中文摘要：

提出了一种基于Parzen窗函数的SAR图像人造目标检测算法.该方法首先以基于数据驱动的Parzen窗核函数逼近实际SAR图像的直方图,完成SAR图像的精确建模;在此基础上,理论推导了全局CFAR检测算法的阈值,设计了阈值求解的数值算法.典型目标的实际图像的实验结果证明,该方法是一种速度较快、精度较高的人造目标检测算法.

英文摘要：

An algorithm based on Parzen-window function to detect man-made target detection in SAR （synthetic aperture radar） images was proposed. The data-driving kernel functions of Parzen-window were used to approximate the histogram of real SAR image, in order to complete the accurate modeling of SAR images. The threshold of global CFAR （constant false alarm rate） was given theoretically, and the numerical solution of the threshold was also devised. The experimental results from the real data of typical target demonstrate that the presented algorithm is rapidly and accurately.