A fast detection method for low-altitude and slow-speed UAVs based on background analysis

Abstract

The invention is a background analysis-based fast detection method for low-altitude and slow-speed unmanned aerial vehicles, which belongs to the field of fast detection of four-rotor unmanned aerial vehicles. The method mainly includes the following steps: obtaining the image sequence data provided by ground monitoring in practical applications, extracting and quantifying the information of moving targets in the images, and extracting image edge information for background analysis, thereby detecting low-altitude and slow-moving drones. In the process of background analysis, first select the adjacent area of ​​the moving target, perform block processing, and count the edge information in each block; then judge the background of the moving target through the statistical results, and retain the suspected target under the sky background information; finally, the background analysis is performed again on the next frame, and if the analysis results are all suspected targets, it is judged to be a drone. This method has the characteristics of strong practicability, good robustness, and easy implementation, and can adapt to more complex meteorological conditions, and has certain resistance to unfavorable factors such as lens shake, lens rotation, etc.

Description

technical field [0001] The invention belongs to the field of image processing, and in particular relates to a fast detection method for low-altitude and slow-speed unmanned aerial vehicles based on background analysis. Background technique [0002] At present, unmanned aerial photography equipment such as quadrotor UAVs are becoming more and more popular. UAVs often break into sensitive places such as airports and military restricted areas. If they are not controlled, it is easy to cause serious consequences. In fact, there are many difficulties in the rapid detection of aircraft through video surveillance. Targets such as drones have the characteristics of fast flight speed, small target imaging, and difficulty in feature extraction. Environmental factors include ground dynamic interference, illumination changes, and complex backgrounds. And other factors will affect the rapid detection of targets, and it is of great practical significance to solve the problem of rapid dete...

AI Technical Summary

Problems solved by technology

However, due to the wide range of video surveillance, the use of various scenarios, and the use of surveillance equipment are also different, which leads to complex background, target occlusion, target ratio change, target and background interference, etc. in the surveillance video. There are currently many fast detection algorithms based on video sequences, but affected by the above problems, it is difficult for conventional algorithms to achieve good detection results.

Images