Mars crater detection based on Yolo structure using TIR data

This article aims to detect Mars craters for the geological research and planetary research. The main method is to use deep learning method to identify craters on Mars using Mars TIR images. YOLOv5 is chosen as the basic neural network in this method. To improve the efficiency, the author chose a system that can automatically generate training dataset for YOLO, and the results shows that it is feasible to use neural network to identify the Mars craters.