Patent · US Active

Method for reconstructing geostationary ocean color satellite data based on data interpolating empirical orthogonal functions

US11790580B2 · kind B2 · utility

0Cited by
2References
10Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJul 23, 2019
Grant dateOct 17, 2023
Priority date
Expiry dateAug 19, 2039

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06T2207/30181
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

The present disclosure discloses a method for reconstructing geostationary ocean color satellite data based on Data INterpolating Empirical Orthogonal Functions. The method includes steps of: 1) for an original ocean color remote sensing area, using concentration detection and edge detection to eliminate abnormal image elements; 2) in the Data INterpolating Empirical Orthogonal Functions, using Laplace smoothing filter to smoothly filter a time covariance matrix; 3) for temporal modes value obtained by the Data INterpolating Empirical Orthogonal Functions, first decomposing it into sub-temporal mode, then adopting empirical modal decomposition, and finally performing interpolation, to obtain an ocean color remote sensing reanalysis data set that maintains its original accuracy and full coverage of time and space. Advantage of the method is that it fully considers high time resolution characteristics of the geostationary ocean color satellite data and considers conditions that there is a continuous high missing rate and complete missing in data missing, and it is suitable for the reconstruction of the geostationary ocean color satellite data, with good accuracy and high interpretati…

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.