Method and apparatus for inter-frame scene matching based on multi-rotor unmanned aerial vehicle aerial video

Abstract

The invention discloses an inter-frame scene matching method and a device based on multi-rotor unmanned aerial vehicle aerial video. The sparse optical flow method is used to extract the characteristic corner pairs from the target region image. According to the first geometrical space distance of each characteristic corner pair, the well area division of the target area image and the combination of characteristic corner pairs, the combination of characteristic corner pairs of the target is selected. Calculating a second geometrical space distance between two corner pairs of the transformed target characteristic corner pairs according to a perspective transformation matrix, and further calculating a sum of weighted distances of all target characteristic corner pairs in each target characteristic corner pair combination; the combined result of the target feature corner pairs with the smallest sum of weighted distances is taken as the final matching result. The invention can effectively cope with the complex and changeable motion shooting state of the multi-rotor unmanned aerial vehicle imaging device, and does not need to rely on flight data or attitude parameters of the multi-rotorunmanned aerial vehicle and the imaging device thereof.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an inter-frame scene matching method and device based on multi-rotor UAV aerial video. Background technique [0002] Aerial photography, also known as aerial photography or aerial photography, refers to the photography and photography activities of the earth's landforms, urban landscapes, people and other aspects from the air. At present, UAV aerial photography has been widely used in military affairs, transportation construction, water conservancy engineering, ecological research, TV columns and art photography. [0003] However, during the research and practice of the prior art, the inventors of the present invention found that the existing UAV aerial photography scene matching methods generally need to perform three-dimensional analysis such as aerial triangulation based on the flight data or parameters of the imaging device. Data calculation, while the exist...

AI Technical Summary

Problems solved by technology

[0003] However, during the research and practice of the prior art, the inventors of the present invention found that the existing UAV aerial photography scene matching methods generally need to perform three-dimensional analysis such as aerial triangulation based on the flight data or parameters of the imaging device. Data calculation, while the existing scene matching method that does not depend on the parameters of the imaging device assumes that the scenes shot twice or more are all shot at the same spatial position with different shooting parameters and angles of the imaging device

That is to say, the existing technology cannot effectively cope with the complex and changeable motion shooting state of the multi-rotor UAV imaging device, and relies heavily on the flight data or attitude parameters of the multi-rotor UAV and its imaging device, and the multi-rotor UAV must be deliberately limited. Flight Control of Rotary Wing UAV

Images