FIELD: physics.
SUBSTANCE: invention relates to computer engineering. Method comprises counting the number of connected cameras, continuous sequential reading of frames from all connected cameras, searching for optical markers on each read frame, forming a packet with an array of geometric centres of each found marker and sending to a computing device, conversion of marker image local coordinates in camera coordinates system into global coordinates, wherein in order to measure the spatial coordinates of the marker, all possible combinations of simultaneously observing it l cameras are analysed, for each combination a pseudo-solution is found for the spatial coordinates of reference points of the marker and the result of the photogrammetric measurement is taken to be a pseudo-decision, at which the root-mean-square error of determining linear dimensions between reference points of the marker is minimized.
EFFECT: reduced error in estimating 3D coordinates of markers used in multi-camera optical tracking systems.
2 cl, 1 dwg
