[gdal-dev] GDALWARP Chunk Problem (Tiling Effect)

[Heng Feng]

Thanks for your prompt reply, Brian.

I did try GDALWARP command on the same height data without the compression option and the tiling effect still exists, but I will try the option "-et 0.0" and let you know it goes. 

BR,
Chris


Home: www.aircominternational.com/Home-v2 
About Us: www.aircominternational.com/About-Us-v2 
Consultancy: www.aircominternational.com/Consultancy-v2 
Products: www.aircominternational.com/Products-v2 
I-View: www.aircominternational.com/I-VIEW-v2 

Disclaimer: 
This 
e-mail message is confidential and should not be used by, or disclosed to, 
anyone except the addressee. If you receive this message in error, please advise 
us immediately on +44 (0) 1932 442000. Since e-mail transmission is not secure 
or error free, we do not accept responsibility for changes to any e-mail which 
occur after it has been sent. Attachments to this e-mail may contain software 
viruses which could damage your systems. You should therefore virus-check all 
attachments before opening. AIRCOM may monitor incoming and outgoing emails 
through its networks and by responding to this email, you consent to such 
monitoring. The views and opinions expressed in this e-mail are those of the 
author only and not of AIRCOM. We do not intend to enter into any legal 
commitments or contracts by e-mail. Registered Office: Cassini Court, Randalls Research Park, Randalls 
Way, Leatherhead, Surrey, KT22 7TW, UK. AIRCOM International Ltd registered in 
England No. 3052022. VAT number 811 5323 68.
 
Save a tree. Don't print this e-mail unless it's really 
necessary

-----Original Message-----
From: Brian Case [mailto:rush at winkey.org] 
Sent: 01 September 2011 07:27
To: Heng Feng
Cc: gdal-dev at lists.osgeo.org
Subject: Re: [gdal-dev] GDALWARP Chunk Problem (Tiling Effect)

Chris,

Try passing gdalwarp -et 0.0

I am also suspicious of doing compression with the warp. you might want to do that as a second step

Brian

On Wed, 2011-08-31 at 23:16 -0700, heng.feng wrote:
> Hi,
> 
> I have used the (OSGeo4W) GDALWARP command to warp some height data 
> image files (NTv2 transformation). When displaying one warped height 
> data in our own application, there appear to be visible tiles on the 
> image and some misalignments exist between those square tiles (chunks).
> 
> After warping, I did make some conversion on the height data from the 
> original format (ERDAS Imagine) to another format, which can be 
> accepted by our own application. However, I think this tiling effect 
> is not caused by the format conversion, but most likely by something in the GDALWARP command.
> The reason for saying this is that I have done the same format 
> conversion on the original height data without warping and there are 
> no tiling effects at all for the converted height data when shown in 
> our application. As attached, please see the two screenshots for the 
> same area of the height data being displayed in our application. One 
> is with warping while the other is without warping.
> 
> I seem to remember that the GDALWAP command does the chunk warping. 
> Could this be the reason why?
> 
> The GDALWARP command I have been using is something like this:
> 
> gdalwarp --config GDAL_CACHEMAX 250 -wm 250 -co COMPRESS=LZW -co 
> BIGTIFF=YES -co TILED=YES -multi -overwrite -s_srs "+proj=tmerc 
> +lat_0=0 +lon_0=12
> +k=1.000000 +x_0=4500000 +y_0=0 +ellps=bessel +datum=potsdam 
> +nadgrids=c:\height \BETA2007.gsb,null +units=m +wktext" -t_srs 
> +"+proj=utm
> +datum=WGS84 +zone=32" "c:\height\test.img" "c:\height\test.tif"
> 
> I have tried the GDALWARP command on the same height data by taking 
> away all the above command options and got the same tiling effect. 
> Also I have managed to open the warped height map data (tif format) in 
> another GIS application and could see the tiling effects as well 
> although the effect is not as visible as seen in our own application. 
> I think this is because the color interpretation type is gray in that warped height tif file.
> 
> The other thing worth mentioning is: with the same warping and 
> conversion process, lower resolution height data (or much smaller 
> height data) do not seem to have such visible tiling effects.
> 
> Anyway, does anyone here have any clue about this? And is there any 
> way to get rid of the tiling effects in the warped height data?
> 
> Many thanks & BR,
> Chris Feng
> 
> 
> 
> 
> 
> --
> View this message in context: 
> http://osgeo-org.1803224.n2.nabble.com/GDALWARP-Chunk-Problem-Tiling-E
> ffect-tp6749055p6749055.html Sent from the GDAL - Dev mailing list 
> archive at Nabble.com.
> _______________________________________________
> gdal-dev mailing list
> gdal-dev at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/gdal-dev