Anti-shake method of UAV image transmission video and UAV

Abstract

The invention discloses an anti-shake method for UAV image transmission video and the UAV. The method includes: acquiring feature points and feature vectors of the current frame image of the video; combining the feature vectors with the previous frame in the video Match the feature vectors of the corresponding feature points of the image, calculate the first optical flow vector and the first rotation angle between the two frames of images; judge the first optical flow vector and the first rotation angle according to the first optical flow vector, the first rotation angle and the gyroscope data Whether the first rotation angle is generated by the shaking of the UAV; if the first optical flow vector and the first rotation angle are generated by the shaking of the UAV, adjust the current frame image according to the first optical flow vector and the first rotation angle to obtain The current frame image after dejittering. The technical solution of the present invention corrects the error caused by the shaking of the drone through the software solution, does not introduce the hardware used for anti-shake, reduces the cost and complexity, can be accurate to the sub-pixel level, and has higher anti-shake accuracy, eliminating The shaking effect is better.

Description

technical field [0001] The invention relates to the technical field of unmanned aerial vehicles, in particular to an anti-shake method for unmanned aerial vehicle image transmission video and an unmanned aerial vehicle. Background technique [0002] At present, drones equipped with cameras have gradually become a popular trend. From the drone side, the video images captured by the camera are transmitted back to the terminal device, which can realize multi-angle and wide-range wide-angle shooting. However, there are many unavoidable shakes during the flight of the drone, which will cause the captured video images to be shaken and unstable, which affects the user's experience and expectations of the video image. [0003] In view of this, some high-end drones add a mechanical gimbal to the camera to correct the video screen shake caused by the camera shaking at the same frequency due to the shaking of the drone. This solution can effectively remove the video shake and improve t...

AI Technical Summary

Problems solved by technology

However, there are many unavoidable shakes during the flight of the drone, which will cause the captured video screen to shake and be unstable, which affects the user's experience and expectation of the video image.

[0003] In view of this, some high-end drones add a mechanical gimbal to the camera to correct the video screen shake caused by the camera shaking at the same frequency due to the shaking of the drone. This solution can effectively remove the video shake and improve the user experience. experience effect, but installing a mechanical gimbal on the UAV increases the weight of the UAV, thereby shortening the life time of the UAV. In addition, adding a mechanical gimbal also increases the cost of the UAV

[0004] There is also a plan to use the gyroscope to detect and determine the shake matrix of the camera, and correct the image captured by the camera through the shake matrix to remove the shake of the video image. However, this technology needs to match the data of the gyroscope with the field of view of the current camera lens. The focal length and pixel size of each camera are different, so every camera needs to match the field of view with the gyroscope data, which increases the complexity of the work. If the field of view is not matched accurately, the anti-shake effect will also be affected

Images