The use of unmanned aerial vehicles (UAV) to obtain cartographic information for multiple applications has increased at present. Different researches have focused on specific applications, from their implementation to their automation. This project focuses on the determination of the cartographic precision of the orthophotomosaics generated from images obtained with UAV processed with the Agisoft® software. For this, a TAROT 680 PRO hexacopter was used (with GPS and INS included) to cover one hectare in the Moorlands of Chingaza, reserve area for Espeletia in the northeast of Bogotá (Colombia). A total of 167 photographs were acquired with a Canon A2300 camera. Five ground control points (GCP) were pre-signalized (GCP-P) and positioned with a Gr-5 receiver, in order to adjust the block of images. Due to the weather conditions not all GCP-P were captured in the images. To improve the adjustment of the project, three GCP were subsequently photo-identified (GCP-F) and positioned with the same type of equipment. Three projects were created, one in which GCP-P are included; the second including the GCP-P and GCP-F; and the third one in which GCP is not included. The first two projects correspond to the photogrammetric SfM process and the last corresponds to the direct georeferencing technique. The best results were obtained in the project that used only three GCP-P. Likewise, it was found that including the GCP-F in the adjustment of the block decreases the accuracy of the project, moving from a millimeter to a submetric precision; while not using GCP does not allow knowing the accuracy of the project with respect to the terrain.