Abstract

A new upper-bound spectrum (UBS) method for removal of cosmic spike artifacts from spectra or images collected using a charge-coupled device (CCD) detector is proposed. This algorithm, which is shown to outperform previous methods, relies on an upper-bound spectrum, derived from scaled copies of consecutively collected spectra, which serves as a threshold for the detection of suspected cosmic spikes. Detected spikes are removed by replacement with the corresponding points in other spectra. Thus, unlike other commonly used methods, the UBS algorithm requires no smoothing or noise filtering and more reliably removes cosmic spikes of all magnitudes while introducing far less (if any) spectral distortion. The UBS method is tested using both synthetic and experimental (gypsum and gypsum/hematite mixture) spectra containing variable background (fluorescence), noise, and cosmic spike interference. The UBS method may in rare instances mistakenly identify spectral noise or photo-induced changes in band intensity (or shape) as cosmic spikes. However, as demonstrated through the use of both synthetic and experimental examples, such misidentifications produce little or no spectral distortion or artifacts in the resulting cosmic-spike-free output spectra.

An upper-bound metric for characterizing spectral and spatial coregistration errors in spectral imaging

Torbjørn Skauli
Opt. Express 20(2) 918-933 (2012)

Parallel Monte Carlo simulation algorithm for the spectral reflectance and transmittance of the wind-generated bubble layers in the upper ocean using CUDA

Yu Yang and Lixin Guo
Opt. Express 28(22) 33538-33555 (2020)

Upper bounds on focusing efficiency

Kurt Schab, Lukas Jelinek, Miloslav Capek, and Mats Gustafsson
Opt. Express 30(25) 45705-45723 (2022)

Blind spectral deconvolution algorithm for Raman spectrum with Poisson noise

Hai Liu, Zhaoli Zhang, Jianwen Sun, and Sanya Liu
Photon. Res. 2(6) 168-171 (2014)

Baseline correction for Raman spectra using a spectral estimation-based asymmetrically reweighted penalized least squares method

Yixin Guo, Weiqi Jin, Weilin Wang, Yuqing He, and Su Qiu
Appl. Opt. 62(18) 4766-4776 (2023)