Skip to main content
ARS Home » Pacific West Area » Logan, Utah » Poisonous Plant Research » Research » Publications at this Location » Publication #299467
Title: Discrimination of grassland species and their classification in botanical families by laboratory scale hyperspectral imaging NIR: preliminary results

Author
item DALE, LAURA - University Of Liege
item THEWIS, ANDRE - University Of Liege
item BOUDRY, CHRISTELLE - University Of Liege
item ROTAR, IOAN - University Of Agricultural Sciences And Veterinary Medicine - Romania
item PACURAR, FLORIN - University Of Agricultural Sciences And Veterinary Medicine - Romania
item ABBAS, OUISSAM - Gembloux Agricultural University
item DARDENNE, PIERRE - Gembloux Agricultural University
item BAETEN, VINCENT - Gembloux Agricultural University
item Pfister, James
item FERNANDEZ PIERNA, JUAN - Gembloux Agricultural University

Submitted to: Talanta
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/4/2013
Publication Date: 11/15/2013
Citation: Dale, L.M., Thewis, A., Boudry, C., Rotar, I., Pacurar, F.S., Abbas, O., Dardenne, P., Baeten, V., Pfister, J.A., Fernandez Pierna, J.A. 2013. Discrimination of grassland species and their classification in botanical families by laboratory scale hyperspectral imaging NIR: preliminary results. Talanta. 116:149-154.

Interpretive Summary: Current methods of measuring the botanical composition of pastures and rangelands are labor intensive and costly. The work sought to use modern imaging techniques to determine the composition of pastures, with a special focus on potentially toxic plant species. The study used on-line hyperspectral imaging to discriminate taxonomic plant families comprised of different grassland species. Plants were collected from semi-natural meadows of the National Apuseni Park, Apuseni Mountains, Gârda area (Romania) according to botanical families. Species within the Poacea family and Other Botanical Families could be distinguished with greater accuracy than those species in the Fabacea family. A correct classification rate of 99% was obtained in the assignment of the various species to the proper family. This work is a first step for the development of a complete procedure for the detection and quantification of possible toxic plant species in semi-natural meadows used by grazing animals.

Technical Abstract: The objective of this study was to discriminate by on-line hyperspectral imaging, taxonomic plant families comprised of different grassland species. Plants were collected from semi-natural meadows of the National Apuseni Park, Apuseni Mountains, Gârda area (Romania) according to botanical families. Species within the Poacea family and Other Botanical Families could be distinguished (R2 = 0.91 and 0.90, respectively) with greater accuracy than those species in the Fabacea family (R2 = 0.60). Chemometric tools such as PLS-DA were used to discriminate distinct grassland species, and assign the different species to botanical families. A correct classification rate of 99% was obtained in the assignment of the various species to the proper family. This work is a first step for the development of a complete procedure for the detection and quantification of possible toxic plant species in semi-natural meadows used by grazing animals.