合成孔径雷达图像中乘性噪音的存在使合成孔径雷达图像分割变得非常困难.针对这一难题,本文以提高分割准确度,保护图像的几何结构边缘和提高算法的鲁棒性为目的,提出了一种适用于处理合成孔径雷达图像分割的新模型.新模型结合合成孔径雷达图像的区域和边缘信息,首先通过引入非凸的正则化项,定义了能量泛函;然后极小化能量泛函,建立了水平集函数演化的偏微分方程;最后对水平集演化方程的数值求解,实现了对合成孔径雷达图像感兴趣区域的分割.分别采用仿真图像和实测合成孔径雷达图像对新模型进行验证,结果表明,新模型对合成孔径雷达图像具有很强的边缘定位能力,能使目标区域分割更完整.

With multiplicative noise present, SAR image segmentation becomes difficult to implement.In this paper, a new energy functional model in introduced which is based on nonconvex regularization. This new model combines the regions as well as boundary properties, which can improve the accuracy of segmentation, and protect the geometric edges of the SAR images as far as possible. Segmentation of regions of interest is performed by numerical solution of the partial differential equations that are derived through minimizing the energy formulation. Experiments on both synthetic and real SAR images demonstrate the effectiveness of the proposed algorithm.