20 juin 2013
Ce document est lié à :
info:eu-repo/grantAgreement/EC/FP7/263479
info:eu-repo/semantics/openAccess
Geoff Smith, « Hybrid Pixel- and Object-Based Approach to Habitat Condition Monitoring. GI_Forum 2013 – Creating the GISociety| », Elektronisches Publikationsportal der Österreichischen Akademie der Wissenschafte, ID : 10670/1.ll0wxo
The analysis of EO data is generally either pixel- or object-based, but this paper outlines a hybrid approach which maximises the information content of the image at its finest spatial resolution (the pixel) and then analyses this information at the scale of landscape features (objects). This approach is useful where the characteristics within features are important to their assessment and is highly relevant in the current context of habitat legislation which focuses on changes in condition as well as extent. A vital component of the approach is local expert knowledge to improve both the results and the engagement of key user communities in the process. Examples are provided for monitoring a 5 year wetland restoration programme where it was aimed to demonstrate a viable operational workflow.