removeBands

Read hyperspectral data into the workspace.

hcube = imhypercube("paviaU.dat");

Inspect the properties of the hypercube object.

hcube

hcube = 
  hypercube with properties:

     ImageDims: "[610x340x103 double]"
    Wavelength: [103×1 double]
      Metadata: [1×1 struct]
     BlockSize: [610 340]

Find the spectral wavelength range of the hyperspectral data cube.

range = [min(hcube.Wavelength) max(hcube.Wavelength)]

range = 1×2

   430   838

Specify the ranges of wavelengths to be remove from the hyperspectral data cube.

wlrange = [410 450; 620 850];

Remove the spectral bands that lie in the specified wavelength ranges. The function returns a new hypercube object without the removed bands.

newhcube = removeBands(hcube,Wavelength=wlrange)

newhcube = 
  hypercube with properties:

     ImageDims: "[610x340x42 double]"
    Wavelength: [42×1 double]
      Metadata: [1×1 struct]
     BlockSize: [610 340]

Plot the original and the new wavelength values.

figure
plot(hcube.Wavelength,"o")
hold on
plot(newhcube.Wavelength,"or")
xlabel("Band Number")
ylabel("Wavelength")
legend("Original Values","New Values",Location="SouthEast")

Read a hyperspectral data into the workspace.

hcube = imhypercube("paviaU.dat")

hcube = 
  hypercube with properties:

     ImageDims: "[610x340x103 double]"
    Wavelength: [103×1 double]
      Metadata: [1×1 struct]
     BlockSize: [610 340]

Compute five spectrally distinct endmembers of the hyperspectral data cube by using the ppi function.

endmembers = fippi(hcube,5);

Determine the 10 most informative bands of the input data cube based on the endmembers spectra.

[~,informativeband] = selectBands(hcube,endmembers,NumberOfBands=10);

Find the band numbers of the noninformative bands of the data cube by using the band numbers of the informative bands.

datacube = gather(hcube);
band = setdiff(1:size(datacube,3),informativeband);

Remove the noninformative bands from the hyperspectral data cube. The function returns a new hypercube object with only the most informative bands.

newhcube = removeBands(hcube,BandNumber=band)

newhcube = 
  hypercube with properties:

     ImageDims: "[610x340x10 double]"
    Wavelength: [10×1 double]
      Metadata: [1×1 struct]
     BlockSize: [610 340]

Plot the original and the new wavelength values.

figure
plot(hcube.Wavelength,"o")
hold on
plot(newhcube.Wavelength,"or")
xlabel("Band Number")
ylabel("Wavelength")
legend("Original Values","New Values",Location="SouthEast")