PrevPrev Go to previous topic
NextNext Go to next topic
Last Post 16 Dec 2002 07:17 AM by  Alexandre Leroux
Spectral band maths
 1 Replies
Sort:
You are not authorized to post a reply.
Author Messages

Alexandre Leroux



New Member


Posts:
New Member


--
16 Dec 2002 07:17 AM
    A simple issue where I see no solution with the runtime version of ENVI. What do you think ? (A) In ENVI RT 3.5sp1, if I try to use "Menu/Spectral/Spectral resampling" for data which wavelength is in nm (example: 400 nm to 2500 nm) and try to use ENVI's "Pre-defined filter function", it will not work (error message telling me all bands are out of range) because ENVI is waiting for wavelenght data between 0.4 and 2.5 instead of between 400 nm and 2500 nm. There is no way to specify the unit factor. What is the workaround ? I don't see any way to modify band wavelenght and FWHM values other than going in the Header info and modify each band one by one manually. (B) On a similar issue, in the runtime version of ENVI, I can't find a way to do spectral band maths. What I want to do is quite simple, for one pixel, I'd like to smooth the spectra by averaging each band with previous and next band values (or better: median instead of average). Spectral math allow addition of whole spectra with other pixels, while I'd like to apply a mathematical function within one pixel's spectra. Thanks in any case, Alex :)

    Alexandre Leroux



    New Member


    Posts:
    New Member


    --
    16 Dec 2002 07:17 AM
    This brilliant workaround was provided by Chris Olsen and it works perfectly: redefine the spectral cube orientation so that the spectral dimension is in x or y, and apply a spatial filter in that direction, then redefine the dimensions ( you change bsq to bil, for example, and swap the appropriate sample, lines, bands numbers) Thanks Olsen !
    You are not authorized to post a reply.


    SIGN UP AND STAY INFORMED

    Sign up to receive the latest news, events, technologies, and special offers.

    SIGN ME UP