Cloud detection is one of the main problems in operational applications of satellite data. The IMAA research activity in this field is mainly devoted to develop new algorithms for cloud detection and classification from different instruments (HIRS/3, IASI, AIRS, SEVIRI). Physical, statistical and temporal approaches have been explored in order to improve the detection of cirrus and of very low clouds. Moreover the water vapour MSG-SEVIRI channels as well as the texture properties of the MSG-SEVIRI images are considered in physical and statistical algorithms in order to screen the no-raining from raining clouds, in the attempt to distinguish no-raining very thick cirrus from raining convective clouds. The following figure shows an example of the cloud Mask Coupling of Statistical and Physical methods (MACSP) application.