SYSTEMATIC MONITORING AND EVALUATION OF M7 SCANNER PERFORMANCE AND DATA QUALITY

摘要

This report presents the results of an investigation to provide the information re¬quired to maintain data quality of the Michigan M7 Multispectral scanner by systematic checks on specific system performance characteristics. Data processing techniques which use calibration data (called "standardization" runs) gathered routinely every mission have been developed to assess current data quality. Significant changes from past data quality are thus identified and attempts made to discover their causes. Pro¬cedures for systematic monitoring of scanner data quality are discussed. In the solar reflective region, calculations of NEAL (Noise Equivalent Change in Radiance) on a per¬mission basis are compared to theoretical tape-recorder limits to provide an estimate of overall scanner performance. M7 signal/noise characteristics (from μsee to days) are examined. In the thermal region, calculations of NEAT (Noise Equivalent Change in Temperature) are used to evaluate scanner performance after each mission. Tech¬niques are developed using the hot, cold, and ambient temperature reference plates to determine if they are working properly. Finally, overall improvements in the process¬ing of radiance calibration data are reported.