SpectraSentinel: LightWeight Dual-Stream Real-Time Drone Detection, Tracking and Payload Identification

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Kabir, Shahriar, Rifti, Istiak Ahmmed, Tabib, H. M. Shadman, Rahman, Mushfiqur, Sadi, Sadatul Islam, Adil, Hasnaen, Rumi, Ahmed Mahir Sultan, Haider, Ch Md Rakin
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918107737489408
author Kabir, Shahriar
Rifti, Istiak Ahmmed
Tabib, H. M. Shadman
Rahman, Mushfiqur
Sadi, Sadatul Islam
Adil, Hasnaen
Rumi, Ahmed Mahir Sultan
Haider, Ch Md Rakin
author_facet Kabir, Shahriar
Rifti, Istiak Ahmmed
Tabib, H. M. Shadman
Rahman, Mushfiqur
Sadi, Sadatul Islam
Adil, Hasnaen
Rumi, Ahmed Mahir Sultan
Haider, Ch Md Rakin
contents The proliferation of drones in civilian airspace has raised urgent security concerns, necessitating robust real-time surveillance systems. In response to the 2025 VIP Cup challenge tasks - drone detection, tracking, and payload identification - we propose a dual-stream drone monitoring framework. Our approach deploys independent You Only Look Once v11-nano (YOLOv11n) object detectors on parallel infrared (thermal) and visible (RGB) data streams, deliberately avoiding early fusion. This separation allows each model to be specifically optimized for the distinct characteristics of its input modality, addressing the unique challenges posed by small aerial objects in diverse environmental conditions. We customize data preprocessing and augmentation strategies per domain - such as limiting color jitter for IR imagery - and fine-tune training hyperparameters to enhance detection performance under conditions of heavy noise, low light, and motion blur. The resulting lightweight YOLOv11n models demonstrate high accuracy in distinguishing drones from birds and in classifying payload types, all while maintaining real-time performance. This report details the rationale for a dual-modality design, the specialized training pipelines, and the architectural optimizations that collectively enable efficient and accurate drone surveillance across RGB and IR channels.
format Preprint
id arxiv_https___arxiv_org_abs_2507_22650
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SpectraSentinel: LightWeight Dual-Stream Real-Time Drone Detection, Tracking and Payload Identification
Kabir, Shahriar
Rifti, Istiak Ahmmed
Tabib, H. M. Shadman
Rahman, Mushfiqur
Sadi, Sadatul Islam
Adil, Hasnaen
Rumi, Ahmed Mahir Sultan
Haider, Ch Md Rakin
Computer Vision and Pattern Recognition
The proliferation of drones in civilian airspace has raised urgent security concerns, necessitating robust real-time surveillance systems. In response to the 2025 VIP Cup challenge tasks - drone detection, tracking, and payload identification - we propose a dual-stream drone monitoring framework. Our approach deploys independent You Only Look Once v11-nano (YOLOv11n) object detectors on parallel infrared (thermal) and visible (RGB) data streams, deliberately avoiding early fusion. This separation allows each model to be specifically optimized for the distinct characteristics of its input modality, addressing the unique challenges posed by small aerial objects in diverse environmental conditions. We customize data preprocessing and augmentation strategies per domain - such as limiting color jitter for IR imagery - and fine-tune training hyperparameters to enhance detection performance under conditions of heavy noise, low light, and motion blur. The resulting lightweight YOLOv11n models demonstrate high accuracy in distinguishing drones from birds and in classifying payload types, all while maintaining real-time performance. This report details the rationale for a dual-modality design, the specialized training pipelines, and the architectural optimizations that collectively enable efficient and accurate drone surveillance across RGB and IR channels.
title SpectraSentinel: LightWeight Dual-Stream Real-Time Drone Detection, Tracking and Payload Identification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.22650