Lunar titanium characterization is an important goal of the China Lunar Exploration Program. We suggest a method to determine the lunar titanium abundance using Chang'E-1 IIM (Interference Imaging Spectrometer) im...Lunar titanium characterization is an important goal of the China Lunar Exploration Program. We suggest a method to determine the lunar titanium abundance using Chang'E-1 IIM (Interference Imaging Spectrometer) imagery. Using samples from Apollo and Luna landing sites, the method firstly establishes the spectral parameters that possess good non-linear correlations with lunar titanium abundance. Secondly, the method estimates lunar titanium abundance using a DT-SVM (Decision Tree Method C5.0-Support Vector Machine) method. Namely, according to the established spectral parameters, it uses the C5.0 algorithm to classify the titanium abundance into the 4 classes of very low, low, intermediate and high. Then, in terms of the spectral parameters and the corresponding classes, it employs the SVM to estimate the titanium abundance. The method makes good use of hyperspectral information, analyzes the nonlinear correlations between spectral characteristics of lunar soils and the composition parameter, and well determines the titanium abundance. Validated by the Apollo and Luna station samples, the RMSE (root mean square error) is 0.72wt% TiO2 and the correlation coefficient of the measured and predicted values is 97.29%. So, the method proposed in this paper has a good predictive capability for TiO2 abundance on the lunar surface. The maps of TiO2 content in the partial region of Sinus Iridium, the Apollo 17 landing site and the Apollo 16 landing site are constructed by our method. This paper demonstrates the potential of IIM data for the investigation of lunar surface chemistry and mineralogy.展开更多
基金supported by the National Natural Science Foundation of China (Grant No. 40902099)the Fundamental Research Funds for the Central Universities,China University of Geosciences (Wuhan) (Grant No. CUG100702)
文摘Lunar titanium characterization is an important goal of the China Lunar Exploration Program. We suggest a method to determine the lunar titanium abundance using Chang'E-1 IIM (Interference Imaging Spectrometer) imagery. Using samples from Apollo and Luna landing sites, the method firstly establishes the spectral parameters that possess good non-linear correlations with lunar titanium abundance. Secondly, the method estimates lunar titanium abundance using a DT-SVM (Decision Tree Method C5.0-Support Vector Machine) method. Namely, according to the established spectral parameters, it uses the C5.0 algorithm to classify the titanium abundance into the 4 classes of very low, low, intermediate and high. Then, in terms of the spectral parameters and the corresponding classes, it employs the SVM to estimate the titanium abundance. The method makes good use of hyperspectral information, analyzes the nonlinear correlations between spectral characteristics of lunar soils and the composition parameter, and well determines the titanium abundance. Validated by the Apollo and Luna station samples, the RMSE (root mean square error) is 0.72wt% TiO2 and the correlation coefficient of the measured and predicted values is 97.29%. So, the method proposed in this paper has a good predictive capability for TiO2 abundance on the lunar surface. The maps of TiO2 content in the partial region of Sinus Iridium, the Apollo 17 landing site and the Apollo 16 landing site are constructed by our method. This paper demonstrates the potential of IIM data for the investigation of lunar surface chemistry and mineralogy.
