Links
- The Publications Warehouse does not have links to digital versions of this publication at this time
- Download citation as: RIS | Dublin Core
Abstract
Muitispectral remotely sensed images have been treated as arbitrary multivariate spectral data for purposes of clustering and classifying. However, the spatial properties of image data can also be exploited. AMOEBA is a clustering and classification method that is based on a spatially derived model for image data. In an evaluation test, Landsat data were classified with both AMOEBA and a widely used spectral classifier. The test showed that irrigated crop types can be classified as accurately with the AMOEBA method as with the generally used spectral method ISOCLS; the AMOEBA method, however, requires less computer time.
| Publication type | Article |
|---|---|
| Publication Subtype | Journal Article |
| Title | Evaluation of AMOEBA: a spectral-spatial classification method |
| Series title | Journal of Applied Photographic Engineering |
| Volume | 8 |
| Issue | 3 |
| Year Published | 1982 |
| Language | English |
| Publisher | Society for Imaging Science and Technology |
| Contributing office(s) | Earth Resources Observation and Science (EROS) Center |
| Description | 4 p. |
| First page | 159 |
| Last page | 162 |
| Online Only (Y/N) | N |
| Additional Online Files (Y/N) | N |
| Google Analytic Metrics | Metrics page |