Sie können Bookmarks mittels Listen verwalten, loggen Sie sich dafür bitte in Ihr SLUB Benutzerkonto ein.
Medientyp:
E-Artikel
Titel:
Identification for genetic differentiation of scots pine seedlings using hyperspectral imaging
Beteiligte:
Juodkienė, Vytautė;
Danusevičius, Darius;
Mozgeris, Gintautas
Erschienen:
Lithuanian Academy of Sciences, 2017
Erschienen in:
Žemės ūkio mokslai, 24 (2017) 2
Sprache:
Nicht zu entscheiden
DOI:
10.6001/zemesukiomokslai.v24i2.3499
ISSN:
2424-4120;
1392-0200
Entstehung:
Anmerkungen:
Beschreibung:
Genetic diversity is an important component of biological diversity and therefore there are attempts to search for effective solutions that increase the effectiveness of genetic diversity assessment methods. The article represents the study that aims for genetic differentiation of Scots pine seedlings based on data, obtained using the hyperspectral imaging in the laboratory. Hyperspectral imaging is the technology that combines image and spectroscopy. Seedlings of 4 distant Scots pine populations from Lithuania (LT), Russia (RU), Spain (SP) and Finland (FIN) were selected for testing. One year old seedlings grown in containers under laboratory conditions were scanned with a hyperspectral camera with an integrated sensitive spectroscope that splits the electromagnetic radiation into 955 spectrum bands in the range of 400 to 1000 nm. The seedlings of different populations had a different reflectance intensity of the near-infrared radiation (750–955 nm). Meanwhile, in the visible spectrum area (400–700 nm) their spectral properties were similar. The Analysis of the Principal Components (PCA) showed that for the differentiating of seedling populations five spectrum bands might be used, which accumulated significant information in the range of 700–720 nm. It was found that the genetic origin classes of the Scots pine can be distinguished reliably (in 85.2% of cases) using the Discriminant Analysis of the seedling’s spectral data. The most geographically distant Spanish and Finnish populations are the most mutually separated populations. The results demonstrate the potential of defining tree genetic properties/origin remotely by using hyperspectral scanners installed in aircrafts.