Forests 2014, 5(6), 1304-1330; doi:10.3390/f5061304
Article

Evaluation and Comparison of QuickBird and ADS40-SH52 Multispectral Imagery for Mapping Iberian Wild Pear Trees (Pyrus bourgaeana, Decne) in a Mediterranean Mixed Forest

1,* email, 2 and 2
Received: 9 January 2014; in revised form: 15 April 2014 / Accepted: 29 May 2014 / Published: 11 June 2014
(This article belongs to the Special Issue Applications of Remote Sensing to Forestry)
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract: The availability of images with very high spatial and spectral resolution from airborne sensors or those aboard satellites is opening new possibilities for the analysis of fine-scale vegetation, such as the identification and classification of individual tree species. To evaluate the potential of these images, a study was carried out to compare the spatial, spectral and temporal resolution between QuickBird and ADS40-SH52 imagery, in order to discriminate and identify, within the mixed Mediterranean forest, individuals of the Iberian wild pear (Pyrus bourgaeana). This is a typical species of the Mediterranean forest, but its biology and ecology are still poorly known. The images were subjected to different correction processes and data were homogenized. Vegetation classes and individual trees were identified on the images, which were classified from two types of supervised classification (Maximum Likelihood and Support Vector Machines) on a pixel-by-pixel basis. The classification values were satisfactory. The classifiers were compared, and Support Vector Machines was the algorithm that provided the best results in terms of overall accuracy. The QuickBird image showed higher overall accuracy (86.16%) when the Support Vector Machines algorithm was applied. In addition, individuals of Iberian wild pear were discriminated with probability of over 55%, when the Maximum Likelihood algorithm was applied. From the perspective of improving the sampling effort, these results are a starting point for facilitating research on the abundance, distribution and spatial structure of P. bourgaeana at different scales, in order to quantify the conservation status of this species.
Keywords: vegetation mapping; QuickBird imagery; ADS40-SH52 imagery; multispectral image; mosaicing; Maximum Likelihood; Support Vector Machines; accuracy assessment; Iberian wild pear; Pyrus bourgaeana
PDF Full-text Download PDF Full-Text [2182 KB, Updated Version, uploaded 16 June 2014 11:13 CEST]
The original version is still available [2182 KB, uploaded 11 June 2014 13:50 CEST]

Export to BibTeX |
EndNote


MDPI and ACS Style

Arenas-Castro, S.; Fernández-Haeger, J.; Jordano-Barbudo, D. Evaluation and Comparison of QuickBird and ADS40-SH52 Multispectral Imagery for Mapping Iberian Wild Pear Trees (Pyrus bourgaeana, Decne) in a Mediterranean Mixed Forest. Forests 2014, 5, 1304-1330.

AMA Style

Arenas-Castro S, Fernández-Haeger J, Jordano-Barbudo D. Evaluation and Comparison of QuickBird and ADS40-SH52 Multispectral Imagery for Mapping Iberian Wild Pear Trees (Pyrus bourgaeana, Decne) in a Mediterranean Mixed Forest. Forests. 2014; 5(6):1304-1330.

Chicago/Turabian Style

Arenas-Castro, Salvador; Fernández-Haeger, Juan; Jordano-Barbudo, Diego. 2014. "Evaluation and Comparison of QuickBird and ADS40-SH52 Multispectral Imagery for Mapping Iberian Wild Pear Trees (Pyrus bourgaeana, Decne) in a Mediterranean Mixed Forest." Forests 5, no. 6: 1304-1330.

Forests EISSN 1999-4907 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert