A low-altitude follow-up system and method aiming at the mobile ground object by unmanned aircraft

Abstract

The invention relates to a low-attitude tracking system of unmanned aircraft on ground movable target and relative method. Wherein, said system comprises carrier system and ground system; the carrier system is formed by camera, table, carrier image sender and carrier data transmitter; the ground system is formed by image receiver, image collector, PC, and ground data transmitter; the tracking method comprises that extracting target via threshold value division, calculating out the position of target gravity center to find its position in track field, and manually interfering the track state. It uses several threshold values, which can be replaced to improve the track stability. The invention has high real-time property and stability.

Description

technical field [0001] The invention relates to visual tracking based on an unmanned aerial vehicle, in particular to a low-altitude tracking system and method for an unmanned rotorcraft or an unmanned airship to a moving target on the ground. Background technique [0002] Vision-based target recognition and tracking is a key technology in visual surveillance systems. In a specific monitoring system, the purpose of visual tracking is mainly to assist monitoring personnel to track and monitor suspicious targets, expand the scope of monitoring, reduce the labor intensity of monitoring personnel, and enhance the reliability of monitoring. [0003] The current vision-based surveillance systems are mainly concentrated in some public places, important departments, etc. Surveillance cameras are usually fixed installations, and the range of surveillance scenes is relatively limited. Surveillance personnel operate the camera movement (pan-tilt-zoom) to track suspicious targets , bas...

AI Technical Summary

Problems solved by technology

In order to expand the monitoring range, multi-camera monitoring is often used. The monitoring ranges of adjacent cameras partially overlap, the scene is fixed, and the cameras cannot track and monitor suspicious targets.

At the same time, the existing automatic identification and tracking technology first establishes the background model of the monitored place, and then divides the moving area and tracks the moving area. These characteristics are not suitable for long-distance and large-scale mobile tracking of ground moving targets.

Images