Digital Elevation Model enhancement using Deep Learning [CL]

http://arxiv.org/abs/2101.04812


We demonstrate high fidelity enhancement of planetary digital elevation models (DEMs) using optical images and deep learning with convolutional neural networks. Enhancement can be applied recursively to the limit of available optical data, representing a 90x resolution improvement in global Mars DEMs. Deep learning-based photoclinometry robustly recovers features obscured by non-ideal lighting conditions. Method can be automated at global scale. Analysis shows enhanced DEM slope errors are comparable with high resolution maps using conventional, labor intensive methods.

Read this paper on arXiv…

C. Handmer
Thu, 14 Jan 21
78/79

Comments: 11 pages, 13 figures