Relative Orientation Method of UAV Images without Ground Control Points

In order to solve the problem of low accuracy in image relative orientation due to the posture instability of unmanned aerial vehicle (UAV), a method to generate UAV image strips and constitute pair of stereoscopic images based on POS (position and orientation system) data was proposed. In accordance with the principle of relative orientation, UAV POS data were used as the priori constraint in relative orientation and adjusted with error equations to reconstruct the cost function and conduct the relative orientation of images. In order to verify the validity of this method, it was compared with the conventional relative orientation method. The experimental results show that it can raise the accuracy of relative orientation by 10% to use UAV POS data as the priori constraint. Especially to images with large distortion, the orientation accuracy can be improved greatly.