CNN-Based Large Area Pixel-Resolution Topography Retrieval From Single-View LROC NAC Images Constrained With SLDEM

dc.contributor.authorChen, Hao
dc.contributor.authorXiao, Haifeng
dc.contributor.authorYe, Zhen
dc.contributor.authorZhang, Hanyue
dc.contributor.authorTong, Xiaohua
dc.contributor.authorOberst, Jürgen
dc.contributor.authorGläser, Philipp
dc.contributor.authorHu, Xuanyu
dc.date.accessioned2023-01-26T14:49:41Z
dc.date.available2023-01-26T14:49:41Z
dc.date.issued2022
dc.description.abstractThe 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.sponsorshipTU Berlin, Open-Access-Mittel – 2022en
dc.identifier.eissn2151-1535
dc.identifier.urihttps://depositonce.tu-berlin.de/handle/11303/18079
dc.identifier.urihttps://doi.org/10.14279/depositonce-16872
dc.language.isoen
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subject.ddc520 Astronomie und zugeordnete Wissenschaftende
dc.subject.otherDeep learning-based reconstructionen
dc.subject.otherdigital terrain model mosaicsen
dc.subject.otherpixel-resolution DTMen
dc.subject.othersingle-view narrow angle camera imagesen
dc.subject.otherSelenological and Engineering Explorer and LRO LOLA Elevation Modelen
dc.subject.otherSLDEMen
dc.titleCNN-Based Large Area Pixel-Resolution Topography Retrieval From Single-View LROC NAC Images Constrained With SLDEMen
dc.typeArticle
dc.type.versionpublishedVersion
dcterms.bibliographicCitation.doi10.1109/JSTARS.2022.3214926
dcterms.bibliographicCitation.journaltitleIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
dcterms.bibliographicCitation.originalpublishernameIEEE
dcterms.bibliographicCitation.originalpublisherplaceNew York
dcterms.bibliographicCitation.pageend9416
dcterms.bibliographicCitation.pagestart9398
dcterms.bibliographicCitation.volume15
dcterms.rightsHolder.referenceCreative-Commons-Lizenz
tub.accessrights.dnbfree
tub.affiliationFak. 6 Planen Bauen Umwelt::Inst. Geodäsie und Geoinformationstechnik::FG Planetengeodäsie
tub.publisher.universityorinstitutionTechnische Universität Berlin

Files

Original bundle
Now showing 1 - 1 of 1
Loading…
Thumbnail Image
Name:
Chen_etal_2022.pdf
Size:
24.05 MB
Format:
Adobe Portable Document Format
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
4.23 KB
Format:
Item-specific license agreed upon to submission
Description:

Collections