The Feature Extraction And Classification System Of Lunar Surface Impact Craters For Digital Terrain Analysis

The moon is the only satellite of the earth,and it is also a model planet for humans to study the space of the solar system and the history of the universe.The crater is the most typical and most obvious feature of the moon.It studies the evolution of the moon's shape,the history of the meteorite impact,and the universe.The space environment and so on have very important meanings.Since the 1960s,humans have begun to observe,record,and study the shape of the lunar surface by means of telescopes and other observations.Since the 1990s,with the advancement of space exploration technology,the development and utilization of lunar resources has been The goal is to set off a new round of lunar exploration in the international arena.The quantitative description of the scientific indicators of the lunar crater can provide a more reasonable basis for in-depth study of the evolution history of the moon's spherical appearance and the design of the automatic identification algorithm of the crater.The research results can be applied to the exploration of basic research on the moon.Although the existing crater classification scheme has achieved relatively mature results,due to the complexity of the crater's own structure,the comprehensiveness of the formation reasons,and the constraints of experimental methods and data quality,the current research on crater classification is concentrated on The qualitative description of the shape of the impact crater involves more detailed characterization of the inside and outside of the crater,and the quantitative index factors describing the morphological characteristics of the crater are not uniform,and the calculation methods have different degrees of difference.Most of them are quantitative descriptions of individual indicator factors based on subjective experience,and the simulation analysis of the relationship between index factors has yet to be resolved.Based on the high-precision DEM data obtained from LRO and the method of digital terrain analysis,the individual shape,spatial combination,crater development and lunar lithology of the crater are analyzed based on the analysis of the typical crater characteristics of the whole month.In this paper,a comprehensive analysis is carried out to select the indicators that describe the shape of the impact crater from different perspectives,and a new crater classification system for digital terrain analysis is constructed.The crater type is coded by four-level coding combination.On the basis of this,1407 typical impact craters were selected to carry out the grading description of the crater morphological index factors and the quantitative simulation of the relationship between the index factors,and to study the linear fitting relationship of the morphological development of the crater.Firstly,this paper systematically introduces the morphological characteristics,evolution characteristics and crater classification history of the lunar crater.The crater shape of the lunar surface is mostly dominated by a single crater.On the actual lunar surface,the crater exists in different spatial combinations,establishing the spatial relationship(phase separation,intersection,inclusion)and size range from the impact crater.Small,medium,large),developmental features(newborn,young,mature,old)and lunar lithologic components(moon sea basalt,klip rock,highland rock and breccia)impact from different perspectives The characterization model of pit topography is a scientific way to study the classification and evolution of crater.Secondly,eight crater morphological description indexes are selected,and the impact crater is quantitatively and gradually described from the perspectives of the size and shape of the crater.the development of the impact crater and the development of the crater.According to the frequency distribution of the quantitative data,the quantitative indicators are divided.It is level 10,and then it is integrated into three classification intervals according to the classification requirements.The impact craters are classified into six types(discrete craters,clustering pits,clustering pits,degenerate discrete craters,degenerate cluster pits,degenerate cluster pits)The average depth-to-diameter ratio of the fresh impact crater is 0.187,and the average depth-to-diameter ratio of the degraded impact crater is 0.0771.The partial correlation and complex correlation analysis of the eight crater morphological indicators are carried out to determine the depth-to-diameter ratio as the dependent variable.Volume,roundness,body ratio and pit wall slope are the quantitative simulations of the index factors of the independent variables,and the linear fitting formula is obtained.The goodness of fit of the model is 0.637,and the adjusted goodness of fit is 0.639.The fitted model can better express the morphological development characteristics of the impact crater.The complex correlation coefficient of the model is 0.798,and the significance at the 0.05 confidence level is F=492.037>>Fa,indicating that the constructed model is significantly more significant.It can be used to predict and estimate the morphological development characteristics of lunar impact craters.Then,through the screening of the impact crater index factors and the modern digital terrain analysis method,this paper constructs a crater classification scheme suitable for the crater automatic recognition algorithm and quantifies the different types of indicators.The experimental results show that the basic topographical features of the impact crater and its spatial combination structure characteristics can be described more comprehensively and effectively.Based on the morphological characteristics of the crater and its description index,this paper selects the typical lunar crater data as the research sample,systematically studies the morphological characteristics of the crater and the quantitative grading description of the index,and establishes a comprehensive crater shape.The description describes the index classification system and discusses the calculation of the impact crater index.The research content of this paper is of great significance for the comprehensive characterization of crater and the design of automatic recognition algorithm.