On radar polarization target decomposition theorems with application to target classification, by using neural network method

In the first part the authors recall the basic principles of `radar holography´ and high resolution electromagnetic radar imaging construction, and present a real measurement of canonical radar targets. Then they introduce in the second part, the concept of `radar polarization target decomposition theorems´ to separate the Mueller matrix in two parts which are related to a single averaged target and to a residue component. In the last part, they introduce a new polarimetric method of classification based on a neural network modelization, and show how it is possible to identify a scatterer from the knowledge of the four high resolution electromagnetic radar images