A background correction method for energy‐dispersive x‐ray fluorescence spectra based on morphological operation

Abstract

AbstractEnergy‐dispersive x‐ray fluorescence spectra often suffer from background caused by primary x‐ray. Therefore, background correction is of great significance that must be performed before subsequent analysis. In this article, a background subtraction method based on morphological operation was proposed. The morphological operation and the smoothing factor could effectively reduce the possibility of background underestimation, especially for a noisy spectrum. The method was verified by simulated and experimental data, and comparisons with other peak‐clipping and penalized least square methods were made. The results showed that the method proposed in this article could get a smooth continuous estimated background with the least root mean square error compared with others. Meanwhile, it can achieve automated background correction, for there is no need to set physical parameters manually. In conclusion, those results indicated that the proposed method could be applied to the rapid in situ x‐ray fluorescence analysis to effectively correct the background.