[GRASS-user] i.rectify - how long is too long

I’m using 1st order and 13 GCPs on an image and i.rectify has been running over 3 hours. Is the process locking up or is this amount of time normal in image rectification? Is i.rectify processing just the area of the image to be rectified, of is it calculating for the whole target mapset area? In other words, should I have applied a mask to the target mapset before running the georectify tool in order to cut down on processing time?

Any info would be greatly appreciated.
Thanks!

Image info:

+----------------------------------------------------------------------------+
| Layer: D-10@PERMANENT Date: Fri Jan 18 08:17:14 2008 |
| Mapset: PERMANENT Login of Creator: me |
| Location: x-y |
| DataBase: /gis/oahu_flightlines |
| Title: ( D-10 ) |
| Timestamp: none |
|----------------------------------------------------------------------------|
| |
| Type of Map: raster Number of Categories: 255 |
| Data Type: CELL |
| Rows: 4171 |
| Columns: 3280 |
| Total Cells: 13680880 |
| Projection: x,y |
| N: 4171 S: 0 Res: 1 |
| E: 3280 W: 0 Res: 1 |
| Range of data: min = 0 max = 255 |
| |
| Data Description: |
| generated by r.in.gdal |
| |
| Comments: |
| r.in.gdal input="/gis/oahu_flightlines/D/D-10\ |
| .jpg" output="D-10" |
| |
+----------------------------------------------------------------------------+

Computer:

Mac Pro 2.66GHz, 2 dualcore Intel Xeon (4 cores total), 4Gb RAM

GRASS:

GRASS 6.3.0RC4 (2008)

Luis E. Menoyo
U.S. Geological Survey
677 Ala Moana Blvd., Suite 415
Honolulu, HI 96813
lemenoyo@usgs.gov
808-587-2428

Very strange... I have a Dual Core 2.8Ghz, 2 Gb RAM laptop and jsut
today I've run i.rectify for a raster with similar dimensions of
yours, 35 GCPs and 2nd order: 15/20 seconds to process it.
GRASS 6.2.3

Giovanni

2008/1/19, Luis E Menoyo <lemenoyo@usgs.gov>:

I'm using 1st order and 13 GCPs on an image and i.rectify has been running
over 3 hours. Is the process locking up or is this amount of time normal in
image rectification? Is i.rectify processing just the area of the image to
be rectified, of is it calculating for the whole target mapset area? In
other words, should I have applied a mask to the target mapset before
running the georectify tool in order to cut down on processing time?

Any info would be greatly appreciated.
Thanks!

Image info:

+----------------------------------------------------------------------------+
| Layer: D-10@PERMANENT Date: Fri Jan 18 08:17:14 2008
|
| Mapset: PERMANENT Login of Creator: me |
| Location: x-y
|
| DataBase: /gis/oahu_flightlines |
| Title: ( D-10 )
|
| Timestamp: none
|
|----------------------------------------------------------------------------|
|
|
| Type of Map: raster Number of Categories: 255
|
| Data Type: CELL
|
| Rows: 4171
|
| Columns: 3280
|
| Total Cells: 13680880
|
| Projection: x,y
|
| N: 4171 S: 0 Res: 1
|
| E: 3280 W: 0 Res: 1
|
| Range of data: min = 0 max = 255
|
|
|
| Data Description:
|
| generated by r.in.gdal
|
|
|
| Comments:
|
| r.in.gdal input="/gis/oahu_flightlines/D/D-10\ |
| .jpg" output="D-10"
|
|
|
+----------------------------------------------------------------------------+

Computer:

Mac Pro 2.66GHz, 2 dualcore Intel Xeon (4 cores total), 4Gb RAM

GRASS:

GRASS 6.3.0RC4 (2008)

Luis E. Menoyo
U.S. Geological Survey
677 Ala Moana Blvd., Suite 415
Honolulu, HI 96813
lemenoyo@usgs.gov
808-587-2428
_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

On 19/01/08 00:25, Luis E Menoyo wrote:

I'm using 1st order and 13 GCPs on an image and i.rectify has been running over 3 hours. Is the process locking up or is this amount of time normal in image rectification? Is i.rectify processing just the area of the image to be rectified, of is it calculating for the whole target mapset area? In other words, should I have applied a mask to the target mapset before running the georectify tool in order to cut down on processing time?

What are your region settings in the target location, especially the resolution ? This can make an enormous difference in terms of calculation time.

Moritz

Any info would be greatly appreciated.
Thanks!

Image info:

+----------------------------------------------------------------------------+

| Layer: D-10@PERMANENT Date: Fri Jan 18 08:17:14 2008 |
| Mapset: PERMANENT Login of Creator: me |
| Location: x-y |
| DataBase: /gis/oahu_flightlines |
| Title: ( D-10 ) |
| Timestamp: none |
|----------------------------------------------------------------------------|

| |
| Type of Map: raster Number of Categories: 255 |
| Data Type: CELL |
| Rows: 4171 |
| Columns: 3280 |
| Total Cells: 13680880 |
| Projection: x,y |
| N: 4171 S: 0 Res: 1 |
| E: 3280 W: 0 Res: 1 |
| Range of data: min = 0 max = 255 |
| |
| Data Description: |
| generated by r.in.gdal |
| |
| Comments: |
| r.in.gdal input="/gis/oahu_flightlines/D/D-10\ |
| .jpg" output="D-10" |
| |
+----------------------------------------------------------------------------+

Computer:

Mac Pro 2.66GHz, 2 dualcore Intel Xeon (4 cores total), 4Gb RAM

GRASS:

GRASS 6.3.0RC4 (2008)

Luis E. Menoyo
U.S. Geological Survey
677 Ala Moana Blvd., Suite 415
Honolulu, HI 96813
lemenoyo@usgs.gov
808-587-2428

------------------------------------------------------------------------

_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

It looks like i.rectify processes the complete target area instead of just the image area to be rectified.

If I want to restrict the calculations to the area of interest, can I use r.mask or should I create a target location of only the area of interest?

I would rather use r.mask, since I could change the mask depending on the area of interest. If that won’t work, then I will create subset locations for each image or flight line.

Thanks to Giovanni and Moritz. Your information is what leads me to believe that I must restrict calculations to the area of interest in order to make better use of computing resources.

x-y location

g.region -p

projection: 0 (x,y)
zone: 0
north: 4171
south: 0
west: 0
east: 3280
nsres: 1
ewres: 1
rows: 4171
cols: 3280
cells: 13680880

Target location

g.region -p

projection: 1 (UTM)
zone: 4
datum: nad83
ellipsoid: grs80
north: 2401345.47908
south: 2350807.60197392
west: 574439.74229936
east: 640164.771271
nsres: 1.2962086
ewres: 1.29619826
rows: 38989
cols: 50706
cells: 1976976234

Luis E. Menoyo
U.S. Geological Survey
677 Ala Moana Blvd., Suite 415
Honolulu, HI 96813
lemenoyo@usgs.gov
808-587-2428

Moritz Lennert <mlennert@club.worldonline.be>

01/19/08 10:32 AM

To

Luis E Menoyo <lemenoyo@usgs.gov>

cc

grass-user@lists.osgeo.org

Subject

Re: [GRASS-user] i.rectify - how long is too long

On 19/01/08 00:25, Luis E Menoyo wrote:
>
> I'm using 1st order and 13 GCPs on an image and i.rectify has been
> running over 3 hours. Is the process locking up or is this amount of
> time normal in image rectification? Is i.rectify processing just the
> area of the image to be rectified, of is it calculating for the whole
> target mapset area? In other words, should I have applied a mask to the
> target mapset before running the georectify tool in order to cut down on
> processing time?

What are your region settings in the target location, especially the
resolution ? This can make an enormous difference in terms of
calculation time.

Moritz

>
> Any info would be greatly appreciated.
> Thanks!
>
> Image info:
>
> +----------------------------------------------------------------------------+
>
> | Layer: D-10@PERMANENT Date: Fri Jan 18 08:17:14
> 2008 |
> | Mapset: PERMANENT Login of Creator: me |
> | Location: x-y
> |
> | DataBase: /gis/oahu_flightlines |
> | Title: ( D-10 )
> |
> | Timestamp: none
> |
> |----------------------------------------------------------------------------|
>
> |
> |
> | Type of Map: raster Number of Categories: 255
> |
> | Data Type: CELL
> |
> | Rows: 4171
> |
> | Columns: 3280
> |
> | Total Cells: 13680880
> |
> | Projection: x,y
> |
> | N: 4171 S: 0 Res: 1
> |
> | E: 3280 W: 0 Res: 1
> |
> | Range of data: min = 0 max = 255
> |
> |
> |
> | Data Description:
> |
> | generated by r.in.gdal
> |
> |
> |
> | Comments:
> |
> | r.in.gdal input="/gis/oahu_flightlines/D/D-10\ |
> | .jpg" output="D-10"
> |
> |
> |
> +----------------------------------------------------------------------------+
>
>
> Computer:
>
> Mac Pro 2.66GHz, 2 dualcore Intel Xeon (4 cores total), 4Gb RAM
>
> GRASS:
>
> GRASS 6.3.0RC4 (2008)
>
>
>
> Luis E. Menoyo
> U.S. Geological Survey
> 677 Ala Moana Blvd., Suite 415
> Honolulu, HI 96813
> lemenoyo@usgs.gov
> 808-587-2428
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> grass-user mailing list
> grass-user@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/grass-user

Luis E Menoyo wrote:

It looks like i.rectify processes the complete target area instead of
just the image area to be rectified.

tip: use v.in.region + v.proj to set up target region / resolution
prior to using r.proj or i.rectify.

Hamish

      ____________________________________________________________________________________
Be a better friend, newshound, and
know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

I started using the display to set the target region and g.region to set the resolution. I also run the georeferencing rutine at low resolution first to more accuratly determine the target area and then run at higher resolution for the final product. The whole process (setting GCPs and running at different resolutions) takes about an hour per aerial photo, which is acceptable. The quality of the final image is very close to that of the source image.

Thanks to everyone for their help.

Luis E. Menoyo
U.S. Geological Survey
677 Ala Moana Blvd., Suite 415
Honolulu, HI 96813
lemenoyo@usgs.gov
808-587-2428

Hamish <hamish_b@yahoo.com>

01/22/08 03:14 PM

To

Luis E Menoyo <lemenoyo@usgs.gov>, grass-user@lists.osgeo.org

cc

Subject

Re: [GRASS-user] i.rectify - how long is too long

Luis E Menoyo wrote:
> It looks like i.rectify processes the complete target area instead of
> just the image area to be rectified.

tip: use v.in.region + v.proj to set up target region / resolution
prior to using r.proj or i.rectify.

Hamish

____________________________________________________________________________________
Be a better friend, newshound, and
know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ