Title :
Unsupervised retraining of a maximum likelihood classifier for the analysis of multitemporal remote sensing images
Author :
Bruzzone, Lorenzo ; Prieto, Diego Fernàndez
Author_Institution :
Dept. of Civil & Environ. Eng., Trento Univ., Italy
fDate :
2/1/2001 12:00:00 AM
Abstract :
An unsupervised retraining technique for a maximum likelihood (ML) classifier is presented. The proposed technique allows the classifier´s parameters, obtained by supervised learning on a specific image, to be updated in a totally unsupervised way on the basis of the distribution of a new image to be classified. This enables the classifier to provide a high accuracy for the new image even when the corresponding training set is not available
Keywords :
geophysical techniques; image classification; image sequences; maximum likelihood estimation; remote sensing; terrain mapping; geophysical measurement technique; image classification; image sequence; land surface; maximum likelihood classifier; multitemporal images; remote sensing; terrain mapping; unsupervised retraining; Availability; Image analysis; Image sensors; Maximum likelihood estimation; Pixel; Remote monitoring; Remote sensing; Sensor phenomena and characterization; Sensor systems; Supervised learning;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
