Rigorous Photogrammetric Processing of HiRISE Stereo Imagery for Mars Topographic Mapping.

Rongxing Li,Juwon Hwangbo,Yunhang Chen,Kaichang Di

DOI: https://doi.org/10.1109/tgrs.2011.2107522

IF: 8.2

2011-01-01

IEEE Transactions on Geoscience and Remote Sensing

Abstract:High-resolution (submeter) orbital imagers have opened up new possibilities for Mars topographic mapping with unprecedented precision. While the typical sensor model for Martian orbiters has been the linear-array charge-coupled device (CCD), the High-Resolution Imaging Science Experiment (HiRISE) instrument is based on a more complicated structure involving a combination of 14 separate linear-array CCDs. To take full advantage of this high-resolution capability without compromising imaging geometry, we have developed a rigorous photogrammetric model for HiRISE stereo image processing in which third-order polynomials are used to model the change in exterior-orientation parameters over time. A coarse-to-fine hierarchical matching approach was developed, and its performance was evaluated based on manually matched image points and manually measured features for a test area at the Mars Exploration Rover Spirit landing site. Using automatically selected tie points, we performed bundle adjustment (BA) to improve the accuracy of image pointing data and remove or reduce inconsistencies between the stereo pair and inconsistencies between overlapping CCDs in the same image mosaic. A method for the incorporation of jitter terms into the BA was developed and proved to be effective. We created a 1-m-resolution digital elevation model and an orthophoto using this methodology and compared them with topographic products from the U.S. Geological Survey.