图像幅度和值最小化自聚焦算法  

Minimizing Sum of Image Amplitudes Autofocus Algorithm

作　　者：刘碧丹[1,2] 韩松[1] 王岩飞[1] 

机构地区：[1]中国科学院电子学研究所,北京100190 [2]中国科学院研究生院,北京100039

出　　处：《电子与信息学报》2009年第4期768-771,共4页Journal of Electronics & Information Technology

摘　　要：为了获得高分辨率合成孔径雷达(SAR)图像,需要精确的方位向调频斜率参数,该文提出了图像幅度和值最小化的自聚焦算法。论文分析了方位向信号频域的压缩处理,当频域复乘函数的幅值为常数时,图像对比度与图像幅度和值之间存在一定的关系,又根据图像聚焦程度和图像对比度之间的关系,推导出图像聚焦程度和图像幅度和值之间的联系,根据此联系可以快速准确地确定方位向真实调频斜率,论文给出了该算法的实现流程。与传统的子孔径相关算法相比,该算法由于不存在部分孔径效应,所以具有计算量小、收敛好、估计精度高的特点;与图像对比度最优估计法相比,该文算法收敛速度快,运算量少,易于硬件实现,有助于提高实时成像处理器的性能。对RADARSAT-1数据自聚焦,实验结果验证了该算法的有效性。High resolution Synthetic Aperture Radar （SAR） imaging requires that the azimuth frequency modulation rate is accurately estimated. A new autofocus algorithm is presented which minimizing sum of amplitudes in this paper. This paper analyses the relationship between the image contrast function and the sum of amplitudes, when the matched function＇s amplitude in azimuth direction is invariable. Based on the relationship between the image contrast function and the focus degree of the image, the minimizing sum of amplitudes autofocus algorithm is presented. The performance of this algorithm is verified by the computer simulation and RADARSAT-1 raw data imaging. Experimental results show that the algorithm has better convergence speed and estimation accuracy than traditional Map Drift （MD） auto focus algorithm. It has better convergence speed and less computation cost than the Contrast Optimization Autofocus Algorithm （COAA）. It can be implemented by hardware easily and help to improve the performance of real time imaging processor.

关 键 词：合成孔径雷达 多普勒调频斜率 自聚焦 幅度和值最小 

分 类 号：TN957.52[电子电信—信号与信息处理]