SVM-Based Boosting of Active Learning Strategies for Efficient Domain Adaptation
- 16 November 2012
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 5 (5) , 1335-1343
Abstract
We propose a procedure that efficiently adapts a classifier trained on a source image to a target image with similar spectral properties. The adaptation is carried out by adding new relevant training samples with active queries in the target domain following a strategy specifically designed for the case where class distributions have shifted between the two acquisitions. In fact, the procedure consists of two nested algorithms. An active selection of the pixels to be labeled is performed on a set of candidates of the target image in order to select the most informative pixels. Along the inclusion of the pixels to the training set, the weights associated with these samples are iteratively updated using different criteria, depending on their origin (source or target image). We study this adaptation framework in combination with a SVM classifier accepting instance weights. Experiments on two VHR QuickBird images and on a hyperspectral AVIRIS image prove the validity of the proposed adaptive approach with respect to existing techniques not involving any adjustments to the target domain.Keywords
This publication has 22 references indexed in Scilit:
- A Survey of Active Learning Algorithms for Supervised Remote Sensing Image ClassificationIEEE Journal of Selected Topics in Signal Processing, 2011
- Spatially Adaptive Classification of Land Cover With Remote Sensing DataIEEE Transactions on Geoscience and Remote Sensing, 2011
- Active Supervised Domain Adaptation2011
- Batch-Mode Active-Learning Methods for the Interactive Classification of Remote Sensing ImagesIEEE Transactions on Geoscience and Remote Sensing, 2010
- Adaptive Classification for Hyperspectral Image Data Using Manifold Regularization Kernel MachinesIEEE Transactions on Geoscience and Remote Sensing, 2010
- Efficient SVM training with reduced weighted samples2010
- A Survey on Transfer LearningIEEE Transactions on Knowledge and Data Engineering, 2009
- Mean Map Kernel Methods for Semisupervised Cloud ClassificationIEEE Transactions on Geoscience and Remote Sensing, 2009
- An Efficient Active Learning Algorithm with Knowledge Transfer for Hyperspectral Data Analysis2008
- Adapting SVM Classifiers to Data with Shifted Distributions2007