Semi-automated training approaches for spectral class definition

P. V. Bolstad; T. M. Lillesand

doi:10.1080/01431169208904108

Semi-automated training approaches for spectral class definition

P. V. Bolstad
, T. M. Lillesand

Forest Resources

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

16 Scopus citations

Abstract

A semi-automated approach to spectral training for a maximum likelihood classification is shown to maintain or improve classification accuracy while reducing analyst input five-fold. The semi-automated approach is based on spectral sampling via region growing and training set refinement via transformed-divergence based mergers and deletions. Classification accuracies at the Anderson level II/1II in northern Wisconsin were higher for the semi-automated approach in five of six combinations of imagery, analyst, and study area, at significantly.

Original language	English (US)
Pages (from-to)	3157-3166
Number of pages	10
Journal	International Journal of Remote Sensing
Volume	13
Issue number	16
DOIs	https://doi.org/10.1080/01431169208904108
State	Published - Nov 1992

Bibliographical note

Funding Information:
Acknowledgments This work was supported by the University of Wisconsin Institute of Environmental Studies, the College of Agriculture and Life Sciences, and by Xi Sigma Pi. We also wish to thank M. Mohamed for his contributions to this research.

UN SDGs

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

SDG 7 Affordable and Clean Energy

Publisher link

10.1080/01431169208904108

Find It

Find It at U of M Libraries

Cite this

@article{e6f9fd5c0a1c4e6ca837dd4a00a45235,

title = "Semi-automated training approaches for spectral class definition",

abstract = "A semi-automated approach to spectral training for a maximum likelihood classification is shown to maintain or improve classification accuracy while reducing analyst input five-fold. The semi-automated approach is based on spectral sampling via region growing and training set refinement via transformed-divergence based mergers and deletions. Classification accuracies at the Anderson level II/1II in northern Wisconsin were higher for the semi-automated approach in five of six combinations of imagery, analyst, and study area, at significantly.",

author = "Bolstad, \{P. V.\} and Lillesand, \{T. M.\}",

note = "Funding Information: Acknowledgments This work was supported by the University of Wisconsin Institute of Environmental Studies, the College of Agriculture and Life Sciences, and by Xi Sigma Pi. We also wish to thank M. Mohamed for his contributions to this research.",

year = "1992",

month = nov,

doi = "10.1080/01431169208904108",

language = "English (US)",

volume = "13",

pages = "3157--3166",

journal = "International Journal of Remote Sensing",

issn = "0143-1161",

publisher = "Taylor and Francis Ltd.",

number = "16",

}

TY - JOUR

T1 - Semi-automated training approaches for spectral class definition

AU - Bolstad, P. V.

AU - Lillesand, T. M.

N1 - Funding Information: Acknowledgments This work was supported by the University of Wisconsin Institute of Environmental Studies, the College of Agriculture and Life Sciences, and by Xi Sigma Pi. We also wish to thank M. Mohamed for his contributions to this research.

PY - 1992/11

Y1 - 1992/11

N2 - A semi-automated approach to spectral training for a maximum likelihood classification is shown to maintain or improve classification accuracy while reducing analyst input five-fold. The semi-automated approach is based on spectral sampling via region growing and training set refinement via transformed-divergence based mergers and deletions. Classification accuracies at the Anderson level II/1II in northern Wisconsin were higher for the semi-automated approach in five of six combinations of imagery, analyst, and study area, at significantly.

AB - A semi-automated approach to spectral training for a maximum likelihood classification is shown to maintain or improve classification accuracy while reducing analyst input five-fold. The semi-automated approach is based on spectral sampling via region growing and training set refinement via transformed-divergence based mergers and deletions. Classification accuracies at the Anderson level II/1II in northern Wisconsin were higher for the semi-automated approach in five of six combinations of imagery, analyst, and study area, at significantly.

UR - https://www.scopus.com/pages/publications/0027009290

UR - https://www.scopus.com/pages/publications/0027009290#tab=citedBy

U2 - 10.1080/01431169208904108

DO - 10.1080/01431169208904108

M3 - Article

AN - SCOPUS:0027009290

SN - 0143-1161

VL - 13

SP - 3157

EP - 3166

JO - International Journal of Remote Sensing

JF - International Journal of Remote Sensing

IS - 16

ER -

Semi-automated training approaches for spectral class definition

Abstract

Bibliographical note

UN SDGs

Publisher link

Find It

Other files and links

Fingerprint

Cite this