一种基于局部梯度比率特征度量SAR图像相似性的新方法

A New Algorithm for SAR Imagery Similarity Measure Based on Local Gradient Ratio Pattern

本文针对SAR图像相干斑噪声和图像局部梯度特性,提出一种基于局部梯度比率特征的相似度准则.首先提取SAR图像各个像素的梯度比率特征,进而构建局部梯度比率特征直方图LGRPH,并进一步分析多尺度LGRPH,最后通过计算多尺度LGRPH的K-L距离定义相似度.基于SAR仿真和实测图像的实验结果表明,本文提出的相似度准则对SAR图像上相干斑噪声和局部梯度变化不敏感,能有效应用于SAR图像的目标识别.

In this paper, a new similarity measure for SAR imagery based on local gradient ratio pattern is proposed.Firstly, we extract the gradient ratio pattern for each pixel. Secondly, the local gradient ratio pattern histogram（LGRPH ）is computed. Final- ly, the similarity is obtained by utilizing K-L discrepancy to measure the distance of LGRPH. Experimental results based on simulat- ed and real SAR images demonstrate that the proposed similarity measure is robust to the speckle noise and local gradient variation in SAR images.