Mapping invasive Veratrum album from UAV imagery: neural network training for deep learning novices with no intention of scripting, using a consumer-grade laptop.
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F67985939%3A_____%2F25%3A00647620" target="_blank" >RIV/67985939:_____/25:00647620 - isvavai.cz</a>
Result on the web
<a href="https://www.eo4plantinvasions.cz/wp-content/uploads/2026/01/V14_EARSEL_2025_Srolleru_et_al_poster_lowQ.pdf" target="_blank" >https://www.eo4plantinvasions.cz/wp-content/uploads/2026/01/V14_EARSEL_2025_Srolleru_et_al_poster_lowQ.pdf</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Mapping invasive Veratrum album from UAV imagery: neural network training for deep learning novices with no intention of scripting, using a consumer-grade laptop.
Original language description
The study explores the use of deep learning in remote sensing through ArcGIS Pro for detecting the species Veratrum album from high-resolution UAV imagery. Training data were generated directly from imagery using segmentation and vectorization, with the process streamlined by AutoDL tools. The results achieved high detection accuracy (F1 score 0.90), demonstrating that effective deep learning applications can be implemented with minimal expertise and standard hardware.
Czech name
—
Czech description
—
Classification
Type
O - Miscellaneous
CEP classification
—
OECD FORD branch
10611 - Plant sciences, botany
Result continuities
Project
<a href="/en/project/SS07020317" target="_blank" >SS07020317: Monitoring the spread and management of invasive and expansive species using advanced earth observation methods</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2025
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů