Search

To address the global phenomenon of the salinisation of large land areas, a quantitative inversion model of the salinity of saline soils and soil visible–near-infrared (NIR) spectral data was developed by considering saline soils in Zhenlai County, Jilin Province, China as the research object. The original spectral data were first subjected to Savitzky–Golay (SG) smoothing, multiplicative scattering correction (MSC) pre-processing, and a combined transformation technique. The pre-processed spectral data were then analysed to construct the difference index (DI), ratio index (RI), and normalised difference index (NDI), and the Spearman rank correlation coefficient (r) between these three spectral indices and the salt content in the samples was calculated, while a combined spectral index (r > 0.8) was eventually selected as a sensitive spectral index. Finally, a quantitative inversion model for the salinity of saline soils was developed, and the model's accuracy was evaluated based on partial least squares regression (PLSR), the random forest (RF) algorithm, and the radial basis function (RBF) neural network algorithm. The results indicated that the inversion of soil salt content using the selected combination of spectral indices based on the RBF neural network algorithm was the most effective, with the prediction model yielding an R2 value of 0.950, a root mean square error (RMSE) of 1.014, and a relative percentage deviation (RPD) of 4.479, which suggested a good prediction effect. [ABSTRACT FROM AUTHOR]

• Pretreatment combined algorithms combining multiple mathematical transformations and dimensionality reduction algorithms to comprehensively process raw spectral data. • The advantage of this method is high accuracy". • A method for selecting the optimal Pretreatment combined algorithm is proposed and can accurately predict the accuracy of skarn-type iron ore. Iron ore grade determination is a key step in mining production, and its efficiency has a significant impact on mining efficiency and profitability. At present, the most commonly used method for ore grade determination is chemical analysis, which not only is expensive but also has a long assay cycle and lag effect relative to the ore allocation process and cannot effectively reduce the loss depletion rate of ore mining. Therefore, in situ iron ore grade data determination technology based on visible and near-infrared hyperspectral data analysis is an effective way to solve the above problems, but the poor quality of raw hyperspectral data and low accuracy of the ore grade inversion model are still one of the bottlenecks in the application of this technology. In this paper, 225 ore samples were collected from Hongling skarn-type iron ore in Chifeng, Inner Mongolia Autonomous Region, China. The visible and near-infrared hyperspectral data of the sample set and the grade data obtained from chemical analysis were used as data sources. The raw spectral data were first smoothed, and then the smoothed spectral data were processed using two algorithms, continuum removal (CR) and standard normal variate transform (SNV). Then, two dimensionality reduction algorithms, principal component analysis (PCA) and genetic algorithm (GA), were used to reduce the dimensionality of the spectral data before and after preprocessing, and the data sources were obtained after the four different preprocessing combination algorithms were processed. Finally, several quantitative inversion models were established based on the random forest algorithm and the backpropagation neural network (BPNN) algorithm. The stability, accuracy and credibility of the models were evaluated by four metrics: the coefficient of determination (R2), root mean square error (RMSE), residual prediction deviation(RPD) and mean relative error (MRE). The results showed that the quantitative inversion model (SNV-PCA-BPNN) established based on the BP neural network algorithm using the data processed by SNV and PCA for dimensionality reduction was the best, with an R2 of 0.99, an RMSE of 0.0048, an MRE of 1.10% and a RPD of 37.70, providing an effective method for the high-precision inversion of iron ore grades. [ABSTRACT FROM AUTHOR]

• The spectral data are pre-processed using three dimensionality reduction algorithms. • Optimize the parameters of the dimensionality reduction algorithm. • Establishment of a high-precision grade inversion model for Derni copper deposit. At present, chemical analysis is the main method for copper ore grade verification. Because of its long test cycle and lag effect relative to mining and ore blending processes, this approach cannot effectively reduce losses in mineral resource mining. An effective way to solve this problem is to conduct research on in situ grade determination methods for ore bodies based on visible-near-infrared hyperspectral grade modelling techniques. However, the accuracy of the modelling is seriously reduced by the large redundancy of hyperspectral data and the self-limiting nature of some machine learning methods. Therefore, the accurate selection of dimensionality reduction algorithms and machine learning algorithms determines the accuracy of the inverse model, and the setting of parameters such as dimensionality reduction, number of nearest neighbour points and number of hidden layers in neural networks is crucial. In this paper, 190 ore samples were collected from the open pit of the Derni copper deposit in Dawu town, Guoluo Tibetan Autonomous Prefecture, Qinghai Province, China. First, the visible-near-infrared spectral data for the sample set were obtained with an SVC HR-1024 spectrometer. The copper grade of the sample set was verified by chemical analysis. Three algorithms—the Laplacian eigenmap (LE), local tangent space alignment (LTSA) and local linear embedding (LLE)—were used to reduce the dimensionality of the original hyperspectral data. The original data and the three hyperspectral datasets after dimensionality reduction were used as data sources. Several Derni copper grade inversion models based on two machine learning algorithms, the backpropagation neural network and radial basis function (RBF) neural network, were developed. Finally, the coefficient of determination (R2), mean absolute error (MAE) and root mean square error (RMSE) were used to evaluate the accuracy of each model. The results show that the accuracy of the model built with spectral data processed by the dimension reduction algorithm is higher than that of the model constructed with the original spectral data. Among the above combined models, the LLE-RBF model has the highest inversion accuracy. The MAE of the model is 0.117%, the RMSE is 0.136, and R2 is 0.934. [ABSTRACT FROM AUTHOR]