舰船目标快速检测是星载合成孔径雷达的重要应用之一，也是SAR星上处理首先发展的应用方向。目前SAR舰船快速检测普遍使用的图像恒虚警（CFAR）检测方法中，低虚警一直是研究的重点，多通道SAR中舰船目标存在多通道虚假目标，进一步增加了虚警剔除的难度。为此，本文提出了一种基于置信度计算和SVM判别的低虚警CFAR（Constant False-Alarm Rate）舰船快速检测方法。首先，在传统CFAR算法中引入分块计算，同时针对目前舰船的长宽、面积等几何参数统计，设计舰船目标置信度计算方法，并对几何参数相近的虚假目标利用SVM分类器进行舰船目标与虚假目标的快速区分，并设计了CFAR检测总体流程。最后，采用GF-3卫星双通道模式实测数据对已有方法和新方法进行了算法运行时间及目标检测性能对比。实验结果表明，在基于TX2构建的星上处理平台上，新方法运行时间比传统CFAR方法缩短30倍，具有良好的实时性，同时对舰船目标检测检测率达到97.8%的同时，虚警率控制在5.2%，具有良好的虚假目标剔除效果。

The rapid detection of ship targets is one of the important applications of spaceborne synthetic aperture radar (SAR) and the first application direction of SAR on-board processing. At present, the constant false alarm (CFAR) detection method is the most commonly used in the rapid detection of SAR ships, and low false alarm has always been the focus of research. In multi-channel SAR, ship targets have multi-channel false targets, further increasing the difficulty of false alarm elimination.This paper proposes a fast detection method for low false alarm CFAR ships according to confidence calculation and SVM discrimination. First, the traditional CFAR algorithm introduces block calculation while designing the calculation method of ship target confidence and at the same time, according to the statistics of the geometric parameters of the current ship’s length, width, area, and so on, and the SVM classifier is used for the false targets with similar geometric parameters. Quickly distinguishing between ship and false targets and designing the overall process of CFAR detection. Finally, the GF-3 satellite dual-channel mode measured data is used to compare the algorithm running time and the target detection performance between the existing and new methods.Experimental results show that on the on-board processing platform based on TX2, the new method runs 30 times shorter than the traditional CFAR method and has good real-time performance. Meanwhile, the detection rate of ship targets reaches 97.8%, and the false alarm rate is controlled at 5.2%, showing a good ability of eliminating false targets.In view of the low real-time performance of traditional CFAR on spaceborne platforms and the existence of multiple false targets in the detection results, a method for quickly calculating the CFAR panoramic threshold is proposed. Experiments show that this algorithm has good real-time and detection performance for the ship detection problem of multi-channel spaceborne SAR data and can quickly and effectively detect ship targets and remove most false targets.