 blockSize2
> 1 Float32          FALSE           0          1       1320
> apparent band statistics:
>          Bmin       Bmax Bmean Bsd
> 1 -4294967295 4294967295    NA  NA
> Metadata:
> AREA_OR_POINT=Area
> TIFFTAG_RESOLUTIONUNIT=1 (unitless)
> TIFFTAG_SOFTWARE=IMAGINE TIFF Support
> Copyright 1991 - 1999 by ERDAS, Inc. All Rights Reserved
> @(#)$RCSfile: etif.c $ $Revision: 1.11 $ $Date$
> TIFFTAG_XRESOLUTION=1
> TIFFTAG_YRESOLUTION=1
> Warning message:
> statistics not supported by this driver
>>
>
> I was wondering about this since GeoTiff is one of the supported drivers.

 Some geospatial image formats store 'statistics' of their data in
themselves - things like the mean and variance of the data - are kept
in little metadata chunks so that software doesn't have to loop over
the data and compute it whenever needed.

 The warning message is just saying that the GeoTiff driver in GDAL
doesn't support this kind of metadata. If GDAL has to get say, the
mean value of a band, it will have to loop over and compute it rather
than just getting the number from the metadata statistics block.

 Thomas Adams doesn't get this message - either he's got a different
version of the GDAL library that does handle statistics metadata, or
he's got a different version of sp or rgdal (wherever GDALInfo comes
from) that doesn't bother warning you.

 Either way, it doesn't really matter!

Barry



More information about the R-sig-Geo mailing list