Comparison And Analysis Of The Inversion Methods For The Contents Of The Lunar Surface Substances Based On Multi-source Remote Sensing Data

Since ancient times,the vast sky has always attracted human exploration and research.The exploration of the vast universe and space enables humans to increase their understanding of the earth and the evolution of the earth and moon.It is of great significance to the progress of human civilization.Since the 1950s,humans have launched multiple probe launches and landing samples on the Moon.All countries in the world have established a lunar exploration program with national characteristics to conduct a series of explorations on the mysterious moon.In a series of probes,spectral and resolution-rich data were obtained,such as Apollo spectral data,Clementine spectral data,LRO Diviner thermal infrared data,a series of data obtained from the Nippon Engineering Co.,Ltd.,Japan's Moon Goddess data,and ESA.Sentinel data,etc.There are numerous types of data,and the methods used to perform the inversion of the composition of the monthly surface materials are also diverse.Different countries have different detection data on lunar exploration.However,in the face of inversion of lunar material components in this research field,the inversion results of different data inversion methods and the consistency of error accuracy are the components of the lunar surface analysis.Have to face the problem.China's orbiting moon project is about to launch sampling satellites.China's Lunar Exploration Project is expected to launch sampling satellites in 2020.The acquisition of moon samples marks the progress of China's lunar exploration technology.How to use the samples brought back by the helium project is good,and the selection of appropriate data for inversion of the material components on the monthly surface is an important issue to be solved.Therefore,in this paper,comprehensive use of domestic and foreign relevant spectral data,Clementine data,Chang'e IIM data,LRO Diviner data,laboratory measured spectral data of the US RELAB spectral library and LSCC sample data,Apollo17 landing area as the experimental area,the area The main elements and mineral components were inverted and analyzed,and the relative errors between the measured values and the inversion values of the samples were obtained,and the existing mainstream spectral data and main inversion model methods were systematically analyzed.Analyze and compare the effectiveness of different data and inversion methods,and define the optimal inversion reference model for different elements.The Rainbow Bay area was used as a test area to obtain the distribution of main elements and minerals in the Rainbow Bay area.The results of Clementine data,Chang'e 1 IIM data and LRO Diviner data are analyzed and fitted to the measured sample data of the monthly table,so as to fit the goodness of the data and the validity of the inversion model.The results show that among the three data,in the case of using the inversion model described in the text,t for the inversion of FeO content,the fitting degree of Clementine data FeO content inversion is the best,R² is 0.924.For TiO₂ content inversion,the CE-1 IIM data TiO₂content inversion value fitting degree is the best,R² equals 0.7997;C For Al₂O₃content inversion,Clementine data has the best effect on Al₂O₃ content inversion,R²is 0.6833.The systematic review of the major inversion model methods for the content of major minerals,pyroxenes and plagioclases on the monthly surface.After research and analysis,for the inversion of pyroxene content,the best-fitness of the inversion value of the pyroxene content in Clementine data is the best,R² is equal to 0.7586;the inversion of the pyroxene content of the IIM data is the second,R² is 0.5264;the Diviner data is inverse of the pyroxene content.The performance was only 0.5074,so it is best to use Clementine data for inversion of pyroxene content.For the inversion of plagioclase content,the inversion value of CE-1 IIM data is larger than the measured value,and the trend line has a certain deviation angle,resulting in R² of only 0.5339;Clementine data content inversion value has a good fit The degree of R²is 0.7171;the goodness of fit of Diviner data inversion values is low,indicating that the inversion effect is not as good as the above two data,R² is only 0.4513.The element inversion model is applied to the rainbow Bay area,and three quantitative inversion results for the main elements of FeO,TiO₂ and Al₂O₃ are obtained.By contrast,the content of elements has obvious zoning.The mean values of FeO and TiO₂ in the three kinds of data are not very different,the average content is around 11.6wt.%,2.4wt.%,and the Al₂O₃ content data differ greatly.The difference between the IIM data and the mean value of the Diviner data is approximately6.4wt.%,indicating that the Diviner data is not high for the inversion of Al₂O₃.Based on spectral band ratio method and multiple regression analysis,thequantitative inversion of plagioclase and augite is carried out,and the inversion model is applied to the rainbow Bay area.The distribution of minerals in rainbow bay can be roughly divided into three parts:cold sea,ring mountain highland and rain sea.The dominant minerals in the cold sea are plagioclase,and the dominant minerals in the highland are also plagioclase,and the dominant mineral in the rain sea area is pyroxene.In the cold sea area,the total content of plagioclase is high,and the high area content is moderate,about 20 wt.%to 30wt.%.Clementine data,IIM data,and LRO Diviner data are 17.7882wt.%,23.529 wt.%,and 23.059wt.%respectively.The peak value of the IIM and Clementine data histogram is about 10wt.%,the mean value is about 8wt.%,while the peak value in the Diviner data is low,only 7.404wt.%,and the mean is also low.It may be because the missing of Diviner data strip is more serious,and it has a certain gap with the actual distribution of the pyroxene on the surface of the moon.By applying the inversion model to the Rainbow Bay area and analyzing the content distribution map and the content histogram,the distribution of minerals in different areas of the Rainbow Bay area was obtained.