Forest height extraction with polarimetric SAR interferometry（POLInSAR） is a hot research field of imaging SAR remote sensing.Several available forest height inversion methods using POLInSAR data were validated and compared with repeat pass E-SAR datasets and the corresponding ground measured forest stand height through the analysis of the Random Volume over Ground（RVoG） scattering model.After analyzing the experiment results in the view of physical mechanisms

we developed an integrated inversion method combining interferometric coherence optimization and compensation of non-volumetric scattering decorrelation.Validation result shows that the general performance of the developed forest height inversion method is superior to the others.