Statistical multisource classification of remote sensing data and geographic information by means of a method based on Bayesian classification theory is investigated. Extensions are made to the method to control the influence of the data sources involved in the classification process. Reliability measures are introduced to rank the quality of the data sources. The data sources are then weighted according to these rankings in the multisource classification. Four data sources are used in experiments: One is Landsat multispectral scanner (MSS) data; the other three contain topographic data (elevation, slope, and aspect data). The data classes in the Landsat MSS data are treated as Gaussian, but the data classes in the other sources are treated as non-Gaussian.
|Útgáfustaða||Útgefið - 1989|
|Viðburður||IGARSS'89 - Twelfth Canadian Symposium on Remote Sensing Part 2 (of 5) - Vancouver, BC, Can|
Tímalengd: 10 júl. 1989 → 14 júl. 1989
|Ráðstefna||IGARSS'89 - Twelfth Canadian Symposium on Remote Sensing Part 2 (of 5)|
|Borg/bær||Vancouver, BC, Can|
|Tímabil||10/07/89 → 14/07/89|