Influence of Imaging Condition on Matching Point Detection

Matching point detection between a template image and a target image is a key for mosaicking of two or more images. Robustness of a common of template matching algorithm, Sequential Similarity Detection Algorithm (SSDA), was evaluated. Influence of imaging condition, such as solar azimuth, solar elevation, and image azimuth, is also investigated. We particularly consider a production of regional mosaics of the lunar surface from images obtained by an exploration satellite. A Japanese lunar explorer SELENE is going to be launched in the summer of 2007. Thus, synthesized lunar images with various imaging conditions are used as test data in this study. Effects on matching results of features appeared in the templates are also evaluated to know how we choose the location of a template in lunar images to obtain the best matching result.