Numerical Analysis of Multispectral Images for Detection of Buried Archaeological Remains

Preservation of the historical heritage of each nation is a task for everyone. Studying our history, as well as that of other countries, would allow us to rediscover ourselves and know where we come from our customs, and our characteristics. Archaeology is a fundamental part of this link of cultural development, which, together with geomatic engineering, as well as other fields of engineering, are rediscovering new everyday techniques that allow us to locate historical settlements buried by the passage of time. Remote sensing, however, is a non-intrusive technique that allows optimal results to be obtained in the shortest possible time. The localization of crop marks (sites with a radiometric response different from their surroundings) using infrared imagery is a long-standing task with increasingly rich results, which, together with the use of vegetation indices derived from multispectral data and the high spatial resolution of the imagery, narrows the narrow research path towards the discovery of new historical urban settlements. The present research aimed to improve the analysis of numerical data obtained in the classification of multispectral images to identify areas likely to harbor archaeological remains. These images were acquired using a Parrot Sequoia multispectral camera onboard an unmanned aerial vehicle, which was divided into four spectral bands (NIR, RED_EDGE, RED, and GREEN), which were processed to obtain new images corresponding to vegetation indices (NDVI, RVI, and TVI).