A transplantable unmanned aerial vehicle detection system combined with an imaging simulation technology

Abstract

The invention provides a portable unmanned aerial vehicle detection system combined with an imaging simulation technology, and relates to the technical field of imaging simulation and machine vision crossing, the system uses the imaging simulation technology, the data collection process is simplified, the model training period is shortened, calculation is carried out by using a network after internal weight adjustment, and finally unmanned aerial vehicle detection is realized. The invention discloses a transplantable unmanned aerial vehicle detection system combined with an imaging simulationtechnology, a plurality of complex backgrounds can be obtained; various types of unmanned aerial vehicles are detected in real time; the problem that a traditional template matching method is poor ingeneralization ability is solved through the deep learning technology, the data collection process is simplified through the combination with the imaging simulation technology, the model training period is shortened, meanwhile, the portable intelligent unmanned aerial vehicle detection system can be set up on various embedded devices, and the portable intelligent unmanned aerial vehicle detectionsystem is provided.

Description

technical field [0001] The invention belongs to the cross technical field of imaging simulation and machine vision, and relates to a portable unmanned aerial vehicle detection system combined with imaging simulation technology. Background technique [0002] In recent years, drone technology has developed rapidly, and more and more people have begun to use drones to complete various creative tasks. Therefore, in drone no-fly zones or public places, the flight of drones must be regulated, otherwise the safety and privacy of the public will be threatened. A very critical step in drone supervision is drone detection, which is to quickly find the position of the drone in each frame of the video. There is no doubt that the future will also develop in the direction of intelligent supervision, so it is very necessary to design a portable intelligent detection system. [0003] Today, artificial intelligence technology is gradually subverting people's way of life. In order to meet t...

AI Technical Summary

Problems solved by technology

However, the UAV detection technology based on deep learning relies on a large amount of training data, and objects such as UAVs are special compared with daily detection objects, and it is difficult to find UAVs from public datasets used to train neural networks. The relevant training data

If you manually obtain videos of specific background environments and drone types from the Internet, firstly, it will take a lot of time to collect, and secondly, you need to consider the cost of manual labeling

Images