Active Learning in the Spatial Domain for Remote Sensing Image Classification

Active learning (AL) algorithms have been proven useful in reducing the number of required training samples for remote sensing applications; however, most methods query samples pointwise without considering spatial constraints on their distribution. This may often lead to a spatially dispersed distr...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	IEEE transactions on geoscience and remote sensing Ročník 52; číslo 5; s. 2492 - 2507
Hlavní autoři:	Stumpf, Andre, Lachiche, Nicolas, Malet, Jean-Philippe, Kerle, Norman, Puissant, Anne
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	New York, NY IEEE 01.05.2014 Institute of Electrical and Electronics Engineers The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:	Active learning (AL) Algorithms Applied geophysics Assessments batch-mode class imbalance Computer Science Earth sciences Earth, ocean, space Entropy Exact sciences and technology ground truth uncertainty image classification Internal geophysics landslide inventory mapping Landslides Landslides & mudslides Radio frequency Remote sensing Sampling Sensitivity analysis Signal and Image Processing Spatial distribution spatial information Terrain factors Training Training data Uncertainty class imbalance image classification landslide inventory mapping Active learning (AL) ground truth uncertainty batch-mode spatial information algorithms ground truth sensitivity analysis Space remote sensing sampling satellite measurements classification inventory trees spatial distribution interpretation landslides imagery uncertainties errors
ISSN:	0196-2892, 1558-0644
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	Active learning (AL) algorithms have been proven useful in reducing the number of required training samples for remote sensing applications; however, most methods query samples pointwise without considering spatial constraints on their distribution. This may often lead to a spatially dispersed distribution of training points unfavorable for visual image interpretation or field surveys. The aim of this study is to develop region-based AL heuristics to guide user attention toward a limited number of compact spatial batches rather than distributed points. The proposed query functions are based on a tree ensemble classifier and combine criteria of sample uncertainty and diversity to select regions of interest. Class imbalance, which is inherent to many remote sensing applications, is addressed through stratified bootstrap sampling. Empirical tests of the proposed methods are performed with multitemporal and multisensor satellite images capturing, in particular, sites recently affected by large-scale landslide events. The assessment includes an experimental evaluation of the labeling time required by the user and the computational runtime, and a sensitivity analysis of the main algorithm parameters. Region-based heuristics that consider sample uncertainty and diversity are found to outperform pointwise sampling and region-based methods that consider only uncertainty. Reference landslide inventories from five different experts enable a detailed assessment of the spatial distribution of remaining errors and the uncertainty of the reference data.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 ObjectType-Article-2 ObjectType-Feature-1 content type line 23
ISSN:	0196-2892 1558-0644
DOI:	10.1109/TGRS.2013.2262052