Share Email Print

Proceedings Paper

Transfer component analysis for domain adaptation in image classification
Author(s): Giona Matasci; Michele Volpi; Devis Tuia; Mikhail Kanevski
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

This contribution studies a feature extraction technique aiming at reducing differences between domains in image classification. The purpose is to find a common feature space between labeled samples issued from a source image and test samples belonging to a related target image. The presented approach, Transfer Component Analysis, finds a transformation matrix performing a joint mapping of the two domains by minimizing a probability distribution distance measure, the Maximum Mean Discrepancy criterion. When predicting on a target image, such a projection allows to apply a supervised classifier trained exclusively on labeled source pixels mapped in this common latent subspace. Promising results are observed on a urban scene captured by a hyperspectral image. The experiments reveal improvements with respect to a standard classification model built on the original source image and other feature extraction techniques.

Paper Details

Date Published: 26 October 2011
PDF: 9 pages
Proc. SPIE 8180, Image and Signal Processing for Remote Sensing XVII, 81800F (26 October 2011); doi: 10.1117/12.898229
Show Author Affiliations
Giona Matasci, Univ. de Lausanne (Switzerland)
Michele Volpi, Univ. de Lausanne (Switzerland)
Devis Tuia, Univ de València (Spain)
Ecole Polytechnique Fédérale de Lausanne (Switzerland)
Mikhail Kanevski, Univ. de Lausanne (Switzerland)

Published in SPIE Proceedings Vol. 8180:
Image and Signal Processing for Remote Sensing XVII
Lorenzo Bruzzone, Editor(s)

© SPIE. Terms of Use
Back to Top