Preprocessing Method for Spectra Interpreted by the Power Law Using Segment Regression

Abstract

We report an automated preprocessing method for extracting analysis regions from measured spectral data interpreted by the power law using segment regression. By applying segment regression to spectra, it is possible to obtain the characteristics of the change points and the slopes before and after the change points in the spectrum. By evaluating and judging these features, the analysis region can be automatically extracted. An example of analysis applying this method to a photoelectron yield spectrum with a power law characteristic is shown.