Unmanned aerial vehicle tracking shooting system and RGBD tracking method thereof

Abstract

The invention discloses an unmanned aerial vehicle tracking shooting system and an RGBD tracking method thereof. The system comprises a storage and service module, a tracking control module and a data transmission module, and each module is realized by a basic program language and a machine learning algorithm framework and is applied to unmanned aerial vehicle equipment. The storage and service module provides a unified model training interface for a tracking task and is used for storing and optimizing a model; the tracking control module uses a target tracking algorithm to track a shooting object in real time, and unmanned aerial vehicle tracking shooting in a complex environment is achieved; and the data transmission module is used for transmitting image data and tracking control instructions. According to the invention, the automatic tracking shooting function of an unmanned aerial vehicle without manual control is realized, the problem that the unmanned aerial vehicle is liable to lose a target due to a shielding factor in a complex shooting environment can be solved, and the availability of an unmanned aerial vehicle shooting technology in more scenes is improved.

Description

technical field [0001] The invention relates to the technical field of visual target tracking, in particular to an unmanned aerial vehicle tracking and shooting system and an RGBD tracking method thereof. Background technique [0002] UAV technology has been widely used in many fields such as advertising photography, express transportation, urban planning, etc. Among them, UAV aerial photography technology has the advantages of high definition, large shooting area, and small scene limitation, which replaces the traditional in many scenes. means of shooting. When using drones to shoot moving objects, the photographer usually needs to use ground equipment to manually control the flight status of the drone to complete the follow-up task and ensure the shooting quality. In recent years, artificial intelligence technology has developed rapidly, especially visual image recognition technology has realized the landing of a large number of intelligent products in the fields of video...

AI Technical Summary

Problems solved by technology

[0004] The optimization of target tracking algorithm in network, model, feature extraction and other aspects has achieved remarkable results, but the needs of drone shooting may appear in various complex environments, In some specific scenes, high-quality follow-up shots cannot be guaranteed, especially when the target is completely occluded or the background is similar to the target, it is easy to lose the target, causing the target to deviate from the shooting center or exceed the shooting angle of view

Since only the color, texture and other information of the surface of the object under a single viewing angle can be obtained from the RGB video image, 2D visual tracking has great limitations.

Images