Identifying saltcedar with hyperspectral data and support vectormachines

Abstract

Saltcedar (Tamarix spp.) are a group of dense phreatophytic shrubs and trees that are invasive to riparian areas throughout the United States. This study determined the feasibility of using hyperspectral data and a support vector machine (SVM) classifier to discriminate saltcedar from other cover types in west Texas. Spectral measurements were collected with a ground-based hyperspectral spectroradiometer (spectral range 350–2500 nm) in December 2008 and April 2009. Spectral data consisting of 1698 spectral bands (400–1349, 1441–1789, 1991–2359 nm) were subjected to a support vector machine classification to differentiate saltcedar from other vegetative and non-vegetative classes. For both dates, a linear kernel model with a C value (error penalty) of 100 was found optimum for separating saltcedar from the other classes. It identified saltcedar with accuracies ranging from 95% to 100%. Findings support further exploration of hyperspectral remote sensing technology and SVM classifiers for differentiating saltcedar from other cover types.

Keywords:

• support vector machine
• hyperspectral
• saltcedar
• Tamarix spp.

Acknowledgements

The authors are grateful to Isabel Cavazos, Jim Forward, Alfredo Gomez, and Juan Ramos for their assistance in this study.

Notes

Note: Trade names are included for the benefit of the reader and do not imply endorsement of or preference for the products listed by the United States Department of Agriculture.

Additional information

Notes on contributors

Reginald S. Fletcher

This work was authored as part of the Contributors' official duties as Employees of the United States Government and is therefore a work of the United States Government. In accordance with 17 U.S.C. 105, no copyright protection is available for such works under U.S. Law.