Filtering remote sensing data in the spatial and feature domains
The effects of applying pre-processing and post-processing to remote sensing data are studied both in the spatial image domain and the feature domain. A neural network is used for classification since it is not biased by a priori assumptions about the distributions of the spectral values of the classes. Spatial smoothing was applied both as pre- and post-processing steps. Pre-processing involved smoothing the image spectral values by means of anisotropic diffusion, whereas iterative majority filtering is applied as a post-processing step to improve spatial coherence by reclassifying pixels. The effects of all spatial and spectral filtering methods were validated by applying them to three different test cases.
Bibliographic Reference: Paper presented: EUROPTO, Roma (IT), September 26-30, 1994
Availability: Available from (1) as Paper EN 38575 ORA
Record Number: 199510432 / Last updated on: 1995-04-11
Original language: en
Available languages: en