Graphic Documentation of Archaeological Heritage Using Lidar on Uav. Experimental Analysis of Optimal Flight Parameters

Abstract

The introduction of new technologies has opened new methodological avenues and theoretical approaches of great interest for archaeological research and the conservation of historical heritage. This study delves into the advantages of LiDAR systems in the analysis of landscapes and archaeological elements in areas with dense vegetation. The acquisition of three-dimensional graphical documentation of reality for topographic, heritage, architectural, or industrial purposes has been revolutionized in the last decade by the application of unmanned aerial vehicles (UAVs). UAVs have become a fundamental tool in the daily work of technicians, allowing the documentation of large land areas or inaccessible places with great efficiency and better results using traditional methods. UAVs, commonly known as drones, can carry various types of sensors, always conditioned by their size and weight. These limitations mainly resulted in the use of small passive sensors, such as RGB, multispectral, or infrared cameras. The evolution of the sector has enabled the availability of active remote sensing LiDAR (Light Detection and Ranging) sensors at the consumer level, opening up new possibilities that multi-image photogrammetry does not allow, such as acquiring information with reduced ambient light, in shaded areas, or under vegetation. The aim of this study is to perform a comparative analysis of the three-dimensional representation of terrain and structural elements present in densely vegetated landscapes, analysing the penetration index of the airborne LiDAR pulse recording archaeological features based on flight parameters: altitude and speed. The study also offers a series of methodological innovations based on the data extracted and the specifics of these types of sensors. At this point, we can pose the question: what is the difference between obtaining surveying for technical versus archaeological purposes? In topography, the precision of the digital model is established based on the working scale, generating a regulated surface where the final result undergoes a generalization process that smooths terrain forms to obtain contour lines that represent the terrain and are easily interpretable based on the scale. Conversely, in archaeology, evidence of anthropogenic activity is sought, requiring a higher definition of shapes, a greater number of points, and the application of specific visualization systems. For our study, we have proposed two distinct scenarios: an ideal one (without any vegetation or obstacles and horizontal topography, specifically a model airplane landing strip), and a very complex one with dense pine vegetation and steep slopes, conducting a series of flights at different altitudes (between 70 and 120 m) and various aircraft speeds (between 5 and 10 m/s). Finally, we conducted a statistical study of the results to determine the optimal parameters for the desired goals, proposing an optimal workflow and scheme.