Yajnavalkya Bandyopadhyay

In lunar exploration missions, the detection of lunar craters is essential for scientific inquiry, navigation, and terrain analysis. Conventional approaches for identifying craters depend on labor- and time-intensive manual inspection or semi-automated procedures. An effective and precise way to automate this procedure is through the use of deep learning algorithms. In this brief message, we introduce our implementation of the cutting-edge object detection method, YOLOv8, for the purpose of detecting lunar craters. The YOLOv8 architecture, which is well-known for its quickness and precision in object identification tasks, was employed. YOLO (You Only Look Once) predicts bounding boxes and class probabilities for several items in an image at once using a single neural network. We used a dataset of high-resolution lunar surface photos with crater annotations to train the YOLOv8 model.