Low-illumination unmanned aerial vehicle target tracking system and method

By combining the target tracking module within the edge vision camera of the UAV with a lighting-level guided dual-path interactive feature enhancement network and temporal optimization, the problem of target tracking drift and interruption in complex dynamic scenarios under low illumination is solved, achieving high-precision, real-time UAV target tracking.

CN122434979APending Publication Date: 2026-07-21TONGJI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
TONGJI UNIV
Filing Date
2026-06-24
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

Existing drone target tracking algorithms are susceptible to lighting interference, day-night cross-domain feature shift, and motion blur in low-light complex dynamic scenarios, leading to tracking drift and interruption. Furthermore, they are difficult to balance high precision and real-time requirements on drone edge embedded platforms with limited computing resources.

Method used

Design a low-light UAV target tracking system, including a target tracking module deployed in the edge vision camera of an industrial-grade UAV. Through a backbone network trained with domain adaptation and a tracking result output network, combined with an illumination-level guided dual-path interactive feature enhancement network, a temporal optimization and enhancement layer, and a dynamic sparsity and semantic constraint alignment network, feature enhancement, domain adaptation, compensation calibration and lightweight deployment are achieved.

Benefits of technology

It effectively improves the accuracy and stability of target tracking in complex dynamic scenarios with low illumination, meets the real-time and lightweight requirements of UAV edge deployment, solves the problems of illumination interference, cross-domain feature offset and motion blur, and provides high-purity and robust target feature support.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122434979A_ABST
    Figure CN122434979A_ABST
Patent Text Reader

Abstract

The application relates to a low-illumination unmanned aerial vehicle target tracking system and method, which comprises a target tracking module obtained through field self-adaption training, is used for processing and analyzing low-illumination complex dynamic images, and outputs a target tracking result; the target tracking module comprises a pre-trained backbone network and a tracking result output network, the two are connected with an illumination level-oriented double-path interactive feature enhancement network, a time sequence optimization and enhancement layer, and a field self-adaption scoring network and a dynamic sparse and semantic constraint alignment network are arranged; after the field self-adaption training process is completed, model compression is completed through a field adaptation sparse distillation lightweight mechanism, so that the target tracking module is deployed to an industrial-grade unmanned aerial vehicle edge visual camera. Compared with the prior art, the application can solve the problems of tracking drift and interruption caused by illumination interference, day-night cross-domain feature deviation and motion blur in the low-illumination complex dynamic scene, and can also consider the tracking performance and lightweight deployment requirements.
Need to check novelty before this filing date? Find Prior Art