Using the U-Net-like deep convolutional neural networks for precise tree recognition in very high resolution RGB (red, green, blue) satellite images

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F67985939%3A_____%2F21%3A00547970" target="_blank" >RIV/67985939:_____/21:00547970 - isvavai.cz</a>

Result on the web

<a href="http://hdl.handle.net/11104/0324109" target="_blank" >http://hdl.handle.net/11104/0324109</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.3390/f12010066" target="_blank" >10.3390/f12010066</a>

Result language

angličtina

Original language name

Using the U-Net-like deep convolutional neural networks for precise tree recognition in very high resolution RGB (red, green, blue) satellite images

Original language description

In this study, we have demonstrated an example of the use of the DL algorithm, relying on the proposed U-Net-like CNN architecture for the recognition of particular tree species in high-resolution RGB satellite images. We showed that traditional pixel-based ML approaches are influenced by false-positive decisions when objects captured in satellite images have the same color composition as tree crowns.

Czech name

—

Czech description

—

Type

J_imp - Article in a specialist periodical, which is included in the Web of Science database

CEP classification

—

OECD FORD branch

10618 - Ecology

Project

—

Continuities

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Publication year

2021

Confidentiality

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

Name of the periodical

Forests

ISSN

1999-4907

e-ISSN

1999-4907

Volume of the periodical

12

Issue of the periodical within the volume

1

Country of publishing house

CH - SWITZERLAND

Number of pages

17

Pages from-to

66

UT code for WoS article

000610224900001

EID of the result in the Scopus database

2-s2.0-85099743219