In the course of this master¿s thesis, an attempt is being made for the first time to derive, analyze and interpret geological, petrophysical, and geological parameters from drone image data. However, the focus is not on the development of artificial intelligence (AI) architectures, but on the acquisition of parameters from rock samples and the generation as well as the processing of data for the training of these algorithms. In the first chapter, the task of this thesis is explained in more detail and an introduction to the project at present or to be worked on is given. This is followed by an overview of the current state of knowledge and research regarding the use of drones in the geotechnical sector and the mining and tunnelling industry. Chapter three deals exclusively with the methodology of data collection, starting with a brief introduction of the project area, and a description of the geological inventory including the presentation of different geotechnical approaches for the classification of rock masses in the field (including "scanline mapping"). Furthermore, the sampling procedure as well as the subsequent sample preparation and investigations in the laboratory will be discussed. In addition, this section covers the fundamentals of Unmanned Aerial Vehicles (UAV) a.k.a. drones as well as the basics of planning flights, taking into account local factors and prevailing conditions. The fourth main point is dedicated to the processing of the drone data collected during the flight. All process steps, from the acquired raw data to the finished 3D point clouds, are explained and described individually, broken down by software. The final fifth chapter, "Interpretations", begins with the evaluation of the laboratory data. By comparing and contrasting selected plots and diagrams, trend properties are identified and correspondences between sample parameters and literature values are found. Subsequently, the topic of rock mechanical classification systems (e. g. Geological Strength Index - GSI) will be revisited and their influences assessed in relation to the given project areas. After the processing and analysis of the laboratory data, the following subsections are dedicated to the AI algorithms to be tested or trained. Preparing for the interpretation of the segmented 3D point clouds, the structure of the applied artificial intelligence workflow, and the creation of the training and verification data are discussed. At the end of this chapter, a comparison of all segmentation variants performed on the 3D models (using RGB and/or normal vectors) is made, followed by an evaluation using a feature matrix. To conclude this work, the main results and conclusions are summed up and an outlook, as well as impulses for future projects in this field, are given.