Optimal enhancement of features in digital spectra
JACK W. SULENTIC, Halton C. Arp, Jean J. Lorre · The Astronomical Journal · 1985
The optimization of spectral data processing in astronomy by the use of median digital filters is described. The approach is to use a spatial frequency filtering algorithm to eliminate, in this case, extragalactic background contaminants. The algorithm works in terms of the median of the input at each pixel in relation to a cumulative histogram of all pixels. The cleaned image field then retains otherwise obliterated faint spectral features. Sample optimal spectra for knot C of the M87 jet and the interacting galaxies NGC 5679ab are treated with the algorithm to demonstrate the faint emission lines which are retrievable.