Title :
An entropy minimization principle for semi-supervised terrain classification
Author :
Guerrero-Curieses, Alicia ; Cid-Sueiro, Jesús
Author_Institution :
Escuela Politecnica Superior, Univ. Carlos III de Madrid, Spain
Abstract :
Analyzing the structure of the family of cost functions that are minimum when the classifier outputs are equal to the class probabilities, we found that all of them can be expressed as sum of a generalized entropy measure and an error component. This suggests a novel algorithm for classification that uses both labeled an unlabeled data and is based on the following idea: use labeled data to minimize the cost function and unlabeled data to minimize the corresponding entropy measure. This entropy minimization principle is applied to terrain classification of Landsat images
Keywords :
image classification; learning (artificial intelligence); minimum entropy methods; neural net architecture; probability; remote sensing; Landsat images; class probabilities; classifier outputs; cost functions; entropy minimization; error component; generalized entropy measure; labeled data; neural network architecture; remote sensing data; semi-supervised learning algorithms; semi-supervised terrain classification; unlabeled data; Cost function; Data mining; Entropy; Image databases; Labeling; Remote sensing; Satellites; Spatial databases; Supervised learning; Training data;
Conference_Titel :
Image Processing, 2000. Proceedings. 2000 International Conference on
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-6297-7
DOI :
10.1109/ICIP.2000.899370