Volume Coherence Estimation for Random Forest Height Retrieval Based on Polinsar Data

The volume coherence is an important parameter in the two-layer model for random forest height retrieval. Though it is "invisible" from the observation, it can be estimated from the data. Considering the argument range of the coherences for all the possible polarizations, six polarizations can be obtained for each scattering cell. The volume coherence can be estimated more reasonably and the ground phase center can be determined more accurately. It benefits to improve the height retrieval accuracy by this model. The proposed method is validated by using L-band single-baseline polarimetric interferometric SAR data collected with the DLR Experimental SAR system.