Semisupervised Feature Selection for Unbalanced Sample Sets of VHR Images

A semisupervised feature selection method, named asymmetrically local discriminant selection (ALDS), is proposed to evaluate the class separability of unbalanced sample sets from very high resolution (VHR) imagery in an object-oriented classification. In order to cope with class imbalance, ALDS inco...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE geoscience and remote sensing letters Jg. 7; H. 4; S. 781 - 785
Hauptverfasser: Chen, Xi, Fang, Tao, Huo, Hong, Li, Deren
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Piscataway IEEE 01.10.2010
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Asymmetrically local discriminant selection (ALDS)
Asymmetry
class imbalance
Classification
Costs
Face recognition
Filters
graph-based filter model
Image processing
Image resolution
Imagery
Machine learning
Mathematical analysis
Object oriented
Object oriented modeling
Object-oriented programming
objectoriented classification
Remote sensing
Research and development
Satellites
semisupervised feature selection
ISSN:1545-598X, 1558-0571
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:A semisupervised feature selection method, named asymmetrically local discriminant selection (ALDS), is proposed to evaluate the class separability of unbalanced sample sets from very high resolution (VHR) imagery in an object-oriented classification. In order to cope with class imbalance, ALDS incorporates asymmetric misclassification costs of classes into weight matrices. Furthermore, this method locally exploits multiple kinds of relationships between sample pairs to more accurately assess the ability of features in preserving the geometrical and discriminant structures. The experimental results on VHR satellite and airborne imagery attest to the effectiveness and practicability of ALDS.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ObjectType-Article-2
ObjectType-Feature-1
ISSN:1545-598X
1558-0571
DOI:10.1109/LGRS.2010.2048197