EUVIMEDEUVIMED entdecken
Crossref

Satellite imagery and deep learning for discovering lost civilisations

Parthip Prasooth, Nimah Jaleel, Avin Reji, Amaan Mir, Sahithi Pashapu, Efstratios Ntantis

Multidisciplinary Science Journal · 15.10.2026 · 9 (4) · 2027077

DOI: 10.31893/multiscience.2027077 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Multidisciplinary Science Journal, online erschienen am 15.10.2026. Verfasst von Parthip Prasooth und 5 weiteren Autor:innen. Die Arbeit nennt 66 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Geographical constraints, high operational costs, environmental challenges, and the inherently time-consuming nature of excavation have long limited the effectiveness of traditional archaeological methods. Many potential archaeological sites remain inaccessible due to dense vegetation, remote locations, political instability, or urban development. In recent years, however, advances in artificial intelligence (AI), deep learning, and remote sensing technologies have transformed archaeological exploration by enabling the detection of ancient structures and lost civilisations through the analysis of satellite imagery. This paper investigates the use of AI-driven deep learning frameworks, particularly convolutional neural networks (CNNs), for identifying and mapping previously unknown archaeological sites. By processing large volumes of high-resolution satellite data, AI algorithms can recognise subtle spatial patterns, geometric features, and landscape anomalies that may indicate the presence of buried or obscured archaeological remains. To improve detection accuracy, several image preprocessing techniques are employed, including edge detection, spectral analysis, image enhancement, and noise reduction. These methods increase image clarity and allow AI models to distinguish archaeological features from natural terrain with greater precision. The effectiveness of AI-assisted archaeology is demonstrated through several notable case studies, including the rediscovery of lost Mayan settlements hidden beneath dense tropical forests and the identification of previously unknown geoglyphs in Peru’s Nazca Desert. These successes highlight the ability of AI to accelerate archaeological surveys while significantly reducing the need for extensive and invasive excavation campaigns. Despite its considerable advantages, AI-based archaeological exploration faces challenges such as false-positive detections, limited training datasets, algorithmic bias, and ethical concerns related to cultural heritage management. Nevertheless, continued improvements in machine learning algorithms, combined with real-time satellite monitoring, drone-based imaging, and geophysical validation techniques, are expected to further enhance discovery rates. Overall, AI offers a scalable, cost-effective, and non-invasive approach that is reshaping modern archaeology and expanding our ability to uncover and preserve evidence of ancient civilisations.

Zitieren

APA

Prasooth, P., Jaleel, N., Reji, A., Mir, A., Pashapu, S., & Ntantis, E. (2026). Satellite imagery and deep learning for discovering lost civilisations. Multidisciplinary Science Journal, 9(4), 2027077. https://doi.org/10.31893/multiscience.2027077

Vancouver

Prasooth P, Jaleel N, Reji A, Mir A, Pashapu S, Ntantis E. Satellite imagery and deep learning for discovering lost civilisations. Multidisciplinary Science Journal. 2026;9(4):2027077. doi:10.31893/multiscience.2027077

BibTeX
@article{Prasooth2026,
  author = {Prasooth, Parthip and Jaleel, Nimah and Reji, Avin and Mir, Amaan and Pashapu, Sahithi and Ntantis, Efstratios},
  title = {Satellite imagery and deep learning for discovering lost civilisations},
  journal = {Multidisciplinary Science Journal},
  year = {2026},
  volume = {9},
  number = {4},
  pages = {2027077},
  doi = {10.31893/multiscience.2027077}
}
Kostenlos starten →