Support Vector Machine Based Tool for Plant Species Taxonomic Classification

Manimekalai K and Vijaya Ms

Journal of Asian Scientific Research, 2014, vol. 4, issue 4, 159-173

Abstract: Plant species are living things and are generally categorized in terms of Domain, Kingdom, Phylum, Class, Order, Family, Genus and name of Species in a hierarchical fashion. This paper formulates the taxonomic leaf categorization problem as the hierarchical classification task and provides a suitable solution using a supervised learning technique namely support vector machine. Features are extracted from scanned images of plant leaves and trained using SVM. Only class, order, family of plants and species are considered for hierarchical classification in this research work. The trained models corresponding to class - Magnoliopsida, orders - Brassicales, Rosales, families - Cariaceae, Brassicacea, Rosaceae, Rhamnaceae have been used to develop a three level hierarchical classification model for hierarchical classification of plant species and the results are analyzed. A flat multiclass classifier has been built for species which is at last level of hierarchy and the results are analyzed. A user interactive taxonomic tool has developed for displaying taxonomic ranks of species.

Keywords: Pre-processing; Feature extraction; Hierarchical classification; Multi-class classification; Plant species taxonomy; Support vector machine. (search for similar items in EconPapers)
Date: 2014
References: Add references at CitEc
Citations:

Downloads: (external link)
https://archive.aessweb.com/index.php/5003/article/view/3619/5735 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:asi:joasrj:v:4:y:2014:i:4:p:159-173:id:3619

Access Statistics for this article

More articles in Journal of Asian Scientific Research from Asian Economic and Social Society
Bibliographic data for series maintained by Robert Allen ().