基于混合矩的极化SAR图像K分布模型参数估计新方法
doi: 10.11999/JEIT140551
Parameter Estimation for the K-distribution in PolSAR Imagery Based on Hybrid Moments
-
摘要: K分布模型在极化合成孔径雷达(PolSAR)图像建模领域中获得广泛应用。其模型参数估计的精度将直接影响到模型拟合的准确性。目前普遍采用的K分布参数估计方法是基于协方差矩阵Mellin变换的对数累积量的估计方法。但是该方法没有解析的表达式,数值计算运算时间较长,另外在形状参数1时估计偏差较大。为此该文提出一种基于|z|rlg|z|混合矩的参数估计新方法,该方法对不同形状参数值下的参数估计具有较好的适应性,并且在值较小时估计性能优于对数累积量方法。同时在r=1/d时该方法有解析的表达式,其运算时间优于对数累积量方法。最后用仿真数据和实测数据对新方法和已有参数估计方法的结果进行了比较,验证了基于混合矩估计方法的准确性与有效性。该方法为PolSAR图像统计模型参数的快速有效估计提供了新手段。Abstract: The K-distribution is usually used to model the Polarimetric Synthetic Aperture Radar (PolSAR) imagery. The parameter estimation method for K-distribution is very important,which affects the fitting degree of the model. However, the classical method of matrix log-cumulants relies upon a nontrivial inversion of a nonlinear equation, which introduces a computationally expensive stage into the estimation procedure. Moreover, the bias is large when the sharp parameter is smaller than 1. Therefore, a new method for estimating the sharp parameter of K-distribution based on|z|rlg|z| is proposed. This method is more adaptable to parameter estimation under different sharp parameters, and the performance is better than matrix log-cumulantes whenis small. In addition, the proposed estimator has an analytical expression at r=1/d, which allows rapid caculation. Finally, the estimation accuracy of this new estimator is compared with previous ones through simulation data and real data. The results show that the new estimator is effective and robust, which is of practical value in solving the accurate parameter estimation of K-distribution.
计量
- 文章访问数: 1640
- HTML全文浏览量: 118
- PDF下载量: 554
- 被引次数: 0