Background analysis-based low-altitude low-speed unmanned aerial vehicle rapid detection method

Abstract

The invention relates to a background analysis-based low-altitude low-speed unmanned aerial vehicle rapid detection method, and belongs to the field of quadrotor unmanned aerial vehicle rapid detection. The method mainly comprises the following steps: acquiring picture sequence data provided by ground monitoring in an actual application occasion, extracting and quantifying information of a movingtarget in an image, and extracting image edge information for background analysis, thereby detecting the low-altitude low-speed unmanned aerial vehicle. The method further comprises the following steps: in a background analysis process, firstly, selecting an adjacent region of a moving target, carrying out partitioning processing, and carrying out statistics on an edge information amount in each block; judging the background condition of the moving target through a statistical result, and reserving suspected target information under the sky background; and finally, carrying out background analysis on the next frame again, and if the analysis results are all suspected targets, judging that the unmanned aerial vehicle is the unmanned aerial vehicle. The method has the advantages of being high in practicability, good in robustness and easy to implement, can adapt to complex meteorological conditions, and has certain resistance to adverse factors such as lens jitter and lens rotation.

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