基于改进支持向量机的海面目标检测方法

为解决舰载、岸基海防预警雷达在检测掠海飞行的反舰导弹时,受到海杂波严重干扰的问题,基于工程实际应用的需要,本文首先采用易于实现的盒维数提取方法,进而提出了一种基于改进核函数的支持向量机(Support Vector Machine,SVM)分类器目标检测方法。最后,基于S波段雷达实测海杂波数据进行了仿真实验,实验结果验证了本文方法具有较强的检测能力和抗杂波性能。

The Method of Target Detection Based on Ameliorated SVM in the Presence of Sea Clutter

To solve this question that the ship-borne radar and shore-based early warning radar are interfered seriously by the sea clutter when they track the anti-ship missile of flight over the ocean,firstly this paper adopts the feasible method of extracting box dimen-sion based on meeting the need of fact engineering.Furthermore the method of target detection with SVM classifier is presented of the a-meliorated RBF kernel function.Finally,based on the observed sea clutter of S-band radar,the computer simulation is processed and the results prove the effective detection performance and noise tolerance.