Burned area mapping using logistic regression modeling of a single post-fire Landsat-5 Thematic Mapper image

Logistic regression modeling was applied, as an alternative classification procedure, to a single post-fire Landsat-5 Thematic Mapper image for burned land mapping. The nature of the classification problem in this case allowed the structure and application of logistic regression models, since the de...

Full description

Saved in:
Bibliographic Details
Published in:International journal of remote sensing 2000-01, Vol.21 (4), p.673-687
Main Authors: Koutsias, N., Karteris, M.
Format: Article
Language:English
Subjects:
Applied geophysics
Areal geology. Maps
Earth sciences
Earth, ocean, space
Exact sciences and technology
Geologic maps, cartography
Internal geophysics
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Logistic regression modeling was applied, as an alternative classification procedure, to a single post-fire Landsat-5 Thematic Mapper image for burned land mapping. The nature of the classification problem in this case allowed the structure and application of logistic regression models, since the dependent variable could be expressed in a dichotomous way. The two logistic regression models consisted of the TM 4, TM 7, TM 1 and TM 4, TM 7, TM 2 presented an overall accuracy of 97.37% and 97.30%, respectively and proved to be the most well performing three-channel color composites. The discriminator ability in respect to burned area mapping of each one of the six spectral channels of Thematic Mapper, which was achieved by applying six logistic regression models, agreed with the results taken from the separability indices Jeffries-Matusita and Transformed Divergence.
ISSN:0143-1161
1366-5901
DOI:10.1080/014311600210506