CO-LOCATION OF PIXELS IN SATELLITE REMOTE-SENSING IMAGES WITH DEMONSTRATIONS USING SEA-ICE DATA

Quantitative integration of remote sensing data requires accurate co-l ocation of pixels between data from different sensors. A technique has been developed, based on a simple spherical Earth model, to map groun d resolution cells pertaining to pixel/grid points in a coarse-resolut ion sensor into an image acquired by a fine-resolution sensor. Pixels from the fine-resolution sensor, located within the coarse-resolution cells, can then be identified and their radiometric information can be compared to or combined with the coarse-resolution observations to im prove the utility of both data sets. The technique is presented along with examples to demonstrate potential applications such as: (1) corre lation between sensors radiometric parameters, (2) validation of produ cts from a coarse-resolution sensor using sub-pixel information from a coincident fine-resolution data set, and (3) contribution of ground t arget heterogeneity within a single pixel to observed radiometric para meters. Successful application of the technique, however, requires acc urate gee-location of both data sets at the pixel level.