Tao, Yu;
(2022)
On Martian Surface Exploration: Development of Automated 3D Reconstruction and Super-Resolution Restoration Techniques for Mars Orbital Images.
Doctoral thesis (Ph.D), UCL (University College London).
Preview |
Text
Tao_10155880_thesis_sigs_removed.pdf Download (24MB) | Preview |
Abstract
Very high spatial resolution imaging and topographic (3D) data play an important role in modern Mars science research and engineering applications. This work describes a set of image processing and machine learning methods to produce the “best possible” high-resolution and high-quality 3D and imaging products from existing Mars orbital imaging datasets. The research work is described in nine chapters of which seven are based on separate published journal papers. These include a) a hybrid photogrammetric processing chain that combines the advantages of different stereo matching algorithms to compute stereo disparity with optimal completeness, fine-scale details, and minimised matching artefacts; b) image and 3D co-registration methods that correct a target image and/or 3D data to a reference image and/or 3D data to achieve robust cross-instrument multi-resolution 3D and image co-alignment; c) a deep learning network and processing chain to estimate pixel-scale surface topography from single-view imagery that outperforms traditional photogrammetric methods in terms of product quality and processing speed; d) a deep learning-based single-image super-resolution restoration (SRR) method to enhance the quality and effective resolution of Mars orbital imagery; e) a subpixel-scale 3D processing system using a combination of photogrammetric 3D reconstruction, SRR, and photoclinometric 3D refinement; and f) an optimised subpixel-scale 3D processing system using coupled deep learning based single-view SRR and deep learning based 3D estimation to derive the best possible (in terms of visual quality, effective resolution, and accuracy) 3D products out of present epoch Mars orbital images. The resultant 3D imaging products from the above listed new developments are qualitatively and quantitatively evaluated either in comparison with products from the official NASA planetary data system (PDS) and/or ESA planetary science archive (PSA) releases, and/or in comparison with products generated with different open-source systems. Examples of the scientific application of these novel 3D imaging products are discussed.
| Type: | Thesis (Doctoral) |
|---|---|
| Qualification: | Ph.D |
| Title: | On Martian Surface Exploration: Development of Automated 3D Reconstruction and Super-Resolution Restoration Techniques for Mars Orbital Images |
| Open access status: | An open access version is available from UCL Discovery |
| Language: | English |
| Additional information: | Copyright © The Author 2021. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request. |
| UCL classification: | UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences > Dept of Space and Climate Physics UCL > Provost and Vice Provost Offices > UCL BEAMS UCL |
| URI: | https://discovery-pp.ucl.ac.uk/id/eprint/10155880 |
Archive Staff Only
 |
View Item |
