Geospatial Analysis of the Roman Site of Munigua Based on RGB Airborne Imagery
This study investigates the use of high-resolution RGB aerial imagery from Spain’s National Aerial Orthophotography Plan (PNOA) for archeological feature detection through spectral index analysis and unsupervised clustering. Focusing on the Roman site of Munigua, eight orthophotographs acquired betw...
| Autores: | , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2025 |
| País: | España |
| Institución: | Universidad de Sevilla (US) |
| Repositorio: | idUS. Depósito de Investigación de la Universidad de Sevilla |
| OAI Identifier: | oai:idus.us.es:11441/177166 |
| Acceso en línea: | https://hdl.handle.net/11441/177166 https://doi.org/10.3390/rs17183224 |
| Access Level: | acceso abierto |
| Palabra clave: | temporal clustering RGB aerial imagery spectral indices archeological prospection Munigua |
| Sumario: | This study investigates the use of high-resolution RGB aerial imagery from Spain’s National Aerial Orthophotography Plan (PNOA) for archeological feature detection through spectral index analysis and unsupervised clustering. Focusing on the Roman site of Munigua, eight orthophotographs acquired between 2014 and 2024 were analyzed to compute five RGB-based spectral indices: VARI, GLI, ExG, CSI, and BI. These indices were used to detect surface spectral anomalies potentially linked to buried archeological structures. A multi-temporal approach was employed, with Principal Component Analysis (PCA) and K-Means clustering applied independently to each image. This allowed for the identification of temporally persistent anomalies (areas that remained within the same spectral cluster across multiple years), suggesting the presence of underlying anthropogenic features. Despite the lack of near-infrared data, the combination of RGB-based indices and temporal clustering proved effective for non-invasive prospection. The methodology is scalable, repeatable, and relies entirely on open-access datasets, making it suitable for broader applications in heritage monitoring and landscape archeology. The results underscore the potential of RGB imagery and time-series clustering in detecting subtle archeological signals within complex vegetated environments. |
|---|