Unmanned aerial vehicle photoelectric tracking dynamic exposure method and system facing complex sky background, terminal and medium

CN122934073APending Publication Date: 2026-10-16HEFEI GUOXIAN HOLDINGS CO LTD
0 Cites 0 Cited by

Patent Information

Application Number
CN202610989025.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-03
Publication Date
2026-10-16

AI Technical Summary

Technical Problem

这使得后续完全依赖于图像特征的AI目标检测与跟踪算法因有效信息丢失而失效，造成跟踪中断

Benefits of technology

[0040] 1. The UAV photoelectric tracking dynamic exposure method provided by this invention for complex sky backgrounds calculates the illumination contrast ratio between the global average brightness and the target's own average brightness in real time. When the illumination contrast ratio is greater than a first preset threshold, the current scene is determined to be a strong backlight state where the brightness of the sky background far exceeds the brightness of the target. Based on this determination, the traditional automatic exposure closed loop, which aims to balance global image quality, is immediately cut off, and a dedicated strong backlight forced compensation mode is forcibly triggered. Logically, this realizes that the exposure strategy changes from taking into account both the background and the target to prioritizing the target and sacrificing the background. This method can ensure that even under extreme backlight conditions such as facing the sun directly or bright clouds, the UAV target can be continuously and clearly imaged, rather than becoming a black silhouette that loses all texture features, thereby improving the robustness and survivability of the AI ​​visual tracking system in complex lighting environments.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application discloses a complex sky background-oriented unmanned aerial vehicle photoelectric tracking dynamic exposure method and system, a terminal and a medium, and relates to the technical field of unmanned aerial vehicles.The technical solution points are as follows: a target reference frame is acquired, and a target average brightness value is calculated; the target reference frame is taken as the center to generate a dynamic photometry envelope area; a global average brightness value is calculated; a light contrast ratio is calculated; when the light contrast ratio is greater than a first preset threshold, a strong back light forced compensation mode is triggered; a feedforward prediction photometry area is taken as the only photometry area for exposure control, so that the exposure parameters of the camera only respond to the brightness change in the feedforward prediction photometry area.The application can ensure that the unmanned aerial vehicle target can be continuously and clearly imaged even under extreme back light conditions of directly facing the sun or bright clouds, instead of losing all texture features of a black silhouette, and the robustness and survivability of an AI vision tracking system in a complex light environment are improved.
Need to check novelty before this filing date? Find Prior Art