Получение, новые источники, открытость
- Сообщения: 902
- Зарегистрирован: 27 янв 2009, 22:57
- Статьи: 1
- Проекты: 2
- Репутация: 215
http://www.int-arch-photogramm-remote-s ... 3-2014.pdf
One of the most challenging tasks in the automation is the masking of outlier areas covered with water, snow, or cloud. In the process we apply a statistical classification algorithm analyzing the correlation derived from the image matching, the brightness value of the ORI data, and the roughness of DSM data itself. For open water areas the existing global water-body-data in public domain such as SRTM Water Body Data (SWBD) (NASA/NGA, 2003) or Global Self-consistent, Hierarchical, High-resolution Shoreline Database (GSHHS) (Wessel et al., 1996) are utilized as initial masks. However, the initial masks may have some incorrect areas because their source dates were different from the ones of PRISM data. Moreover their ground resolution is larger than the one for PRISM. Therefore first the excessive masks in the initial water-body masks are deleted by analysing the matching correlations on the areas.