An Efficient Adjacent Frame Fusion Mechanism for Airborne Visual Object Detection
Open Access
Abstract
With the continuous advancement of drone technology, drones are demonstrating a trend toward autonomy and clustering. The detection of airborne objects from the perspective of drones is critical for addressing threats posed by aerial targets and ensuring the safety of drones in the flight process. Despite the rapid advancements in general object detection technology in recent years, the task of object detection from the unique perspective of drones remains a formidable challenge. In order to tackle this issue, our research presents a novel and efficient mechanism for adjacent frame fusion to enhance the performance of visual object detection in airborne scenarios. The proposed mechanism primarily consists of two modules: a feature alignment fusion module and a background subtraction module. The feature alignment fusion module aims to fuse features from aligned adjacent frames and key frames based on their similarity weights. The background subtraction module is designed to compute the difference between the foreground features extracted from the key frame and the background features obtained from the adjacent frames. This process enables a more effective enhancement of the target features. Given that this method can significantly enhance performance without a substantial increase in parameters and computational complexity, by effectively leveraging the feature information from adjacent frames, we refer to it as an efficient adjacent frame fusion mechanism. Experiments conducted on two challenging datasets demonstrate that the proposed method achieves superior performance compared to existing algorithms.Keywords
Funding Information
- Comprehensive Research Project on Equipment (WJ20211A030131, ZZKY20223105, WJY202209, WYY202304, JYWJ2023B003)
- Independent Propositional Project of PAP (WJ20211A030131, ZZKY20223105, WJY202209, WYY202304, JYWJ2023B003)
- Basic Frontier Innovation Project at the Engineering University of PAP (WJ20211A030131, ZZKY20223105, WJY202209, WYY202304, JYWJ2023B003)
- Applied Research Advancement Project in Engineering University of PAP (WJ20211A030131, ZZKY20223105, WJY202209, WYY202304, JYWJ2023B003)
- Graduate Student Sponsored Project (WJ20211A030131, ZZKY20223105, WJY202209, WYY202304, JYWJ2023B003)
This publication has 40 references indexed in Scilit:
- LSTFE-Net: Long Short-Term Feature Enhancement Network for Video Small Object Detection2023
- TransVisDrone: Spatio-Temporal Transformer for Vision-based Drone-to-Drone Detection in Aerial Videos2023
- Security Threats Analysis of the Unmanned Aerial Vehicle System2021
- Fast, Reliable, and Secure Drone Communication: A Comprehensive SurveyIEEE Communications Surveys & Tutorials, 2021
- Deformable Non-Local Network for Video Super-ResolutionIEEE Access, 2019
- Classifications, applications, and design challenges of drones: A reviewProgress in Aerospace Sciences, 2017
- Multi-target detection and tracking from a single camera in Unmanned Aerial Vehicles (UAVs)2016
- Visual prediction: Psychophysics and neurophysiology of compensation for time delaysBehavioral and Brain Sciences, 2008
- Background subtraction techniques: a review2004
- Eye movements and visible persistence explain the mislocalization of the final position of a moving targetVision Research, 2000