Spectral and Spatial-Based Classification for Broad-Scale Land Cover Mapping Based on Logistic Regression
Improvement of satellite sensor characteristics motivates the development of new techniques for satellite image classification. Spatial information seems to be critical in classification processes, especially for heterogeneous and complex landscapes such as those observed in the Mediterranean basin....
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2008-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/8/12/8067/ |
| _version_ | 1811299501458063360 |
|---|---|
| author | Georgios Mallinis Nikos Koutsias |
| author_facet | Georgios Mallinis Nikos Koutsias |
| author_sort | Georgios Mallinis |
| collection | DOAJ |
| description | Improvement of satellite sensor characteristics motivates the development of new techniques for satellite image classification. Spatial information seems to be critical in classification processes, especially for heterogeneous and complex landscapes such as those observed in the Mediterranean basin. In our study, a spectral classification method of a LANDSAT-5 TM imagery that uses several binomial logistic regression models was developed, evaluated and compared to the familiar parametric maximum likelihood algorithm. The classification approach based on logistic regression modelling was extended to a contextual one by using autocovariates to consider spatial dependencies of every pixel with its neighbours. Finally, the maximum likelihood algorithm was upgraded to contextual by considering typicality, a measure which indicates the strength of class membership. The use of logistic regression for broad-scale land cover classification presented higher overall accuracy (75.61%), although not statistically significant, than the maximum likelihood algorithm (64.23%), even when the latter was refined following a spatial approach based on Mahalanobis distance (66.67%). However, the consideration of the spatial autocovariate in the logistic models significantly improved the fit of the models and increased the overall accuracy from 75.61% to 80.49%. |
| first_indexed | 2024-04-13T06:36:47Z |
| format | Article |
| id | doaj.art-a9eafebd7b03405e8ff461e494c10f15 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-13T06:36:47Z |
| publishDate | 2008-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-a9eafebd7b03405e8ff461e494c10f152022-12-22T02:57:53ZengMDPI AGSensors1424-82202008-12-018128067808510.3390/s8128067Spectral and Spatial-Based Classification for Broad-Scale Land Cover Mapping Based on Logistic RegressionGeorgios MallinisNikos KoutsiasImprovement of satellite sensor characteristics motivates the development of new techniques for satellite image classification. Spatial information seems to be critical in classification processes, especially for heterogeneous and complex landscapes such as those observed in the Mediterranean basin. In our study, a spectral classification method of a LANDSAT-5 TM imagery that uses several binomial logistic regression models was developed, evaluated and compared to the familiar parametric maximum likelihood algorithm. The classification approach based on logistic regression modelling was extended to a contextual one by using autocovariates to consider spatial dependencies of every pixel with its neighbours. Finally, the maximum likelihood algorithm was upgraded to contextual by considering typicality, a measure which indicates the strength of class membership. The use of logistic regression for broad-scale land cover classification presented higher overall accuracy (75.61%), although not statistically significant, than the maximum likelihood algorithm (64.23%), even when the latter was refined following a spatial approach based on Mahalanobis distance (66.67%). However, the consideration of the spatial autocovariate in the logistic models significantly improved the fit of the models and increased the overall accuracy from 75.61% to 80.49%.http://www.mdpi.com/1424-8220/8/12/8067/Land cover mappinglogistic regressionautocovariatetexture |
| spellingShingle | Georgios Mallinis Nikos Koutsias Spectral and Spatial-Based Classification for Broad-Scale Land Cover Mapping Based on Logistic Regression Sensors Land cover mapping logistic regression autocovariate texture |
| title | Spectral and Spatial-Based Classification for Broad-Scale Land Cover Mapping Based on Logistic Regression |
| title_full | Spectral and Spatial-Based Classification for Broad-Scale Land Cover Mapping Based on Logistic Regression |
| title_fullStr | Spectral and Spatial-Based Classification for Broad-Scale Land Cover Mapping Based on Logistic Regression |
| title_full_unstemmed | Spectral and Spatial-Based Classification for Broad-Scale Land Cover Mapping Based on Logistic Regression |
| title_short | Spectral and Spatial-Based Classification for Broad-Scale Land Cover Mapping Based on Logistic Regression |
| title_sort | spectral and spatial based classification for broad scale land cover mapping based on logistic regression |
| topic | Land cover mapping logistic regression autocovariate texture |
| url | http://www.mdpi.com/1424-8220/8/12/8067/ |
| work_keys_str_mv | AT georgiosmallinis spectralandspatialbasedclassificationforbroadscalelandcovermappingbasedonlogisticregression AT nikoskoutsias spectralandspatialbasedclassificationforbroadscalelandcovermappingbasedonlogisticregression |