GEOMETRIC CORRECTION OF GEOSTATIONARY OCEAN COLOR IMAGER (GOCI) DATA

Abstract

IMPS is consisted DM (Decomposition Module, 1 server), INRSM (ImageNavigation and Registration Software Module, 8 servers), IMPS DB (1 server), PMM (Product Management Module with GUI, 2 servers). GOCI raw images consist of 16 “slots” that are positioned in 4 lines and 4 columns. Each slot contains 8 channels in the visible and near infer-redbands. IMPS produces Level 0, Level 1A, and Level 1B data from GOCI raw data. In the process, INRSM(Image Navigation and Registration Software Module) has a function to perform geometric correction (strictly speaking, image registration) on the input image files of Level 1A, and provide output image files containing level 1B information. In this study, we will review the geometric validation of GOCI L1B data produced at the IMPS level and estimate the INRSM performances.Finally we will suggest a method for actual approach of GOCI level 1B.Figure 1 shows a data flow from GDAS(GOCI Data Acquisition System) to GDPS(GOCI DataProcessing System). INRSM produces the parameters such as INRSM configuration, landmark, tiepoint, image quality and pointing accuracy.Fig. 1 Overview of GOCI data processing INR operational and tuning tests are in process and will be continued until October. Preliminary results for the geometric performance are presented here with configuration parameters adjusted during tuning test. The parameters include cloud detection, landmark accuracy estimation, linear error correction, scheduler and IMC command generation parameters.