High - Precision Lunar Mapping in Complex Terrain using CNN and Multi-Source DEM’s

Automated identification and annotation of lunar craters is critical for lunar remote sensing and planetary exploration. This research presents a unique approach for locating craters on the Aristarchus Plateau and Schröter Valley using Terrain Mapping Camera-2 (TMC-2) high-resolution Digital Elevation Model (DEM) data from Chandrayaan-II and prior lunar missions. The model achieves more than 84% precision by combining Mask R-CNN with the ResNet-50 backbone, outperforming traditional approaches substantially. This technique discovers previously unknown craters of diameter at least 0.87 km in the given region, improving geological mapping and planning. These findings directly affect future lunar missions, improving landing site accuracy and making identifying possible resources easier. These applications encourage sustainable exploration and resource management, emphasizing the possible impact on future lunar missions.