Please use this identifier to cite or link to this item:
Title: Satellite remote sensing of wildlife habitats using a multi-scale, object-based, decision-tree classifier (MOOSC)
Authors: Nichol, J 
Wong, MS
Keywords: Habitat mapping
Object-based classification
Image segmentation
Issue Date: 2006
Publisher: WSEAS Press
Source: WSEAS transactions on environment and development, 2006, v. 4, no. 2, p. 353-359 How to cite?
Journal: WSEAS transactions on environment and development 
Abstract: This paper describes an object-based methodology devised to enable accurate and detailed habitat mapping using IKONOS VHR satellite images. The Multi-scale Object Oriented Segmentation with Decision Tree Classification (MOOSC) comprises a suite of image processing techniques applied interactively by the user in response to study area characteristics, and data dimensionality. Because of the inherent flexibility of the method, MOOSC can obtain high mapping accuracy even in the rugged terrain of the present study area combined with the low solar zenith angle of the wintertime images used. The accuracy of 90% obtained for MOOSC increases to 94% if habitat classes are based purely on physical structure rather than life form, ie. if shrub-sized tree seedlings planted on grassy slopes are classified as shrubby grassland, which they resemble physically and spectrally, although logically they belong to the forest class. Thus the MOOSC method is able to achieve similar accuracy to stereo air photo interpretation from large scale, high quality colour photos (which achieved 95% accuracy), but at only one third of the cost.
ISSN: 1790-5079
EISSN: 2224-3496
Appears in Collections:Journal/Magazine Article

View full-text via PolyU eLinks SFX Query
Show full item record

Page view(s)

Last Week
Last month
Citations as of Oct 14, 2018

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.