Lightweight multiscale information aggregation network for land cover land use semantic segmentation from remote sensing images

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27730%2F25%3A10258603" target="_blank" >RIV/61989100:27730/25:10258603 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.nature.com/articles/s41598-025-07908-4" target="_blank" >https://www.nature.com/articles/s41598-025-07908-4</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1038/s41598-025-07908-4" target="_blank" >10.1038/s41598-025-07908-4</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Lightweight multiscale information aggregation network for land cover land use semantic segmentation from remote sensing images

Popis výsledku v původním jazyce

Land Cover and Land Use (LCLU) segmentation plays a fundamental role in various remote sensing applications, including environmental monitoring, urban planning, and disaster management. Traditional models often face limitations in real-time processing and deployment on resource-constrained devices due to their high computational requirements. This paper presents a lightweight neural network designed to address these challenges by integrating dense dilated convolutions with pyramid depthwise convolutions for multiscale feature extraction. The proposed encoder-decoder architecture utilizes dense connections to aggregate spatial and contextual information across different resolutions, enhancing segmentation accuracy while minimizing computational overhead. The model&apos;s performance was rigorously evaluated using the NITRDrone and UDD6 datasets, demonstrating a segmentation accuracy of 94.8%, with a significantly reduced parameter count compared to state-of-the-art methods. The compact design of the network facilitates its implementation on low-power devices, enabling real-time LCLU analysis across diverse environmental conditions. This work underscores the potential of lightweight neural networks to advance remote sensing image processing, offering scalable and efficient solutions for practical applications in geospatial analysis.

Název v anglickém jazyce

Lightweight multiscale information aggregation network for land cover land use semantic segmentation from remote sensing images

Popis výsledku anglicky

Land Cover and Land Use (LCLU) segmentation plays a fundamental role in various remote sensing applications, including environmental monitoring, urban planning, and disaster management. Traditional models often face limitations in real-time processing and deployment on resource-constrained devices due to their high computational requirements. This paper presents a lightweight neural network designed to address these challenges by integrating dense dilated convolutions with pyramid depthwise convolutions for multiscale feature extraction. The proposed encoder-decoder architecture utilizes dense connections to aggregate spatial and contextual information across different resolutions, enhancing segmentation accuracy while minimizing computational overhead. The model&apos;s performance was rigorously evaluated using the NITRDrone and UDD6 datasets, demonstrating a segmentation accuracy of 94.8%, with a significantly reduced parameter count compared to state-of-the-art methods. The compact design of the network facilitates its implementation on low-power devices, enabling real-time LCLU analysis across diverse environmental conditions. This work underscores the potential of lightweight neural networks to advance remote sensing image processing, offering scalable and efficient solutions for practical applications in geospatial analysis.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10700 - Other natural sciences

Projekt

—

Návaznosti

O - Projekt operacniho programu

Rok uplatnění

2025

Kód důvěrnosti údajů

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

Název periodika

Scientific Reports

ISSN

2045-2322

e-ISSN

2045-2322

Svazek periodika

15

Číslo periodika v rámci svazku

1

Stát vydavatele periodika

GB - Spojené království Velké Británie a Severního Irska

Počet stran výsledku

12

Strana od-do

nestránkováno

Kód UT WoS článku

001553526200007

EID výsledku v databázi Scopus

—