Share Email Print
cover

Proceedings Paper

Adaptive image segmentation neural network: application to Landsat images
Author(s): Jose L. Alba Castro; Susana M. Rey; Laura Docio
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

In this paper we introduce an adaptive image segmentation neural network based on a Gaussian mixture classifier that is able to accommodate unlabeled data in the training process to improve generalization when labeled data is insufficient. The classifier is trained by maximizing the joint-likelihood of features and labels over all the data set (labeled and unlabeled). The classifier builds grey- level images with estimation of class-posteriors (as many images as classes) that feed the segmentation algorithm. The paper is focused on the adaptive classification part of the algorithm. The classification tests are performed over Landsat TM mini-scenes. We assess the efficiency of the adaptive classifier depending on the model complexity and the proportion of labeled/unlabeled data.

Paper Details

Date Published: 1 November 1999
PDF: 8 pages
Proc. SPIE 3812, Applications and Science of Neural Networks, Fuzzy Systems, and Evolutionary Computation II, (1 November 1999); doi: 10.1117/12.367699
Show Author Affiliations
Jose L. Alba Castro, Univ. of Vigo (Spain)
Susana M. Rey, Univ. of Vigo (Spain)
Laura Docio, Univ. of Vigo (Spain)


Published in SPIE Proceedings Vol. 3812:
Applications and Science of Neural Networks, Fuzzy Systems, and Evolutionary Computation II
Bruno Bosacchi; David B. Fogel; James C. Bezdek, Editor(s)

© SPIE. Terms of Use
Back to Top