X
PrevPrev Go to previous topic
NextNext Go to next topic
Last Post 08 Feb 2015 11:36 PM by  anon
FLAASH on Hyperion GeoTiff data
 4 Replies
Sort:
You are not authorized to post a reply.
Author Messages

anon



New Member


Posts:
New Member


--
08 Feb 2015 11:36 PM
    Hello!I am trying to perform FLAASH on Geotiff Hyperion Data. I would like to know the steps to be followed. So far, I have done : Input the .tiff files onto ENVI. Also, 1) I read that VNIR and SWIR bands need to be divided by 40 and 80 respectively. Is this value correct? 2) Can Radiometric Calibration be done directly? Is this is step order:1) Input -> calibration by dividing VNIR and SWIR by 40/80 -> Radiometric Calibration -> FLAASH; or 2)Input -> -> Radiometric Calibration -> FLAASH. Which order is to be followed? Thank you.

    MariM



    Veteran Member


    Posts:2396
    Veteran Member


    --
    09 Feb 2015 07:25 AM
    I believe most Hyperion data provided in GeoTIFF have a metadata file such as 'EO1H1770382003203110PY_MTL_L1T.TXT'. If you import using this file, the scale factors should be extracted and the Radiometric Calibration tool (in ENVI 5.1/5.2) should work. This tool lets you output to radiance in the units expected for FLAASH.

    Deleted User



    New Member


    Posts:59
    New Member


    --
    10 Feb 2015 08:36 AM
    We also have a tutorial on working with Hyperion data that may help you: http://www.exelisvis.com/...nalysistutorial.html It covers all the pre-processing steps, including FLAASH. It uses example data in HDF4 format, but the steps should be similar for GeoTIFF format. The process involves (1) opening the metadata file, (2) radiometric calibration, and (3) atmospheric correction. When you read in the metadata file instead of the image file, ENVI will read the scale factors and apply them automatically, so you shouldn't have to apply the scale factors manually. Please let us know if this tutorial helped you, or if you have further questions that it did not address. Thanks!

    Deleted User



    New Member


    Posts:
    New Member


    --
    11 Feb 2015 06:28 AM
    When I input the data from GeoTIFF files and check metadata, the radiance gains are set as .25 and .125 alternatively. However, other sites ask for division by 40 for [1-70] bands and 80 for the rest. Which of these is correct? Also, when input is done from GeoTIFF files, in Spectral Information the Irradiance column is filled as well. What do these values correspond to? The procedure I followed is - input from MTL file-> Radiometric Calibration -> FLAASH. Is this correct?

    Deleted User



    New Member


    Posts:59
    New Member


    --
    11 Feb 2015 09:48 AM
    Normally with Hyperion data in GeoTIFF or HDF4 format, the radiance gains are set to 0.025 for bands 1-70 and 0.0125 for bands 71-242. Later when you calibrate the data, the ENVI Radiometric Calibration tool will multiply a gain value of 0.025 to each pixel in the first band range, which is the same as dividing each pixel value by 40. It will also multiply a gain value of 0.0125 to each pixel in the second band range, which is the same as dividing each pixel by 80. The fact that your gains are 0.25 and 0.125 (versus 0.025 and 0.0125) indicates that the data may have already been scaled by a factor of 10? In any case, ENVI should automatically apply the correct scale factors from the gain values reported in the metadata. The Irradiance field in the metadata viewer reports the solar irradiance values in units of W/(m2 * µm). If the image has metadata for gains, offsets, solar irradiance, sun elevation, and acquisition time, then you can further calibrate the data to top-of-atmosphere reflectance in the Radiometric Calibration tool. The procedure you followed is correct: input from MTL > radiometric calibration > FLAASH.
    You are not authorized to post a reply.