Aerial photogrammetry and its application in the reconstruction of accident scenarios

Abstract

In this research work, the photogrammetric processing of images obtained from unmanned aerial vehicles is carried out with the purpose of recreating or reconstituting the scenarios of traffic accidents that occurred in the highway area. The particularity of this research is that aerial photogrammetry is obtained under different climatic conditions taking, as reference, ground control points marked with a geodetic equipment under a Global Positioning System (GPS for its acronym in English) that indicates the coordinates in which the captured images are fixed. In addition, with the resources obtained from digital photogrammetric processing, such as 3D models, orthophotos and / or point cloud, the identification, length and positioning of braking tracks in the scene of an accident are established to estimate the speed at which the vehicles were circulating, prior to the occurrence of the accident. The proposed methodology helps to survey the scene of an accident to be faster, more efficient and more precise than the methods currently used in the field by police forces.

The results suggested that, considering the number of traffic accident scenarios recreated with the methodology described in the previous paragraph, the average relative error in estimating the length of braking tracks and distances, in general, does not exceed five percent in relation to the distances obtained in the field on a dry asphalt pavement.