Abstract
Due to its ultra-high spectral resolution, hyperspectral remote sensing has the advantages of identifying and quantitatively detecting ground features.In this paper, three methods, including multiple linear regression, partial least square regression and random forest regression, are used to invert the content of chlorite minerals in Liuyuan, Gansu, China. The results show that the R2 of the multiple linear regression model is 0.73 and the RMSE is 0.0565; the R2 of the partial least squares regression model is 0.75 and the RMSE is 0.0535; the R2 of the random forest regression model is 0.80 and the RMSE is 0.0309. Comparative analysis found that the optimal prediction model for chlorite mineral content inversion was random forest regression, and the optimal prediction model was applied to the hyperspectral data of the GF-5 satellite to show the distribution characteristics of chlorite mineral concentration.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call