| Title | Pre-Processing and SemiVariogram Analysis of RADARSAT Fine Mode Images for Forest Application: Tapajós National Forest, Brazilian Amazon |
| |
| Author | Amaral, S; Shimabukuro, Y E; Ahern, F; Valiquet, J |
| Source | International Symposium, Geomatics in the Era of RADARSAT (GER'97), Ottawa, Canada, May 25-30; 1997 p. 10 |
| Year | 1997 |
| Alt Series | Earth Sciences Sector, Contribution Series 20042229 |
| Document | book |
| Lang. | English |
| Media | paper |
| Abstract | This paper describes part of the activities developed in the PRORADAR (INPE/CCRS) evaluating the potentialities of RADARSAT images for forest application. The study site comprises the Tapajós National
Forest, Pará State, Brazil. Two RADARSAT fine mode images were used: SGF- F2, acquired on May 20, 1996 and SGX-F5 acquired on May 05, 1996. In these images the relative image calibration and ortho-correction using geometrical modeling and the digital
elevation model (DEM) were applied. The speckle filters considering different window sizes. Different (mediana, Lee and gamma) were tested over the images, forest types and land use classes were analyzed in the RADARSAT images: low plateau forest,
two types of high plateau forest, disturbed forest, clearing and water. Since no significant difference between power average values for each class was observed, a semivariogram analyses were performed to evaluate its spatial variability. Although
semivariograms varied according to the resampling procedure, they were considered a useful tool for discriminating land cover classes on RADARSAT images. |
| GEOSCAN ID | 219031 |
|
|