X
PrevPrev Go to previous topic
NextNext Go to next topic
Last Post 25 Jun 2015 02:49 PM by  anon
FLAASH module crashing when processing Hyperion product
 3 Replies
Sort:
You are not authorized to post a reply.
Author Messages

anon



New Member


Posts:
New Member


--
25 Jun 2015 02:49 PM
    Hi! I have an Eo-1 Hyperion L1Gst (GeoTIFF) product that I'm trying to apply FLAASH at. I used the Radiometric Calibration tool in order to convert an extent of the imagery to BIL, as stated in the FLAASH module documentation page. When I attempt to perform atmospheric correction using the FLAASH module, it crashes and displays an error. So far I've attempted to use both 242 band BIL and 115 (calibrated only) band BIL, ending with the same result. Also, I thought that it might have something to do with the blank pixels (due to the swath and geometric correction of the product) so I've selected an extent for the BIL that would include only active pixels of the imagery. However, the FLAASH module keeps crashing. I'm quite clueless at this point, since I'm doing everything according to the do directions on the FLAASH page. Any ideas?

    Deleted User



    Basic Member


    Posts:228
    Basic Member


    --
    26 Jun 2015 12:07 PM
    What is the error message that is displayed when FLAASH crashes? That may provide a good clue about what is going wrong. - Peg Exelis VIS

    Deleted User



    New Member


    Posts:
    New Member


    --
    26 Jun 2015 04:17 PM
    I'm having a problem with uploading images so you will have to excuse me for the external link.

    Deleted User



    Basic Member


    Posts:228
    Basic Member


    --
    29 Jun 2015 12:08 PM
    I looked up the error message you are seeing, and it looks like this is due to a problem ENVI is having when importing Hyperion data. The problem was introduced in ENVI 5.0 SP3, which is the version in which a specific reader for Hyperion data was added to ENVI. The problem is related to the fact that in Hyperion data, the VNIR and SWIR focal planes partially overlap in the spectral dimension. So, ENVI needs to re-arrange the bands on input to order by increasing wavelength. Currently ENVI is not doing that, which results in bands between 50-90 having incorrect spectral metadata. This leads to misinterpretation of which bands are "bad" (i.e., contain no data), which leads to the problem you have run into. This problem only occurs with the GeoTIFF Hyperion product. If there is a way to get your dataset in the non-GeoTIFF format (The (e.g., in HDF4), ENVI should handle it correctly. Regards, Peg Exelis VIS
    You are not authorized to post a reply.