Automated crater shape retrieval using weakly-supervised deep learning. (arXiv:1906.08826v1 [astro-ph.EP])
<a href="http://arxiv.org/find/astro-ph/1/au:+Ali_Dib_M/0/1/0/all/0/1">Mohamad Ali-Dib</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Menou_K/0/1/0/all/0/1">Kristen Menou</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Zhu_C/0/1/0/all/0/1">Chenchong Zhu</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Hammond_N/0/1/0/all/0/1">Noah Hammond</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Jackson_A/0/1/0/all/0/1">Alan P. Jackson</a>

Crater shape determination is a complex and time consuming task that so far
has evaded automation. We train a state of the art computer vision algorithm to
identify craters on the moon and retrieve their sizes and shapes. The
computational backbone of the model is MaskRCNN, an “instance segmentation”
general framework that detects craters in an image while simultaneously
producing a mask for each crater that traces its outer rim. Our post-processing
pipeline then finds the closest fitting ellipse to these masks, allowing us to
retrieve the crater ellipticities. Our model is able to correctly identify 87%
of known craters in the holdout set, while predicting thousands of additional
craters not present in our training data. Manual validation of a subset of
these craters indicates that a majority of them are real, which we take as an
indicator of the strength of our model in learning to identify craters, despite
incomplete training data. The crater size, ellipticity, and depth distributions
predicted by our model are consistent with human-generated results. The model
allows us to perform a large scale search for differences in crater diameter
and shape distributions between the lunar highlands and maria, and we exclude
any such differences with a high statistical significance.

Crater shape determination is a complex and time consuming task that so far
has evaded automation. We train a state of the art computer vision algorithm to
identify craters on the moon and retrieve their sizes and shapes. The
computational backbone of the model is MaskRCNN, an “instance segmentation”
general framework that detects craters in an image while simultaneously
producing a mask for each crater that traces its outer rim. Our post-processing
pipeline then finds the closest fitting ellipse to these masks, allowing us to
retrieve the crater ellipticities. Our model is able to correctly identify 87%
of known craters in the holdout set, while predicting thousands of additional
craters not present in our training data. Manual validation of a subset of
these craters indicates that a majority of them are real, which we take as an
indicator of the strength of our model in learning to identify craters, despite
incomplete training data. The crater size, ellipticity, and depth distributions
predicted by our model are consistent with human-generated results. The model
allows us to perform a large scale search for differences in crater diameter
and shape distributions between the lunar highlands and maria, and we exclude
any such differences with a high statistical significance.

http://arxiv.org/icons/sfx.gif