Finding Optimal Neural Networks for Land Use Classification

Horst Bischof; Ales Leonardis

doi:10.1109/36.655348

Finding Optimal Neural Networks for Land Use Classification

Horst Bischof, Ales Leonardis

Institute of Computer Graphics and Vision (7100)

Research output: Contribution to journal › Article › peer-review

Abstract

The authors present a fully automatic and computationally efficient algorithm based on the minimum description length principle (MDL) for optimizing multilayer perceptron (MLP) classifiers. They demonstrate their method on the problem of multispectral Landsat image classification. They compare their results with a hand-designed MLP and a Gaussian maximum likelihood classifier, in which their method produces better classification accuracy with a smaller number of hidden units.

Original language	English
Pages (from-to)	337-341
Journal	IEEE Transactions on Geoscience and Remote Sensing
Volume	36
Issue number	1
DOIs	https://doi.org/10.1109/36.655348
Publication status	Published - 1998

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/36.655348

Cite this

@article{8b5ae95ce94146f1ad63aee86b8e0bf4,

title = "Finding Optimal Neural Networks for Land Use Classification",

abstract = "The authors present a fully automatic and computationally efficient algorithm based on the minimum description length principle (MDL) for optimizing multilayer perceptron (MLP) classifiers. They demonstrate their method on the problem of multispectral Landsat image classification. They compare their results with a hand-designed MLP and a Gaussian maximum likelihood classifier, in which their method produces better classification accuracy with a smaller number of hidden units.",

author = "Horst Bischof and Ales Leonardis",

year = "1998",

doi = "10.1109/36.655348",

language = "English",

volume = "36",

pages = "337--341",

journal = "IEEE Transactions on Geoscience and Remote Sensing",

issn = "1558-0644",

publisher = "Institute of Electrical and Electronics Engineers",

number = "1",

}

TY - JOUR

T1 - Finding Optimal Neural Networks for Land Use Classification

AU - Bischof, Horst

AU - Leonardis, Ales

PY - 1998

Y1 - 1998

N2 - The authors present a fully automatic and computationally efficient algorithm based on the minimum description length principle (MDL) for optimizing multilayer perceptron (MLP) classifiers. They demonstrate their method on the problem of multispectral Landsat image classification. They compare their results with a hand-designed MLP and a Gaussian maximum likelihood classifier, in which their method produces better classification accuracy with a smaller number of hidden units.

AB - The authors present a fully automatic and computationally efficient algorithm based on the minimum description length principle (MDL) for optimizing multilayer perceptron (MLP) classifiers. They demonstrate their method on the problem of multispectral Landsat image classification. They compare their results with a hand-designed MLP and a Gaussian maximum likelihood classifier, in which their method produces better classification accuracy with a smaller number of hidden units.

U2 - 10.1109/36.655348

DO - 10.1109/36.655348

M3 - Article

SN - 1558-0644

VL - 36

SP - 337

EP - 341

JO - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

IS - 1

ER -

Finding Optimal Neural Networks for Land Use Classification

Abstract

UN SDGs

Access to Document

Fingerprint

Cite this