Effective classification for crater detection: A case study on Mars

Craters are important geographical features caused by the impacts of meteoroids. Craters have been widely studied because they contain crucial information about the age and geologic formations of planets. This paper discusses an automated crater-detection framework using knowledge discovery and data mining (KDD) process including sampling, feature selection and creation, and supervised learning methods. The framework is evaluated on a real world case study of Mars crater detection. Compared with the existing method, the F detection rate is improved from 0.613 to 0.772 using a Martial site of area 451,562,500 m
²
.