TY - JOUR
T1 - Lunar impact crater identification and age estimation with Chang’E data by deep and transfer learning
AU - Yang, Chen
AU - Zhao, Haishi
AU - Bruzzone, Lorenzo
AU - Benediktsson, Jon Atli
AU - Liang, Yanchun
AU - Liu, Bin
AU - Zeng, Xingguo
AU - Guan, Renchu
AU - Li, Chunlai
AU - Ouyang, Ziyuan
N1 - Funding Information: This study was supported by the CE-1 and CE-2 missions of the CLEP. We thank the team members of the Ground Application and Research System (GRAS), who contributed to data receiving and preprocessing. This work was also supported by the National Natural Science Foundation of China grants 61572228 and 61972174, the Open Funds Project of Key Laboratory of Lunar and Deep Space Exploration LDSE201906, the Science-Technology Development Plan Project of Jilin Province of China grants 20190303006SF and 20190302107GX and the Industrial Innovation Special Funds Project of Jilin Province grants 2019C053-5 and 2019C053-7. The authors are grateful to Dr. Zhiyong Xiao from Sun Yat-sen University for providing helpful comments on a draft of the manuscript and useful analysis of crater populations. The authors thank Stuart J. Robbins for providing the D >1 km database, R. Povilaitis and the LROC team for providing the 5–20 km database, James W. Head for providing the D >20 km database and Goran Salamunićcar, Weiming Cheng, Ari Silburt for providing the automated crater catalogues. The authors would like to acknowledge Dr. Liang Chen from Shantou University for improving the figures in this paper. Publisher Copyright: © 2020, The Author(s).
PY - 2020/12
Y1 - 2020/12
N2 - Impact craters, which can be considered the lunar equivalent of fossils, are the most dominant lunar surface features and record the history of the Solar System. We address the problem of automatic crater detection and age estimation. From initially small numbers of recognized craters and dated craters, i.e., 7895 and 1411, respectively, we progressively identify new craters and estimate their ages with Chang’E data and stratigraphic information by transfer learning using deep neural networks. This results in the identification of 109,956 new craters, which is more than a dozen times greater than the initial number of recognized craters. The formation systems of 18,996 newly detected craters larger than 8 km are estimated. Here, a new lunar crater database for the mid- and low-latitude regions of the Moon is derived and distributed to the planetary community together with the related data analysis.
AB - Impact craters, which can be considered the lunar equivalent of fossils, are the most dominant lunar surface features and record the history of the Solar System. We address the problem of automatic crater detection and age estimation. From initially small numbers of recognized craters and dated craters, i.e., 7895 and 1411, respectively, we progressively identify new craters and estimate their ages with Chang’E data and stratigraphic information by transfer learning using deep neural networks. This results in the identification of 109,956 new craters, which is more than a dozen times greater than the initial number of recognized craters. The formation systems of 18,996 newly detected craters larger than 8 km are estimated. Here, a new lunar crater database for the mid- and low-latitude regions of the Moon is derived and distributed to the planetary community together with the related data analysis.
UR - https://www.scopus.com/pages/publications/85097925208
U2 - 10.1038/s41467-020-20215-y
DO - 10.1038/s41467-020-20215-y
M3 - Article
C2 - 33353954
SN - 2041-1723
VL - 11
JO - Nature Communications
JF - Nature Communications
IS - 1
M1 - 6358
ER -