Combining optical satellite data and airborne laser scanner data for vegetation classification

Abstract

The aim of this study was to investigate to which degree the accuracy of vegetation classification could be improved by combining optical satellite data and airborne laser scanner (ALS) data, compared with using satellite data only. A Satellite Pour l'Observation de la Terre (SPOT) 5 scene and Leica ALS 50-II data from 2009, covering a test area in the mid-Sweden (latitude 60° 43′ N, longitude 16° 43′ E), were used in maximum likelihood and decision tree classifications. Training and validation data were obtained from the interpretation of digital aerial photo stereo models. Combination of SPOT and ALS data gave classification accuracies up to 72%, compared with 56% using only SPOT data. This indicates that integrating features from large area laser scanning may lead to significant improvements in satellite data-based vegetation classifications.

Acknowledgements

This work was financially supported by the Swedish National Space Board and the Swedish Environmental Protection Agency. We thank the Swedish NLS (Lantmäteriet) for the ALS data and three anonymous reviewers for their constructive comments.