The International Archives of the Photogrammetrv. Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B4. Beiiins 2008 
to be a powerful approach, in particular with regard to quality: 
high DTM resolution with very good absolute accuracy, mode 
ling of fine surface structures, and simultaneous derivation of 
material properties. Hapke Model parameters that have been 
derived based on the integrated approach nicely match with fin 
dings of independent photometric studies. The benefit from 
combining matching with photoclinometry is clearly visible in 
the resulting DTMs, especially in comparison to HRSC DTM 
Test data, which had been derived using either method. In par 
ticular, advantages over image matching are: 
• A gridded DTM is the direct modeling result whereas 
it has to be interpolated from independent, irregularly 
distributed points after image matching. 
• Therefore, local dependencies are implicitly regarded. 
• The combination with photoclinometry introduces fur 
ther local accuracy and bridges weak areas. It can be 
thought of as “terrain-dependent regularization”. 
Major advantages compared to sole photoclinometry are: 
• With terrain-dependent weighting against matching, 
photoclinometry can be targeted where necessary. 
• DTM artifacts related to the illumination direction are 
reduced. 
• Radiometric parameters for surface and atmosphere 
can be derived within the modeling approach. 
However, the integrated approach is time-consuming and highly 
sophisticated; it can hardly be automated and is far from being 
operational, e.g., for processing entire HRSC orbits. Neverthe 
less, for small regions - that might be mosaicked, if desired - 
fine structures of the Martian surface can be modeled in high 
resolutions of up to 2x2 orthoimage surfels per DTM facet. 
In both examples presented in this paper, the assumption of in 
variant radiometric surface properties has been made. In gene 
ral, this has to be carefully evaluated, as unregarded radiometric 
variations will affect geometry! If applicable, parameters have 
to be modeled in facets as shown in Fig. 1. 
In this context, it suggests itself to extend the approach to mul 
tiple HRSC orbits. Theoretically, this would substantially widen 
the range of illumination and viewing geometries and enable 
better radiometric surface modeling, e.g., in terms of deriving 
more than four parameters of the Hapke Model (see chapter 2). 
Moreover, wavelength-dependencies could be modeled based 
on redundant color information. However, different orbits are 
obtained at different times under different atmospheric condi 
tions that would have to be modeled (and optical depth depends 
on wavelength!). Also the Martian surface might change, even 
over short periods of time caused by eolian activity. 
REFERENCES 
Albertz, J., et al., 2005. HRSC on Mars Express - Photogram- 
metric and Cartographic Research. Photogrammetric Enginee 
ring & Remote Sensing, 71(10), pp. 1153-1166. 
Fua, P., Leclerc, Y.G., 1995: Object-Centered Surface Recon 
struction: Combining Multi-Image Stereo and Shading. Inter 
national Journal of Computer Vision, 16(1), pp. 35-55. 
Gehrke, S., 2007. Zur geometrischen und radiometrischen 
Modellierung der Mars-Oberfläche aus HRSC-Daten. Publika 
tionen der DGPF, 16, pp. 683-694. 
Gehrke, S., Haase, I., 2006. An Integrated Approach for 
Orthoimage and DTM Generation Using Mars Express HRSC 
Data. IAPRS, 36(B4), pp. 326-332. 
Hapke, B.W., 1993: Theory of Reflectance and Emittance 
Spectroscopy. Cambridge University Press. 
Heipke, C., et al., 2007: Evaluating Planetary Digital Terrain 
Models - The HRSC DTM Test. Planetary and Space Science, 
55(14), pp. 2173-2193. 
Hoekzema, N.M., et al., 2006: Atmospheric Optical Depths 
from HRSC Stereo Images of Mars. Not yet published. 
Jaumann, R., et al., 2007: The High Resolution Stereo Camera 
(HRSC) Experiment on Mars Express: Instrument Aspects from 
Interplanetary Cruise through Nominal Mission. Planetary and 
Space Science, 55(7-8), pp. 928-952. 
Jehl, A., et al., 2006: Improved Surface Photometric Mapping 
across Gusev and Apollinaris from an HRSC/Mars Express In 
tegrated Multi-Orbit Dataset: Implication on Hapke Parameters 
Determination. 3 7th LPS, abstract #1219. 
Jehl, A., et al., 2008: Gusev Photometric variability as seen 
from the Orbit by HRSC/Mars Express. Not yet published. 
Kirk, R.L., et al., 2004: Comparison of USGS and DLR Topo 
graphic Models of Comet Borrelly and Photometric Applica 
tions. ICARUS, 167: 54-69. 
Lohse, V., et al., 2006: Derivation of Planetary Topography 
Using Multi-Image Shape-from-Shading. Planetary and Space 
Science, 54(7), pp. 661-674. 
McEwen, A.S., 1991: Photometric Functions for Photoclino 
metry and Other Applications. ICARUS, 92, pp. 298-311. 
Smith, D.E., et al., 2001: Mars Orbiter Laser Altimeter: Ex 
periment Summary after the first Year of Global Mapping of 
Mars. Journal of Geophysical Research, 106(E10), pp. 23689- 
23722. 
Spiegel, M., Neukum, G., 2007: Simultaneous Adjustment of 
Interior and Exterior Orientation of HRSC on Mars Express. 
Proceedings ISPRS WG IV/7 Workshop, Houston. 
Weisensee, M., 1992: Modelle und Algorithmen für das Facet- 
ten-Stereosehen. DGK-Reihe C, 374. 
Wrobel, B.P., 1987. Facets Stereo Vision (FAST Vision) - A 
New Approach to Computer Stereo Vision and Digital Photo- 
grammetry. Intercommission Conference on Fast Processing of 
Photogrammetric Data, Interlaken, pp. 231-258. 
Wrobel, B.P., et al., 1992: Adaptive Surface Regularization - 
A New Method for Stabilization of Surface Reconstruction 
from Images. IAPRS, 29(B3), pp. 824-831.