Lunar Impact Crater Detection Based on Yolo V7 Using Muti-source Data

This paper demonstrates that Yolo V7, the latest version of the single-stage neural network Yolo, has good recognition of lunar impact craters on the Lunar CCD data and DEM data provided by the LROC camera which carried by NASA's Lunar Reconnaissance Orbiter (LRO) used for the experiments. The results show that Yolo V7 has good results in detecting lunar impact craters based on two data types. The best identification results are based on CCD data exceeding 70%.