| タイトル | The role of spatial, spectral and radiometric resolution on information content |
| 著者（英） | Acevedo, W.; Buis, J. S.; Alexander, D. A.; Wrigley, R. C. |
| 著者所属（英） | NASA Ames Research Center |
| 発行日 | 1983-01-01 |
| 言語 | eng |
| 内容記述 | The results of a factorial experiment to evaluate the effects of spatial, spectral, and radiometric resolution on training-data spectral separability and classification accuracy are reported. Aircraft scanner data from five flightlines at 19.8 km over California including croplands, rangeland, forest, water, and urban areas were systematically degraded over a range approximately from Landsat MSS to Thematic Mapper specifications. Reference data were collected on the ground and from aerial photography. The degradations, training-site delineation, data-analysis procedures, and accuracy-assessment techniques are described; the results are presented in tables and graphs and discussed. It is found that while accuracy was increased by higher spectral resolution in 70 percent of the cases and uniformly by increased radiometric resolution, it was decreased by higher spatial resolution. This phenomenon is attributed to classification methods. |
| NASA分類 | EARTH RESOURCES AND REMOTE SENSING |
| レポートNO | 84A13043 |
| 権利 | Copyright |
| URI | https://repository.exst.jaxa.jp/dspace/handle/a-is/400743 |
|
