Least square smoothing algorithm and gauss decomposition spectral analysis method in spectral gamma ray logging

In the exploration of deep mineral resources, natural gamma spectrometer logging technology is a method which can realizes qualitative and quantitative analysis of deep underground radioactive elements. However, the low energy resolution of natural gamma ray spectrometry logging results in low reliability of distinguishing the radioactive element, therefore the study of spectral analysis method is particularly important. In this paper, the least square smoothing algorithm and the Gauss decomposition method are studied. The results show that the third order polynomial with 5 points smooth window has a good smoothing effect on the spectral data. The spectrum processed by using of the Gauss decomposition method is affected by the separation degree and the peak to height ratio. It is fast and accurate to choose the Gauss decomposition method to solve the spectral data when the separation degree of overlapping peaks is greater than 0.4, and the difference of peak heights are large. This paper provides the basis for the application and selection of the generally analytical method in the actual logging processing, which is of great significance.