Textural and contextual land-cover classification using single and multiple classifier systems

Olivier Debeir, Isabelle Van Den Steen, Patrice Latinne, Philippe Van Ham, Éléonore Wolff

2002 · 51 citations · 10 references

Concepts

Abstract

The objective of this study was to improve the quality of the digital land cover and land use classification of high-resolution (10-30 m) remote sensing data. Three classification techniques were compared, which can be divided into two groups: single classifiers (a 5-nearest neighbor and the C4.5 decision tree classifier) and multiple classifier systems (BAGFS). Textural and contextual features (roads, hydrology, relief, etc.) were introduced during the classification process. Eleven land-cover categories, in a Belgian varied landscape, were analyzed and classified using the LANDSAT Thematic Mapper data. The accuracy assessment increased with the introduction of textural features and contextual data, between 0,60 and 0,82 for the Kappa coefficient. The best kappa value was achieved using numerous textural and contextual features with the multiple classifier system (BAGFS).

References

10