Hyperspectral Data Cleaning Towards Camouflage Detection

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F60162694%3AG43__%2F26%3A00564787" target="_blank" >RIV/60162694:G43__/26:00564787 - isvavai.cz</a>

Výsledek na webu

<a href="http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=11061248" target="_blank" >http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=11061248</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICMT65201.2025.11061355" target="_blank" >10.1109/ICMT65201.2025.11061355</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Hyperspectral Data Cleaning Towards Camouflage Detection

Popis výsledku v původním jazyce

Hyperspectral imaging (HSI) is a powerful technique for remote sensing, offering detailed spectral information beyond conventional RGB imaging. This paper presents a method for acquiring and processing hyperspectral data to enhance its applicability in deep learning analysis. We explore visualization strategy under controlled experimental conditions. A key contribution of this study is the introduction of three evaluation criteria: entropy, correlation, and Jaccard score for assessing band pertinence during pre-processing. These criteria help optimize data selection by filtering out noisy or redundant spectral bands, thereby improving the efficiency and accuracy of subsequent analysis. Our results demonstrate that strategic band selection enhances camouflage detection in military and security applications, particularly by leveraging the hyperspectral differentiation of natural and artificial materials. The proposed methodology streamlines hyperspectral data processing, making it more effective for integration with artificial neural networks in various domains, including surveillance, environmental monitoring, and target

Název v anglickém jazyce

Hyperspectral Data Cleaning Towards Camouflage Detection

Popis výsledku anglicky

Hyperspectral imaging (HSI) is a powerful technique for remote sensing, offering detailed spectral information beyond conventional RGB imaging. This paper presents a method for acquiring and processing hyperspectral data to enhance its applicability in deep learning analysis. We explore visualization strategy under controlled experimental conditions. A key contribution of this study is the introduction of three evaluation criteria: entropy, correlation, and Jaccard score for assessing band pertinence during pre-processing. These criteria help optimize data selection by filtering out noisy or redundant spectral bands, thereby improving the efficiency and accuracy of subsequent analysis. Our results demonstrate that strategic band selection enhances camouflage detection in military and security applications, particularly by leveraging the hyperspectral differentiation of natural and artificial materials. The proposed methodology streamlines hyperspectral data processing, making it more effective for integration with artificial neural networks in various domains, including surveillance, environmental monitoring, and target

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20204 - Robotics and automatic control

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

2025 10th International Conference on Military Technologies, ICMT 2025 - Proceedings

ISBN

—

ISSN

2996-4474

e-ISSN

—

Počet stran výsledku

7

Strana od-do

1-7

Název nakladatele

Institute of Electrical and Electronics Engineers Inc.

Místo vydání

Brno

Místo konání akce

Brno, Czech Republic

Datum konání akce

27. 5. 2025

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

001545807300094