Title :
Enhanced algorithm performance for land cover classification from remotely sensed data using bagging and boosting
Author :
Chan, Jonathan Cheung-Wai ; Huang, Chengquan ; DeFries, Ruth
Author_Institution :
Dept. of Geography, Maryland Univ., College Park, MD, USA
fDate :
3/1/2001 12:00:00 AM
Abstract :
Two ensemble methods, bagging and boosting, were investigated for improving algorithm performance. The authors´ results confirmed the theoretical explanation of L. Breiman (1996) that bagging improves unstable, but not stable, learning algorithms. While boosting enhanced accuracy of a weak learner, its behavior is subject to the characteristics of each learning algorithm
Keywords :
geophysical signal processing; geophysical techniques; image classification; learning (artificial intelligence); terrain mapping; accuracy; algorithm; bagging; boosting; enhanced performance; ensemble method; geophysical measurement technique; image classification; image processing; land cover classification; land surface; learning algorithm; remote sensing; terrain mapping; weak learner; Aggregates; Bagging; Boosting; Geography; Image resolution; MODIS; Pressing; Radiometry; Sampling methods; Voting;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
