Lunar Crater Identification via Deep Learning

TL;DR

This paper demonstrates that convolutional neural networks can effectively identify lunar craters from digital elevation maps, significantly improving detection rates and accuracy over traditional visual methods, and generalize well to other planetary bodies.

Contribution

The study introduces a CNN-based approach for crater detection that outperforms previous methods and is adaptable to different planetary surfaces.

Findings

Recovered 92% of craters from human-annotated datasets.

Nearly doubled the total crater detections, including smaller craters.

Achieved median errors of 11% or less in crater position and size.

Abstract

Crater counting on the Moon and other bodies is crucial to constrain the dynamical history of the Solar System. This has traditionally been done by visual inspection of images, thus limiting the scope, efficiency, and/or accuracy of retrieval. In this paper we demonstrate the viability of using convolutional neural networks (CNNs) to determine the positions and sizes of craters from Lunar digital elevation maps (DEMs). We recover 92% of craters from the human-generated test set and almost double the total number of crater detections. Of these new craters, 15% are smaller in diameter than the minimum crater size in the ground-truth dataset. Our median fractional longitude, latitude and radius errors are 11% or less, representing good agreement with the human-generated datasets. From a manual inspection of 361 new craters we estimate the false positive rate of new craters to be 11%. Moreover, our Moon-trained CNN performs well when tested on DEM images of Mercury, detecting a large fraction of craters in each map. Our results suggest that deep learning will be a useful tool for rapidly and automatically extracting craters on various Solar System bodies. We make our code and data publicly available at https://github.com/silburt/DeepMoon.git and https://doi.org/10.5281/zenodo.1133969 .