SAR目标检测中的一种稳健变化检测算法

在合成孔径雷达(SAR)目标检测中，目前采用的变化检测算法主要包括构造似然比法和最小二乘图像相减法，这两种算法在实际变化检测中分别存在如下问题:构造似然比法需要目标、杂波及噪声充分的先验知识，而最小二乘图像相减法则存在稳健性差的问题，鉴于上述问题，该文提出一种稳健可行的变化检测算法，该算法在最小二乘图像相减法的基础上引入了误差影响因子，通过迭代的方法求取待定参数，大大提高了变化检测的稳健性，文章最后基于不同算法进行了变化检测试验，试验结果进一步证明了该文提出算法的稳健性。

A Robust Change Detection Algorithm in SAR Target Detection

The main change detection algorithms include forming likelihood ratio algorithm and Least Mean Square (LMS) image subtraction algorithm in Synthetic Aperture Radar (SAR) target detection. There are respective problems for two change detection algorithms: Forming likelihood ratio algorithm needs enough prior knowledge of target, clutter and noise, and LMS image subtraction algorithm is not robust enough. Considering above problems, this paper presents a robust change detection algorithm. The proposed algorithm introduces the influence factor of error on the basis of LMS image subtraction algorithm and calculates the parameters through iteration technique, which improves the robustness of change detection. Finally the results of change detection experiments based on different change detection algorithms are given to prove the robustness of the proposed algorithm in this paper.