CraterNet : a Fully Convolutional Neural Network for Lunar Crater Detection Based on Remotely Sensed Data

Abstract

In this master thesis, we propose a novel approach to detect lunar craters using new deep learning advances. The architecture of the model employed is a fully convolutional neural network that uses the freely available remotly sensed data from the «Lunar Reconnaissance Orbiter» space probe. In this brief, we discuss the methodology, the choices made and the evaluation of the model with different visual and quantitative results in addition to the source code.