Active one-class classification of remote sensing image

Zhigang Liu; Zhichao Sun

doi:10.1117/12.816115

29 December 2008 Active one-class classification of remote sensing image

Zhigang Liu, Zhichao Sun

Proceedings Volume 7285, International Conference on Earth Observation Data Processing and Analysis (ICEODPA); 72850L (2008) https://doi.org/10.1117/12.816115
Event: International Conference on Earth Observation Data Processing and Analysis, 2008, Wuhan, China

Abstract

In many remote sensing image classification applications, interest focuses on a specific land-cover class. In these cases, one-class classification (OCC) approach is appropriate, because one classifier can be trained with samples of target class and just few or no samples of classes that are not of interest are required. However, it is always hard to build a training sample set effectively to represent the target class completely. In this paper, an active learning is introduced for OCC based on support vector data description (SVDD). In active SVDD learning, a SVDD classifier is started with a small size of training samples and an expert is asked to label supplementary training data by asking only for the labels of the most informative, unlabeled examples. Thus, it is possible to build a training sample set effectively to represent the target class completely. The effectiveness of active SVDD is proved by preliminary experiments.

Citation Download Citation

Zhigang Liu and Zhichao Sun "Active one-class classification of remote sensing image", Proc. SPIE 7285, International Conference on Earth Observation Data Processing and Analysis (ICEODPA), 72850L (29 December 2008); https://doi.org/10.1117/12.816115

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available