基于白顶帽滤波的伽马能谱寻峰算法OA

In natural gamma-ray spectroscopy logging,the characteristic information of radionuclides is typically presented in the form of peaks.Accurate peak detection in data is a crucial step for precise analysis tasks such as nuclide identification and activity calculation.The application principles of mathematical morphology are thoroughly investigated,and a peak seeking algorithm for gamma-ray spectra based on white top-hat filtering,designed according to the characteristics of morphological operations,is proposed.The algorithm separates the background through morphological top-hat transformation to enhance peak features,combines local maximum detection for preliminary peak identification,and uses peak width thresholds and peak-to-valley ratio thresholds as constraints to screen peaks based on the characteristics of gamma spectral peaks,ultimately achieving peak detection.The algorithm is successfully applied to natural gamma spectral data processing,effectively enhancing peak features and enabling peak detection in gamma spectral data.Using standard spectral line data of U-Th mixtures measured by a LaBr3(Ce)detector,the accuracy of peak positioning is analyzed.Results show that the relative energy error for all peaks in each standard spectrum is less than 2.5%,meeting the measurement requirements for peak positioning accuracy.It provides robust technical support for subsequent nuclide identification and activity calculation,demonstrating significant theoretical importance and practical application value.