Translator Disclaimer
27 January 1997 Synthetic multispectral data sets for testing remote sensing algorithms and processing systems
Author Affiliations +
At the same time that we develop new sensors we also need to produce algorithms and processing systems to analyze the data in an operational mode shortly after launch. To develop and test the algorithms and processing systems we need test data. Yet new sensors are often designed to produce combinations of measurements that have never been made before. To resolve this dilemma for the moderate resolution imaging spectroradiometer (MODIS) to be launched in 1998 on the Earth Observing System (EOS) AM platform we are producing synthetic data to test the programing and operational aspects of the algorithms and processing systems. Both MODIS and the resulting synthetic data provide measurements at 36 wavelengths ranging from the visible well into the infrared day and night over the entire globe. The test data covers many of the physical conditions MODIS will observe with a full range of surface and atmospheric characteristics over land and sea with correct instrument and orbital characteristics. The data is sufficiently representational of the radiances MODIS would observe that the processing algorithms run to completion in a reasonable manner and use computing resources similar to those expected with real flight data. Although the simulation is not detailed enough to support theoretical investigations it has proven invaluable in implementing new concepts into operational code. So far we have provided hundreds of gigabytes of data covering many test cases. We describe requirements for synthetic data, tell how the data is produced, what characteristics it models, what limitations it has and what sorts of tests it supports. We show examples of the resulting data sets and describe our plans for future improvements. Sample synthetic MODIS data sets are available and we tell where and how to obtain them.
© (1997) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Albert J. Fleig, Edward J. Masuoka, and Kai Yang "Synthetic multispectral data sets for testing remote sensing algorithms and processing systems", Proc. SPIE 2957, Advanced and Next-Generation Satellites II, (27 January 1997);


Back to Top