BATCH CO-REGISTRATION OF MARS HIGH-RESOLUTION IMAGES TO HRSC MC11-E MOSAIC

Abstract

Abstract. Four NASA missions over the last forty years with onboard instruments for high-resolution orbital imaging have achieved both global coverage (with 6m CTX, 20m THEMIS-VIS and >8m Viking Orbiter cameras) as well as imaging with very high resolution in specific regions of interest (e.g. 25cm HiRISE and ≈1.5-12m MOC-NA cameras). Overall, this set of cameras have acquired more than 400,000 high-quality images of Mars with resolution between 25cm/pixel and 100m/pixel (Sidiropoulos and Muller, 2015). On the other hand, ESA has sent the only high-resolution stereo photogrammetric camera around Mars, HRSC onboard the Mars Express spacecraft, which has been mapping the Martian surface since 2004 with a resolution of 12.5 m/pixel (Jaumann et al., 2015). Initially the raw images are combined through an elaborate photogrammetric process to get (single-strip) 3D products (i.e. digital terrain models (DTMs) and derived orthorectified images (ORIs)). However, recently the processing chain has changed, and the single-strip product release was temporarily halted to be replaced by the production and release of mosaics of Mars quadrangles. The first product of this kind is the mosaic for the East part of quadrangle MC11 (i.e. the MC11-E mosaic), a product with 12.5 metres per pixel resolution in the panchromatic image and 50 metres per pixel resolution in the corresponding DTM (Gwinner et al., 2015). Such a product provides an excellent basemap to co-register and orthorectify all NASA high-resolution (≤100m/pixel) orbital images. The need for this co-registration to HRSC comes from their poor areo-referencing, which often leads to large deviations (reaching up to several kilometres) between the area they are supposed to image and the area they are actually imaging. After co-registration, all products are projected onto an common 3D coordinate system, which allows an examination of dynamic features of Mars through the changes that happen on its surface. In this work, we present the results of the batch coregistration of all NASA high-resolution orbiter images of MC11-E, i.e. almost 8,000 images in total. This task was conducted with an in-house pipeline which was modified in order to handle the different parameters of the mosaic in comparison to single-strip HRSC products and to process the large input data volumes within a realistic time. An outline of the processing pipeline is given, along with examples of co-registered images and statistics of the co-registration performance. We demonstrate how such a time series representation of the surface will open up new areas for exploration and understanding of the Martian surface.