An improved volume coherence optimization method for forest height estimation using PolInSAR images

Forest height is one of the most important forest vertical structure parameter for many forest management and monitoring activities. One of the most widely used methods for forest height estimation using the PolInSAR image, which is the three-stage inversion process. In this method, the complex coherence value of the HV channel is used for forest height estimation. Therefore, the estimated forest height is usually lower than the actual forest height. Moreover, when determining the ground-phase coefficient, this method used a lot of channel coherence coefficients, which increased the amount of calculation complexity. The proposed method in this paper, we determined the optimum volume coherence coefficient of the system by comparing the polarization channels and using the optimal complex unitary vector. The experimental results show that the accuracy of forest height is significantly improved by the proposed method. © 2019 IEEE.