Quantitative Detection of Quartz Sandstone SiO₂ Grade Using Polarized Infrared Absorption Spectroscopy with Convolutional Neural Network Model

Table 4

Assessment of the PLSR prediction results.

Input variable	R^2	RMSE	RPIQ
Original data	0.5948	0.050	2.35
First-order derivative	0.6211	0.049	2.40
Reciprocal logarithm	0.7579	0.034	3.46
MSC	0.8051	0.033	3.56
Original data-PCA	0.6369	0.046	2.55
First-order derivative-PCA	0.6484	0.045	2.61
Reciprocal logarithm-PCA	0.7674	0.034	3.46
MSC-PCA	0.7947	0.033	3.56
Original data-SPA	0.6138	0.047	2.50
First-order derivative-SPA	0.5860	0.038	3.09
Reciprocal logarithm-SPA	0.7712	0.033	3.56
MSC-SPA	0.8437	0.030	3.92