by

Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Aug 02 19:25

    tboggs on 0.22.1

    (compare)

  • Aug 02 19:18

    tboggs on master

    Increase version to 0.22.1 (compare)

  • Aug 02 19:11

    tboggs on master

    Fix package_data to include spe… (compare)

  • Aug 01 20:50

    tboggs on 0.22

    (compare)

  • Aug 01 20:43

    tboggs on master

    Increase version to 0.22 (compare)

  • Aug 01 20:30
    tboggs closed #113
  • Aug 01 20:30
    tboggs commented #113
  • Aug 01 20:21

    tboggs on master

    Loosen remaining SMACC unit tes… (compare)

  • Aug 01 20:13

    tboggs on master

    Loosen additional SMACC unit te… (compare)

  • Aug 01 20:03

    tboggs on master

    Loosen tolerance on test_smacc_… Fix imports in orthogonalize an… (compare)

  • Jul 24 12:44
    tboggs opened #113
  • Jul 07 13:36

    tboggs on master

    Implement SMACC algorithm (#112) (compare)

  • Jul 07 13:36
    tboggs closed #112
  • Jul 07 13:36
    tboggs commented #112
  • Jul 07 13:32

    tboggs on master

    Fix USGS database for python 2.… (compare)

  • Jul 07 13:32
    tboggs closed #111
  • Jul 07 13:32

    tboggs on master

    Improve docs and error message … (compare)

  • Jul 07 13:32
    tboggs closed #108
  • Jul 03 18:06
    kormang commented #112
  • Jul 03 17:48
    kormang opened #112
Thomas Boggs
@tboggs
Testing.....
kidpixo
@kidpixo
Working....
Thomas Boggs
@tboggs
Great. I'll add gitter chat badge to the README soon.
kidpixo
@kidpixo
Nice :-D
Thomas Boggs
@tboggs
Good news: Gitter created an automatic pull request to add the badge to README.rst.
kidpixo
@kidpixo
sounds good
Thomas Boggs
@tboggs
Bad news: Gitters pull request not only doesn't display but broke display of Travis-CI badge.
kidpixo
@kidpixo
Hey, thank you for spectralpython!
sounds less good....
Thomas Boggs
@tboggs
You're welcome and thanks for your contributions!
Hmm. Maybe my browser or connection is wonky. Do the badges on the repo page display for you?
Thomas Boggs
@tboggs
Nevermind. It appears to have been a browser cache issues. Badges display now.
wwwald
@wwwald
hi all
is this room still alive?
if so, can it be used for user questions?
justinkotzebue
@justinkotzebue
I was wondering the same?
kidpixo
@kidpixo
good question
Thomas Boggs
@tboggs
What's the question?
This forum isn't used much but feel free to ask. Alternately, you can open an issue.
justinkotzebue
@justinkotzebue
Hi everyone, My question is how to inverse transform a denoised multispectral band stack? My workflow is :
1)
signal = spectral.calc_stats(data)
noise = spectral.noise_from_diffs(data)
mnfr = spectral.mnf(signal, noise)
denoise_trans = mnfr.get_denoising_transform(num=4), (data is my multispectral stack)
2)
Some image processing on 'denoise_trans'
Now I woul like to inverse transform "denoise_trans" to reflectance values (original format)
Thanks :)
Thomas Boggs
@tboggs
If you call the denoise method, specifying the min snr, then the transformed data will still be in reflectance units.
justinkotzebue
@justinkotzebue
Thanks @tboggs I have been runing that apporach which gives really nice results. But what I'm trying is noise removel by 'performing a forward transform' using these steps:
1) first rotaion to noise whitened data (denoise_trans = mnfr.get_denoising_transform(num=4))
2) apply some image smoothing of noise whitened data
3) rotate back to original dimensions and reflectance units
Thomas Boggs
@tboggs
If you look at the source code for the get_denoising_transform method, you can easily call parts of that algorithm piece-wise, if you want to do some special processing before rotating back to the original feature space.
ShivakumarBR
@ShivakumarBR
hi, Spectral lib is pip installed with 0.19 version. But unable to access mnf(),calc_stats() etc. functions in spyder 3.6? Please help me.
Mehdi saberioon
@saberioon_twitter
Hi everyone, I would like to do the segmentation based on k-means clustering, K-means clustering gave me classmaps and centers , but I don't know how to use them for masking, Any help?
Thomas Boggs
@tboggs
@saberioon_twitter , sorry I haven't been getting notifications for this group for some reason. The first value returned from kmeans is the class_map, which you can pass on to the create_training_classes function.