DocumentCode :
2123341
Title :
Supervised image classification based on adaboost with contextual weak classifiers
Author :
Nishii, Ryuei ; Eguchi, Shinto
Author_Institution :
Graduate Sch. of Math., Kyushu Univ., Fukuoka, Japan
Volume :
2
fYear :
2004
fDate :
20-24 Sept. 2004
Firstpage :
1467
Abstract :
AdaBoost, one of machine learning techniques, is employed for supervised classification of land-cover categories of geostatistical data. We introduce contextual classifiers based on neighboring pixels. First, posterior probabilities are calculated at all pixels. Then, averages of the posteriors in various neighborhoods are calculated, and the averages are used as contextual classifiers. Weights for the classifiers can be determined by minimizing the empirical risk with multiclass. Finally, a linear combination of classifier is obtained. The proposed method is applied to artificial multispectral images and shows an excellent performance similar to the MRF-based classifier with much less computation time.
Keywords :
geophysical signal processing; image classification; learning (artificial intelligence); MRF-based classifier; adaboost; artificial multispectral image; contextual weak classifier; geostatistical data; image classification; land-cover category; linear combination classifier; machine learning techniques; neighboring pixel; posterior probability; supervised classification; Artificial neural networks; Image classification; Machine learning; Mathematics; Multispectral imaging; Pattern recognition; Probability; Support vector machine classification; Support vector machines; Voting;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Geoscience and Remote Sensing Symposium, 2004. IGARSS '04. Proceedings. 2004 IEEE International
Print_ISBN :
0-7803-8742-2
Type :
conf
DOI :
10.1109/IGARSS.2004.1368697
Filename :
1368697
Link To Document :
بازگشت