The purpose of the thesis is to verify the usefulness of aerial photography from an unmanned aerial vehicle for faster production of traffic accidents plans. Time-consuming field measurement of data for the creation of a traffic accident plan is one of the main reasons for the long-term traffic jams. We assume that the police could use a small unmanned aerial vehicle to perform the task faster and with a proper quality. At a suitable location we simulated a traffic accident that was photographed in several different ways from the aircraft by taking into account photogrammetric rules. With classic geodetic measurements, we measured reference and control points. With appropriate software, we oriented our photos, made a dense cloud of points and an orthophoto of the accident area, which serve to create a vector plan. We verified the accuracy of orthophotos with a relative error, which was calculated between the distances determined on the ortophoto and on the field, with the surveying instruments. Throughout the execution of the thesis, we carried out various analyses of the impacts of control distances and ways of aerial photography on the quality of the orthophoto, where we checked the effect of shortening the time of data capturing. We found a method that will enable the fastest way of capturing data, but so that the quality of the finished product is still appropriate.