Quantitative analysis of oxide content in ChemCam spectral data using sparrow search algorithm-kernel extreme learning machine

Laser-induced breakdown spectroscopy (LIBS), as a pivotal technique in the field of planetary exploration, is rapidly advancing its application in precise analysis of target chemical composition. To further enhance the accuracy of LIBS in quantitative analysis, this study employs the least absolute shrinkage and selection operator (LASSO) to extract spectral features as inputs for subsequent quantitative models. Subsequently, the sparrow search algorithm (SSA) is utilized to optimize the regularization coefficient and kernel function parameters of kernel extreme learning machine (KELM), establishing an SSA-KELM quantitative analysis model. This model is applied to predict the ChemCam dataset and compared with other methods to validate its effectiveness. The research findings demonstrate that integrating LIBS technology with LASSO-SSA-KELM method holds great potential for the quantitative analysis of surface elements on Mars, thus providing a reference for rapid and accurate analysis of surface elements on Mars.