Automated detection of lunar craters based on Chang'E-1 CCD data

Craters are one of the most significant features of the lunar surface and have been researched widely because they offer us the relative age of the surface unit as well as crucial geological information. Generally, early studies of craters are based on the result of manual extraction from lunar images which requires a great deal of man power and specific knowledge and skills. With the development of computer image processing technology, several automated crater detection algorithms have been proposed, but they have some flaws and limitations requiring further strengthening. This paper puts forth a new approach to image enhancement and feature extraction and shows a new automated crater detection method. Our approach was validated by using Chang'E-1 CCD data, and experiments indicate that the algorithm was able to detect 84.1% in mare images and 87.2% in highland images.