Deep Learning for In-Orbit Cloud Segmentation and Classification in Hyperspectral Satellite Data

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26220%2F24%3APU155003" target="_blank" >RIV/00216305:26220/24:PU155003 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1109/ICFSP62546.2024.10785468" target="_blank" >https://doi.org/10.1109/ICFSP62546.2024.10785468</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICFSP62546.2024.10785468" target="_blank" >10.1109/ICFSP62546.2024.10785468</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Deep Learning for In-Orbit Cloud Segmentation and Classification in Hyperspectral Satellite Data

Popis výsledku v původním jazyce

This article explores the latest Convolutional Neural Networks (CNNs) for cloud detection aboard hyperspectral satel-lites. The performance of the 1D CNN (1D-Justo-LiuNet) and two 2D CNNs (nnU-net and 2D-Justo-UNet-Simple) for cloud segmentation and classification is assessed. Evaluation criteria include precision and computational efficiency for in-orbit deployment. Experiments utilize NASA's EO-1 Hyperion data, with different spectral channel numbers after Principal Component Analysis. Results indicate that 1D-Justo-LiuNet achieves the highest accu-racy, outperforming 2D CNNs, while maintaining compactness with larger spectral channel sets, albeit with increased inference times. However, the performance of 1D CNN degrades with significant channel reduction. In this context, the 2D-Justo-UNet-Simple offers a good balance for in-orbit deployment, considering precision, memory, and time costs. While nnU-net is suitable for on-ground processing, deployment of lightweight 1D-Justo-LiuNet is recommended for high-precision applications. Alternatively, lightweight 2D-Justo-UNet-Simple balanced better the computational cost and precision for in-orbit deployment.

Název v anglickém jazyce

Deep Learning for In-Orbit Cloud Segmentation and Classification in Hyperspectral Satellite Data

Popis výsledku anglicky

This article explores the latest Convolutional Neural Networks (CNNs) for cloud detection aboard hyperspectral satel-lites. The performance of the 1D CNN (1D-Justo-LiuNet) and two 2D CNNs (nnU-net and 2D-Justo-UNet-Simple) for cloud segmentation and classification is assessed. Evaluation criteria include precision and computational efficiency for in-orbit deployment. Experiments utilize NASA's EO-1 Hyperion data, with different spectral channel numbers after Principal Component Analysis. Results indicate that 1D-Justo-LiuNet achieves the highest accu-racy, outperforming 2D CNNs, while maintaining compactness with larger spectral channel sets, albeit with increased inference times. However, the performance of 1D CNN degrades with significant channel reduction. In this context, the 2D-Justo-UNet-Simple offers a good balance for in-orbit deployment, considering precision, memory, and time costs. While nnU-net is suitable for on-ground processing, deployment of lightweight 1D-Justo-LiuNet is recommended for high-precision applications. Alternatively, lightweight 2D-Justo-UNet-Simple balanced better the computational cost and precision for in-orbit deployment.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/FW09020069" target="_blank" >FW09020069: Palubný družicový systém založený na umelej inteligencii pre efektívnu filtráciu a spracovanie hyperspektrálnych dát</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2024

Kód důvěrnosti údajů

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

Název statě ve sborníku

2024 9th International Conference on Frontiers of Signal Processing (ICFSP)

ISBN

979-8-3503-5323-5

ISSN

—

e-ISSN

—

Počet stran výsledku

5

Strana od-do

68-72

Název nakladatele

IEEE

Místo vydání

neuveden

Místo konání akce

Paříž

Datum konání akce

12. 9. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—