CNN-Based Large Area Pixel-Resolution Topography Retrieval From Single-View LROC NAC Images Constrained With SLDEM
dc.contributor.author | Chen, Hao | |
dc.contributor.author | Xiao, Haifeng | |
dc.contributor.author | Ye, Zhen | |
dc.contributor.author | Zhang, Hanyue | |
dc.contributor.author | Tong, Xiaohua | |
dc.contributor.author | Oberst, Jürgen | |
dc.contributor.author | Gläser, Philipp | |
dc.contributor.author | Hu, Xuanyu | |
dc.date.accessioned | 2023-01-26T14:49:41Z | |
dc.date.available | 2023-01-26T14:49:41Z | |
dc.date.issued | 2022 | |
dc.description.abstract | The production of high-resolution digital terrain models (DTMs) based on images is often hampered by the lack of appropriate stereo observations. Here, we propose a deep learning-based reconstruction of pixel-resolution DTMs from Lunar Reconnaissance Orbiter (LRO) single-view narrow angle camera (NAC) images, constrained by Selenological and Engineering Explorer and LRO LOLA Elevation Models (SLDEM). The procedure is carried out for a set of adjacent images, and the mosaicking of a contiguous large-area DTM is demonstrated. The approach is applied to the CE-3 and CE-4 landing sites, involving six multiple coverage and eight adjacent NAC L/R image pairs, respectively. For the DTM reconstruction, we use an improved convolutional neural network architecture with a weighted sum loss function involving three loss terms. We demonstrate that our method is robust and can deal with images acquired under varying illumination conditions. The DTM mosaic (1.5 m pixel size) covering the CE-4 landing area (72.8 × 30.3 km) is without apparent seams between the individual image boundaries and consistent with the SLDEM (60 m pixel size) in terms of overall elevation, trend, and scale, but is showing considerably more morphologic detail. | en |
dc.description.sponsorship | TU Berlin, Open-Access-Mittel – 2022 | en |
dc.identifier.eissn | 2151-1535 | |
dc.identifier.uri | https://depositonce.tu-berlin.de/handle/11303/18079 | |
dc.identifier.uri | https://doi.org/10.14279/depositonce-16872 | |
dc.language.iso | en | |
dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | |
dc.subject.ddc | 520 Astronomie und zugeordnete Wissenschaften | de |
dc.subject.other | Deep learning-based reconstruction | en |
dc.subject.other | digital terrain model mosaics | en |
dc.subject.other | pixel-resolution DTM | en |
dc.subject.other | single-view narrow angle camera images | en |
dc.subject.other | Selenological and Engineering Explorer and LRO LOLA Elevation Model | en |
dc.subject.other | SLDEM | en |
dc.title | CNN-Based Large Area Pixel-Resolution Topography Retrieval From Single-View LROC NAC Images Constrained With SLDEM | en |
dc.type | Article | |
dc.type.version | publishedVersion | |
dcterms.bibliographicCitation.doi | 10.1109/JSTARS.2022.3214926 | |
dcterms.bibliographicCitation.journaltitle | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing | |
dcterms.bibliographicCitation.originalpublishername | IEEE | |
dcterms.bibliographicCitation.originalpublisherplace | New York | |
dcterms.bibliographicCitation.pageend | 9416 | |
dcterms.bibliographicCitation.pagestart | 9398 | |
dcterms.bibliographicCitation.volume | 15 | |
dcterms.rightsHolder.reference | Creative-Commons-Lizenz | |
tub.accessrights.dnb | free | |
tub.affiliation | Fak. 6 Planen Bauen Umwelt::Inst. Geodäsie und Geoinformationstechnik::FG Planetengeodäsie | |
tub.publisher.universityorinstitution | Technische Universität Berlin |