Product Name

Toolbox of utilities to read, process and visualize Raman spectra/hyperspectra

Installation

Run the following to install:

pip install raman_hyperspectra

Methods

Raman file reading :

• WITEK 1D/2D .txt files, WITEK information files;
• RENISHAW 1D/2D .txt/.wdf files.

  Raman spectra/hyperspectra processing

  1. Spikes (cosmics) elimination
     • on hyperspectrum usind PCA or robust PCA
     • on spectra using Outlier detection in large data sets
  2. Linebase correction using :
     • Linear baseline removal;
     • Morphological fiter (tophat);
     • Rubberband ;
     • Asymmetric least square;
     • Improved asymmetric least square ;
     • Asymmetrically reweighted penalized least squares smoothing.
     • local linebase correction
  3. Spectra shift correction using
     • The location of a sharp well referenced peak;
     • The location of the Rayleigh peak.
  4. Noise estimaton (std) on spectra/hyperspectra
  5. Spectra/hyperspectra denoising using:
     • Savitzky-Golay moving smoothing;
     • Noise reduction by wavelet thresholding.
  6. Peaks finding and sorting
  7. Phase imaging
  8. Other available methods:
     • PCA
     • robust PCA
     • NMF
     • VCA (in test)

Usage example

For more examples and usage, please refer to example-raman-hyperspectra.

read and plot a spectra/hyperspectra

import raman_hyperspectra as rhp

file = "C:\\my_modules_Python\\rhp_demo_files\\RENISHAW_1D_Si.txt"
da_sliced, da_sliced_interp, da, da_interp = rhp.read_RAMAN_RENISHAW_0D(file)
da_sliced_interp.plot()

file = "C:\\my_modules_Python\\rhp_demo_files\\Large Area Scan_000_Spec_As cut.txt"
da_sliced, da_sliced_interp, da_full_spectrum, da_full_spectrum_interp = rhp.read_RAMAN_WITEC_2D(file,200,1000)
da_sliced.sel(x=10,y=20).plot()

linebase correction

z_ials, z_base_ials = rhp.flatten_spectra(da,('ials',10_000,  0.01, 0.001))
z_als,z_base_als = rhp.flatten_spectra(da,('als',10_000, 0.001,20))
z_lin, z_base_lin = rhp.flatten_spectra(da,('linear',500,5))
z_top_hat,z_base_top_hat = rhp.flatten_spectra(da,('top hat',factor))

denoising of a spectrum

z_nrwt = rhp.smooth_spectrum (z, ("Savitzky-Golay",sg_length, sg_order) ) 
z_nrwt = rhp.smooth_spectrum (z, ("NRWT", decNum, wnane, mode, ratio_sigma), std ) 

PCA

hp.Raman_PCA(da_denoise, k_PCA = 6)

NMF

rhp.Raman_NMF(da_denoise, k_NMF = 6)

Release History

0.0.1

Meta

- François Bertin– francois.bertin7@wanadoo.fr
- Amal Chabli- 

Distributed under the MIT license

About the authors

- François Bertin now retired was an expert senior in nanocharacterization at CEA-LETI
- Amal Chabli is director of research in nanocharacterization at CEA-LITEN