2016/03/07 by Leonardo Bennun, Bennun, Leonardo
Computer Science · Engineering · Physics and Astronomy · #Advanced X-ray and CT Imaging #Atomic Physics (physics.atom-ph) #FOS: Physical sciences #Geochemistry and Geologic Mapping #X-ray Spectroscopy and Fluorescence Analysis
paper · pdf · doi:10.48550/arxiv.1603.02061
openalex publication_date 2016/03/07 · openalex created_date 2022/10/02 · openalex updated_date 2026/07/28
A new smoothing method for the improvement on the identification and\nquantification of spectral functions based on the previous knowledge of the\nsignals that are expected to be quantified, is presented. These signals are\nused as weighted coefficients in the smoothing algorithm. This smoothing method\nwas conceived to be applied in atomic and nuclear spectroscopies preferably to\nthese techniques where net counts are proportional to acquisition time, such as\nparticle induced X-ray emission (PIXE) and other X-ray fluorescence\nspectroscopic methods, etc. This algorithm, when properly applied, does not\ndistort the form nor the intensity of the signal, so it is well suited for all\nkind of spectroscopic techniques. This method is extremely effective at\nreducing high-frequency noise in the signal much more efficient than a single\nrectangular smooth of the same width. As all of smoothing techniques, the\nproposed method improves the precision of the results, but in this case we\nfound also a systematic improvement on the accuracy of the results. We still\nhave to evaluate the improvement on the quality of the results when this method\nis applied over real experimental results. We expect better characterization of\nthe net area quantification of the peaks, and smaller Detection and\nQuantification Limits.\n We have applied this method to signals that obey Poisson statistics, but with\nthe same ideas and criteria, it could be applied to time series. In a general\ncase, when this algorithm is applied over experimental results, also it would\nbe required that the sought characteristic functions, required for this\nweighted smoothing method, should be obtained from a system with strong\nstability. If the sought signals are not perfectly clean, this method should be\ncarefully applied\n