UAV Attitude Angle Estimation Based on Door Frame Deformation in Image

In this paper, we introduce a new method for calculating position angle of the barrier-wearing drone through the rectangular change in the image. This method only uses the visual deformation information of rectangular frame to calculate the angle between UAV navigation direction and door frame plane. There is a certain geometric transformation relationship between the change of view angle and the deformation of rectangle. In this paper, we infer the relationship between deformed graphic and UAV attitude under a series of assumptions. This method does not consider the distortion and rotation of rectangle, which are generally very sensitive to noise working conditions. No complicated mathematical computation is involved in the implementation and the required computational storage space is much cheaper compared to the current methods. Experiments with simulated images indicate the effectiveness of the proposed method.