A Study and Implications on the Potential of Satellite Image Spectral to Assess the Iron Ore Grades of Noamundi Iron Deposits Area

Hyperspectral magnetite grade based on characteristic band screening inversion study

“…Many scholars at home and abroad use spectral analysis techniques to analyze the grade and content of minerals, for example: T. Magendran et al used E0-1 Hyperion hyperspectral image data for linear spectral decomposition, which provides the possibility of analyzing the grade of iron ore [2]; Wang Dong et al tested different grades of magnetite and found that the reflectance in the 350-2500 nm band is negatively correlated with the magnetite. The reflectance in the 350-2500 nm band is negatively correlated with magnetite, and the error of the model prediction result is less than 1% through the model established by the law [3] ；SURAJITP S et al used the EO-1 Hyperion data to study the iron ore grade in the iron ore deposit area of Noamundi, and found that in the wavelength region of 752-773 nm, the position of the near-infrared absorption trough moves to the direction of the longer wavelengths with the decreasing of the iron content [4] . XIE H F et al used the hyperspectral data of different grades of copper ore to analyze, and found that with the increase of copper ore grade, its spectral reflectance decreases, and through the correlation analysis, found that the Cu content in the ore is negatively correlated with the spectral reflectance of the ore [5] ; Mengqian Li used hyperspectral detection technology to determine the iron content of iron ore powder, and the relative error of the inversion was 7.26% [6] ; Xu Yanhui et al used spectral analysis technology to spectroscopically analyze magnetite quartzite, and found that the content of TFe and mFe had an exponential negative correlation with the mean value of the reflectance of 850-900 nm [7] ; He Qun et al used spectral analysis technology to carry out the whole iron grade of BIF iron ore model inversion, the inversion error is about 3.43% on average [8] .…”

Hyperspectral magnetite grade based on characteristic band screening inversion study

“…Many scholars at home and abroad use spectral analysis techniques to analyze the grade and content of minerals, for example: T. Magendran et al used E0-1 Hyperion hyperspectral image data for linear spectral decomposition, which provides the possibility of analyzing the grade of iron ore [2]; Wang Dong et al tested different grades of magnetite and found that the reflectance in the 350-2500 nm band is negatively correlated with the magnetite. The reflectance in the 350-2500 nm band is negatively correlated with magnetite, and the error of the model prediction result is less than 1% through the model established by the law [3] ；SURAJITP S et al used the EO-1 Hyperion data to study the iron ore grade in the iron ore deposit area of Noamundi, and found that in the wavelength region of 752-773 nm, the position of the near-infrared absorption trough moves to the direction of the longer wavelengths with the decreasing of the iron content [4] . XIE H F et al used the hyperspectral data of different grades of copper ore to analyze, and found that with the increase of copper ore grade, its spectral reflectance decreases, and through the correlation analysis, found that the Cu content in the ore is negatively correlated with the spectral reflectance of the ore [5] ; Mengqian Li used hyperspectral detection technology to determine the iron content of iron ore powder, and the relative error of the inversion was 7.26% [6] ; Xu Yanhui et al used spectral analysis technology to spectroscopically analyze magnetite quartzite, and found that the content of TFe and mFe had an exponential negative correlation with the mean value of the reflectance of 850-900 nm [7] ; He Qun et al used spectral analysis technology to carry out the whole iron grade of BIF iron ore model inversion, the inversion error is about 3.43% on average [8] .…”