About

I am a Earth observation scientist, working at Stockholm University. I spend an inordinate amount of time looking at glaciers.

Posts will be occasional and irregular, though I will aim for at least one post per week. Feel free to leave comments or questions.

7 Responses to About

  1. Greg says:

    I am wondering if you have any recommendation for my following issue: I am using envi 4.7 and am doing temporal spatial change for one of my class projects using landsat. My problem is that the two regions cover slightly different regions and for some reason have different numbers of lines. Anyways, I dont want to resize the images to match each other because then the pixels wont be square. In the end I want to apply a mask to both images, but since they dont have the same lines I cant. Any tips? Thanks!

    -Greg (University of Victoria, BC)

    • blamannen says:

      Greg,
      you can resize the data and select a new pixel size to make the resulting images square (under ‘Basic Tools’). Or, classify both images and compare classes. In this case you should use a supervised classification and save the training areas (Regions of Interest in ENVI-speak) as a vector file (‘Export EVF’) so that you can use training areas common to both images. Then you can compare the resulting classification. If it is vegetation you are mapping, you might want to try using NDVI or another vegetation index. The advantage with this approach is that there should be less effect from differences in solar illumination (sun angle, atmospheric effects etc), moisture on the ground and in the air, and other second-order effects. If comparing NDVI values it is probably worth binning them into classes (e.g. <1.200 = no veg., 1.201-1.400 = scrub etc).
      Good luck!

  2. conrad says:

    Hi Geonotes,
    Any chance you could do a bit of a walkthrough for the target detection wizard in Envi 5 including spectral library resampling for landsat and aster? I’m having a terrible time getting it to work. I’m trying to make mineral maps using the USGS library but all images keep coming out the same…
    Conrad

  3. kemilla says:

    Hi Geonotes,

    I am trying to use RAT for RADARSAT-2 (RS2) data processing and analysis but I’m running into an issue importing RS2 data. I see in your post from 2011 that you were successful in importing RS2 data. RAT appears to be looking for a tif file instead of the product.xml. I converted the data to both multi-channel tif files (I am working with quad pol data – FQ) and single channel tif files but am unable to import the data for use in RAT. The program keeps crashing and saying that I’ve found a bug in RAT. Do you have any tips or hints?

    • blamannen says:

      Hi,
      I am having the same problem with both Radarsat-2 Quad-pol and ALOS PALSAR Quad-pol. I have tried RAT 0.20 and 0.19 and that does not work either. but I have used it successfully in the recent past. I wonder if it is a Win7 or Win64 bit problem. I’ll test another machine and get back to you.

  4. kemilla says:

    Thanks – I appreciate it. I am using Windows 7 64 bit version….I don’t have access to a 32 bit machine so I can’t really test if that is the issue.

    • blamannen says:

      Hi Kemilla,
      it appears that the problem is using RAT on a 64-bit architecture. This is compounded by not being able to install a 32-bit IDL on a 64-bit Win system; so that’s not a viable work-around. The only thing I can suggest is trying to get hold of a 32-bit machine.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s