Super-Resolution Techniques in Photogrammetric 3D Reconstruction from Close-Range UAV Imagery
Datos Bibliográficos
| ID | 13125791 |
|---|---|
| Autores | Antigoni Panagiotopoulou (0009-0008-9144-1982, University of West Attica), Λάζαρος Γραμματικόπουλος (0000-0002-3858-1352, University of West Attica), Andreas El Saer (0000-0002-1320-7102, University of West Attica), Elli Petsa (University of West Attica), Eleni Charou (0000-0003-0452-1754, National Centre of Scientific Research "Demokritos"), Lemonia Ragia (0000-0002-3232-8671, Hellenic Open University, autor de correspondencia), George C Karras (0000-0002-4045-4715, National Technical University of Athens) |
| Año | 2023 |
| Volumen | 6 |
| Número | 3 |
| Páginas | 2701-2715 |
| Fecha de publicación | 2023-03-06 |
| Peer Reviewed | Sí |
| Open Access | Sí |
| Tipo | ARTICLE |
| Revista | Heritage (JOURNAL) |
| Identificadores de la revista | ISSN: 2571-9408 • E-ISSN: 2571-9408 |
| Editorial | MDPI AG (PUBLISHER • IT) |
| DOI | 10.3390/heritage6030143 |
| OpenAlex | W4323312718 |
| Idioma | EN |
| Citas recibidas | 3 |
| Referencias citadas | 38 |
Current Multi-View Stereo (MVS) algorithms are tools for high-quality 3D model reconstruction, strongly depending on image spatial resolution. In this context, the combination of image Super-Resolution (SR) with image-based 3D reconstruction is turning into an interesting research topic in photogrammetry, around which however only a few works have been reported so far in the literature. Here, a thorough study is carried out on various state-of-the-art image SR techniques to evaluate the suitability of such an approach in terms of its inclusion in the 3D reconstruction process. Deep-learning techniques are tested here on a UAV image dataset, while the MVS task is then performed via the Agisoft Metashape photogrammetric tool. The data under experimentation are oblique cultural heritage imagery. According to results, point clouds from low-resolution images present quality inferior to those from upsampled high-resolution ones. The SR techniques HAT and DRLN outperform bicubic interpolation, yielding high precision/recall scores for the differences of reconstructed 3D point clouds from the reference surface. The current study indicates spatial image resolution increased by SR techniques may indeed be advantageous for state-of-the art photogrammetric 3D reconstruction
3D reconstruction · Archaeology · Bicubic interpolation · Computer vision · Context (archaeology · Geography · Image (mathematics · Image resolution · Interpolation (computer graphics · Iterative reconstruction · Linear interpolation · Orthophoto · Pattern recognition (psychology · Photogrammetry · Point cloud · Remote sensing · 3D Surveying and Cultural Heritage · Advanced Image Processing Techniques · Advanced Vision and Imaging · Computer Science · Artificial Intelligence · Geology
| Obras citantes distintas | 3 |
|---|---|
| Citas por año | 1,5 |
| Intervalo de citas | 2024 - 2026 (3) |
| Velocidad de citación | current |
| Altamente citado | No |
| Tipos de cita | Neutras: 3 |