Cluster-based active learning for compact image classification

Devis Tuia*, Mikhail Kanevski, Jordi Muñoz Marí, Gustavo Camps-Valls

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference paper

9 Citations (Scopus)

Abstract

In this paper, we consider active sampling to label pixels grouped with hierarchical clustering. The objective of the method is to match the data relationships discovered by the clustering algorithm with the user's desired class semantics. The first is represented as a complete tree to be pruned and the second is iteratively provided by the user. The active learning algorithm proposed searches the pruning of the tree that best matches the labels of the sampled points. By choosing the part of the tree to sample from according to current pruning's uncertainty, sampling is focused on most uncertain clusters. This way, large clusters for which the class membership is already fixed are no longer queried and sampling is focused on division of clusters showing mixed labels. The model is tested on a VHR image in a multiclass classification setting. The method clearly outperforms random sampling in a transductive setting, but cannot generalize to unseen data, since it aims at optimizing the classification of a given cluster structure.

Original languageEnglish
Title of host publication2010 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2010
Pages2824-2827
Number of pages4
ISBN (Electronic)9781424495665, 9781424495641
DOIs
Publication statusPublished - 2010
Externally publishedYes
Event2010 30th IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2010 - Honolulu, HI, United States
Duration: 25 Jul 201030 Jul 2010

Publication series

NameInternational Geoscience and Remote Sensing Symposium (IGARSS)

Conference

Conference2010 30th IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2010
CountryUnited States
CityHonolulu, HI
Period25/07/1030/07/10

Fingerprint Dive into the research topics of 'Cluster-based active learning for compact image classification'. Together they form a unique fingerprint.

Cite this