This page contains digital elevation maps (DEMs) of regions of the lunar surface obtained using images of the Moon Mineralogy Mapper (M3) instrument on board the Indian lunar orbiter Chandrayaan-1. The M3 data set, which is available from the PDS Imaging Node, has a resolution of about 140 m per pixel. It is used as an input for our image-based DEM construction algorithm as it provides pixel-synchronous 85-channel multispectral image data and resampled elevation data obtained by the Lunar Orbiter Laser Altimeter (LOLA) instrument on board the US spacecraft Lunar Reconnaissance Orbiter (LRO). To avoid the thermal radiance component, channels with centre wavelength above 2000 nm are neglected.
In order to reconstruct a DEM of the lunar surface from an optical image, a modified version of the shape from shading algorithm proposed by Horn (1989) is used. To maintain accuracy on large scales we added an error term to the optimisation problem which describes the deviation of the resulting DEM from the LOLA DEM at a lower lateral resolution than the original image. Although in principle a single image is sufficient to construct the DEM, our method takes into account the non-uniform surface albedo, where initially a strongly blurred and thus almost uniform albedo map is assumed. In an iterative manner, the DEM is then reconstructed and the albedo map is recovered simultaneously on decreasing spatial scales. The lateral resolution of the resulting DEM comes close to that of the M3 images.
Further details about the DEM construction algorithm and its application to the analysis of spectral features and the automatic detection of lunar craters can be found in the following publications:
S. Herbort, A. Grumpe, C. Wöhler.
Reconstruction of Non-Lambertian Surfaces by Fusion of Shape from Shading and Active Range Scanning.
Proc. IEEE Int. Conf. on Image Processing, Brussels, Belgium, 2011.
C. Wöhler, A. Grumpe.
Correction of Chandrayaan-1 M³ Lunar Hyperspectral Image Data with Respect to Local Topography.
Lunar and Planetary Science Conference XXXXIII, abstract #1906, The Woodlands, Texas, 2012.
S. Loncaric, G. Salamuniccar, A. Grumpe, C. Wöhler.
Automatic Detection of Lunar Craters Based on Topography Reconstruction from Chandrayaan-1 M3 imagery.
Lunar and Planetary Science Conference XXXXII, abstract #1454, The Woodlands, Texas, 2011.
The DEMs are provided as delimited ASCII text files in simple cylindrical projection (the same map projection is used for the LOLA DEM on the PDS server). The elevation values are given in metres relative to the lunar reference radius. The DEMs were constructed based on M3images resampled to a resolution of 1/300°. For compatibility with the 1/512° LOLA DEM, the DEMs are also provided in a resampled enlarged format at 1/512° resolution. The ZIP files below each contain a DEM along with the corresponding pixel-synchronous radiance image of channel 60 (centre wavelength 1978.1 nm) in BMP format. The filenames contain the DEM resolution and the coordinate ranges (selenographic longitude and latitude for 1/300° resolution, pixel coordinates in LOLA DEM for 1/512° resolution).
Stripe of 40° extension in north-south direction from Ptolemaeus to Huggins (4.2° S - 44.2° S, 357.6° E - 358.3° E)
Resolution 1/300°: regional_DEM_04.2S-44.2S_357.6E-358.3E_300ppd.zip
Resolution 1/512°: LU_GDEM_2_512_excludedregion2_90931x91290_48230y68711_native.zip