A Novel Three Steps Method for Forest Parameters Extraction Using PolInSAR Images

摘要

The accuracy of ground phase and volume only coherence estimation are decisive factors to the efficient of the estimated forest height using PolInSAR data. This article approaches a novel three steps method to enhance the ground phase as well as forest height estimation from PolInSAR data. The ground phase is extracted from two fist steps by using conditional optimization approach. In the last steps, the forest parameters are retrieved from the optimal loop by using two optimum coherence channel without any prior information. The accuracy of the suggested method was applied with PolInSAR simulated data, which is generated from PolSARprosim 5.2 software. Experimental results indicate that the proposed method has improved the accuracy of the estimated forest height compared to Tayebe method by more than 1.1m