snr_derived¶
- specutils.analysis.snr_derived(spectrum, region=None)[source]¶
This function computes the signal to noise ratio DER_SNR following the definition set forth by the Spectral Container Working Group of ST-ECF, MAST and CADC.
- Parameters:
- spectrum
Spectrum1D
The spectrum object overwhich the equivalent width will be calculated.
- region: `~specutils.SpectralRegion`
Region within the spectrum to calculate the SNR.
- spectrum
- Returns:
- snr
Quantity
or list (based on region input) Signal to noise ratio of the spectrum or within the regions
- snr
Notes
The DER_SNR algorithm is an unbiased estimator describing the spectrum as a whole as long as the noise is uncorrelated in wavelength bins spaced two pixels apart, the noise is Normal distributed, for large wavelength regions, the signal over the scale of 5 or more pixels can be approximated by a straight line.
The code and some documentation is derived from
http://www.stecf.org/software/ASTROsoft/DER_SNR/der_snr.py
, and the algorithm itself is documented at https://esahubble.org/static/archives/stecfnewsletters/pdf/hst_stecf_0042.pdf