Home »
Public Forums »
archive »
Georeferencing in ENVI
| Georeferencing in ENVI [message #28127] |
Mon, 26 November 2001 08:53 |
glinka
Messages: 2
Registered: November 2001
|
Junior Member |
|
|
First I would like to thank Chris Jengo and Logan Lindquist for
providing insight to my last question. I am an obvious beginner at
this, so help is always very much appreciated.
I have been using ENVI 3.4 to georeference Landsat 7 TM+ imagery using
1:50 000 scale NTS mapsheets in UTM zone 12 with Nad27 datum. I
rewarped the image in the first degree polynomial using nearest
neighbor resampling (my RMS error was 1.5). Here's the issue. When I
go back and compare the NTS mapsheets to my newly warped image the
areas on the NTS map and on the image don't line up. I checked a few
locations and they're off anywhere from 1 to 9 pixels in either the x
or y direction (I used the cursor location indicator). I would chalk
this up to extrememly poor georeferencing, but when I go back to the
original image and enter the Northing and Easting for the point that
was off by 1 (x) and 9 (y) the RMS error for this point was shows to
be about 1.3 with relatively even errors in x (0.4) and y (0.8). I
would have expected the RMS error for this one point to be large
indicating that my georeferencing was poor, but I'm not sure what this
is telling me. Does anyone know why this is happening? I am apt not to
trust the location idicator, but then I don't know how to enter
training areas for a classification nor will I understand the accuracy
of my data. I'm confused!
Thank you in advance,
Kristen Glinka
|
|
|
|
Current Time: Sat Oct 11 07:32:10 PDT 2025
Total time taken to generate the page: 1.91970 seconds
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
