lijinning / continuumremoval

Schematic Continuum Removal approach for hyperspectral data.

ContinuumRemoval

The Continuum Removal approach, applied by Kokaly and Clark, enhances absorption features in spectroscopic measurements. This approach is extensively used in regression-type problems and attributes

The repository requires 'numpy' and 'matplotlib' library and runs smoothly in 'Python 3.X'.

How it works

The algorithm takes in spectra of samples, the wavelength vector, and the feature regions user is interested in.

Feature regions is a list of either numpy arrays of specified wavelengths or length-two tuples (start,end).

The algorithm is a class with a set of functions as below:

• Continuum_Removal(spectra, wavelength, feature_regions): inputs are spectra, wavelengths vector and specified feauture regions either as a list of tuples or a list of numpy arrays.
• find_near(wl_region): finds nearest wavlengths based on the given wavelength region or point
• R_value(spectra,wl_region) identifies reflectance values based on a given wavelength point or region
• slope_intercept(spectra,wl_region) specifies slope and intercept of a spectra based on a given wavelength region
• cont_rem(): calculates the continuum removed spectra based on inputs
• plot_spectra(self): plots all given spectra
• plot_cr(self): plots all continuum removed curves