Satellite sea-fog detection system using the gaussian mixture mode and fog detection method with unSupervised Learning mode based on gaussian mixture model

摘要

The present invention relates to a satellite malfunction detection system using a Gaussian mixture model and a UL fog detection method based on a Gaussian mixture model. According to a first aspect of the present invention, there is provided a clustering module for generating a Gaussian mixture model by performing daily clustering using an EM algorithm; A fog mode determination module (12) that performs fog mode determination among the clustered components; A cloud masking module 14 for performing a COMS cloud masking test between the fog pixels using the cloud masking pixels of the COMS; And a satellite satellites detection system using the Gaussian mixture model. According to a second aspect of the present invention, there is provided a method of generating a Gaussian mixture model by performing daily clustering using an EM algorithm, A second step of performing a fog mode decision among the clustered components; And a third step of performing a COMS cloud masking test using the cloud masking pixels of the COMS between the fog pixels, The present invention also provides a method for detecting a fog based on a Gaussian mixture model.