Generation of pixel-level resolution lunar DEM based on Chang'E-1 three-line imagery and laser altimeter data

• We develop a data processing flow based on Chang'E-1 (CE-1) data.
• We use additional information (LAM data) to generate initial global disparity.
• We achieve pixel-level resolution corresponding points and high resolution DEM.

Acquiring high resolution topography products of the Moon is an important issue for further lunar exploration. In this paper, we present a data processing flow of 3D reconstruction based on Chang'E-1 (CE-1) three-line imagery (120 m) and laser altimeter (LAM) data to acquire pixel-level resolution DEM. First, initial global disparity estimation was generated, and then corresponding points were acquired pixel-by-pixel by Adaptive Support-Weight Approach. A Least Square Image matching method was used to achieve sub-pixel accuracy, ultimately resulting in pixel-level resolution DEM (120 m). Global lunar 100 m DEM (GLD100 DEM) from LROC WAC stereo imagery is employed for comparison analysis and the mean elevation difference between two DEMs is less than 120 m. Experimental results show that the processing flow is effective and reliable for acquiring pixel-level DEM based on CE-1 data.