Classification of MOMS-02 image data using spectral and shape features

Segl, Karl

Item

ITEM ACTIONSEXPORT

Add to Basket

Local TagsRelease HistoryDetailsSummary

Released

Conference Paper

Classification of MOMS-02 image data using spectral and shape features

Authors

/persons/resource/segl

Segl, Karl
1.4 Remote Sensing, 1.0 Geodesy and Remote Sensing, Departments, GFZ Publication Database, Deutsches GeoForschungsZentrum;

External Ressource

No external resources are shared

Fulltext (public)

There are no public fulltexts stored in GFZpublic

Supplementary Material (public)

There is no public supplementary material available

Citation

Segl, K. (1995): Classification of MOMS-02 image data using spectral and shape features, EARSeL Advances in Remote Sensing - MOMS-02 (Köln/Bonn).

https://gfzpublic.gfz-potsdam.de/pubman/item/item_227222

Abstract

The high spatial resolution of the multispectral channels (13.5m x 13.5m), the nadir panchromatic channel (4.5m x 4.5m) and their combination offers significant improvements for geoscientific applications. Common multispectral classification schemes however hardly benefit from the improved spatial resolution. Many objects can be better separated by their individual shape characteristics and not by the spectral information of a single pixel. Therefore a modular classification scheme was developed, which uses spectral as well as shape features. This technic was applied to a landcover classification on two different test sites. First applications show significantly improved results for certain object classes, which have a typical shape, like e.g. trees, houses, roads, and rivers. The common use of this technique is still restricted by the shape extraction.