A survey of active learning algorithms for supervised remote sensing image classification

Devis Tuia*, Michele Volpi, Loris Copa, Mikhail Kanevski, Jordi Muñoz-Marí

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

344 Citations (Scopus)


Defining an efficient training set is one of the most delicate phases for the success of remote sensing image classification routines. The complexity of the problem, the limited temporal and financial resources, as well as the high intraclass variance can make an algorithm fail if it is trained with a suboptimal dataset. Active learning aims at building efficient training sets by iteratively improving the model performance through sampling. A user-defined heuristic ranks the unlabeled pixels according to a function of the uncertainty of their class membership and then the user is asked to provide labels for the most uncertain pixels. This paper reviews and tests the main families of active learning algorithms: committee, large margin, and posterior probability-based. For each of them, the most recent advances in the remote sensing community are discussed and some heuristics are detailed and tested. Several challenging remote sensing scenarios are considered, including very high spatial resolution and hyperspectral image classification. Finally, guidelines for choosing the good architecture are provided for new and/or unexperienced user.

Original languageEnglish
Article number5742970
Pages (from-to)606-617
Number of pages12
JournalIEEE Journal on Selected Topics in Signal Processing
Issue number3
Publication statusPublished - Jun 2011
Externally publishedYes


  • active learning
  • hyperspectral
  • image classification
  • support vector machine (SVM)
  • training set definition
  • very high resolution (VHR)

Fingerprint Dive into the research topics of 'A survey of active learning algorithms for supervised remote sensing image classification'. Together they form a unique fingerprint.

Cite this