Method and device used for marking object

Abstract

The invention provides a method and a device used for marking an object. The method comprises steps that a panorama image is acquired by an unmanned plane, and the panorama image is sent to auxiliary driving equipment; the panorama image is received by the auxiliary driving equipment, visual scene analysis on the panorama image is carried out, and a scene and the object of the panorama image are acquired; target identification of the object under the scene corresponding to the panorama image is carried out through the auxiliary driving equipment; when a target object exists, an image frame including the target object is displayed by the auxiliary driving equipment; and the object in the image frame is marked. Through the method, the panorama image can be acquired in real time, all visual fields around the unmanned plane are guaranteed to monitor, the target object of the scene of the unmanned plane is further marked, reference is provided for auxiliary driving of the unmanned plane, and an operator is made to easily avoid the target object.

Description

technical field [0001] The invention relates to the field of assisted driving of unmanned aerial vehicles, in particular, to a method and device for marking objects. Background technique [0002] In the prior art, UAV assisted driving mainly obtains the data information required for UAV driving through gyroscope, GPS, airspeed head and other sensors, so that the assisted driver can control the UAV through the above data information. Operation, so as to ensure the stability of the drone's flight attitude, and realize the transfer of the spatial position and other functions. [0003] However, under the existing technology, the driver cannot monitor the surrounding environment of the UAV in real time in all directions, and at the same time has no energy to respond to the situation in all directions. Therefore, there are three major defects in the existing technology: [0004] 1. Unable to effectively avoid dynamic obstacles: [0005] The real scene is complex and changeable. ...

AI Technical Summary

Problems solved by technology

[0003] However, under the existing technology, the driver cannot monitor the surrounding environment of the UAV in real time in all directions, and at the same time has no energy to respond to the situation in all directions. Therefore, there are three major defects in the existing technology:

[0004] 1. Unable to effectively avoid dynamic obstacles:

[0005] The real scene is complex and changeable. When controlling the UAV remotely, it is often easy to ignore the situation on the flanks when changing the driving direction.

[0006] 2. Unable to avoid dynamic obstacles such as flying birds suddenly appearing in the air;

[0007] The real scene is complex and changeable. Even if the drone is controlled remotely, even if it is traveling in an open airspace, the operator cannot effectively respond to sudden birds and other objects because the operator cannot perceive them in advance.

[0008] 3. Unable to effectively avoid static obstacles such as complex buildings;

[0009] It is difficult for a UAV to make a good assessment of the environment and surrounding objects after passing through complex building structures or jungles. Therefore, even if it is operated manually, it is difficult to effectively avoid obstacles

Images