Combining transductive and active learning to improve object-based classification of remote sensing images. Issue 4 (2nd April 2016)

• Record Type: Journal Article
• Authors: Güttler, Fabio N.; Ienco, Dino; Poncelet, Pascal; Teisseire, Maguelonne
• Is Part Of: Remote sensing letters. Volume 7:Issue 4(2016)
• Page Start: 358
• Page End: 367
• Publication Date: 2016-04-02
• Subjects: Remote sensing -- Periodicals; Remote sensing; Periodicals; 621.3678
• DOI: 10.1080/2150704X.2016.1142678
• Languages: English
• ISSNs: 2150-704X

Abstract

ABSTRACT: In this letter, we propose a new active transductive learning (ATL) framework for object-based classification of satellite images. The framework couples graph-based label propagation with active learning (AL) to exploit positive aspects of the two learning settings. The transductive approach considers both labelled and unlabelled image objects to perform its classification as they are all available at training time while the AL strategy smartly guides the construction of the training set employed by the learner. The proposed framework was tested in the context of a land cover classification task using RapidEye optical imagery. A reference land cover map was elaborated over the whole study area in order to get reliable information about the performance of the ATL framework. The experimental evaluation underlines that, with a reasonable amount of training data, our framework outperforms state of the art classification methods usually employed in the field of remote sensing.