Automated Detection of Lunar Craters Using Deep Learning

Abstract

In view of the problems of traditional lunar crater detection algorithm (CDA) needs artificial building crater, morphological characteristics, and the extraction precision is not high and the retrieval speed slower. In this paper, based on the Image segmentation convolutional network U-Net model, an automatic crater extraction method based on the improved U-Net model is proposed. The features of craters on the lunar surface were parameterized by deep convolutional neural network, and the residual block and multiple types of dense skip connection were introduced into the convolutional network in multi-scale sampling to further accelerate the convergence speed of the model and improve the detection accuracy, thus realizing the intelligent detection of craters on the lunar surface. Finally, the accuracy of the algorithm is verified by selecting DEM on the lunar surface and craters marking data set. The experimental results show that the Improved U-Net model can quickly and accurately detected craters on the lunar surface.