Importing, spectral resampling, and using field and library spectra

[Elsy]

Hello,

I would appreciate help in the following issue. I have generic files (.txt, .csv) of hyperspectral data from a spectral library and in-situ measurements that I would like to use as reference spectra for supervised classification of a sentinel 2 image (bands spatially resampled to be of the same pixel size).

Is it possible to do that? Can I carry out spectral resampling of the field-spectra in SNAP and then carry out the classification from a “spectral library”? If yes, how? I can’t seem to find any info about the matter.

Thank you

[ABraun]

Not SNAP related but the EnMAP box works with spectral profiles and libaries. I didn’t use this feature so far but had great experiences with the software at all. It is free and open source.

EnMAP

Random Forest NAN at Confidence values below 0.5

thanks for the response. I tried this but the result at these pixels is -1 (NoData). Is this caused by the RF/KD algorithm or the module?

[Elsy]

Thank you for your input. I wonder if I can use SNAP for this.

[ABraun]

have a look at the spectral unmixing module. If you can create a csv with center values for each class and band you could use this as training input and have your scene classified. Worked fine for me.

The CSV must however be exactly as described in the help menu

Weird result from the Spectral Unmixing of a Sentinel2

ss_1.csv (550 Bytes) Thank you for answering. I assumed the file with the signatures is ok, because when I open the Spectral Unmixing window and I select it, all the signatures appear in the small spectrum view inside the window itself. I have anyway attached the file here. For the spectral unmixing I chose the following spectral signatures: Veg_1, Bright_1, Dark_4.

[marpet]

I also wanted to suggest the spectral unmixing.

Unfortunately, we can’t resample high-res spectra. But I take this as idea for further development.

[Elsy]

Thank you. I will give it a try.