基于聚类的对海雷达目标检测算法

A Sea Radar Target Detection Algorithm Based on Clustering

强海杂波是对于对海雷达目标检测性能影响最大的干扰项,目前已有的单一检测量和多特征联合检测算法性能极不稳定。针对上述难题,提出基于聚类的对海雷达目标检测算法。所提算法提取了3个特征量——相对幅度方差(RAV)、相对平均能量(RAP)和相对向量熵(RVE),对聚类算法中的k-近邻(kNN)算法进行模型改造,完成了虚警可控的kNN检测器,在该特征空间中利用kNN检测器实现目标与杂波分离。实测雷达数据实验结果显示,在观测时间0.512 s与1.024 s时,所提算法比基于分形的检测器的平均检测概率分别提高了56.2%和58.3%(HH极化模式下),相比基于三特征的检测器的平均检测概率分别提高了29.2%和31.3%。可以得出结论:所提算法能够实现复杂海况下的对海雷达目标检测,且检测效果明显优于基于三特征的检测器算法。

Strong sea clutter is the interference term that has the greatest impact on the target detection performance of sea radar,and the performance of the existing single detection quantity and multi-feature joint detection algorithms is extremely unstable.To solve the above problems,a sea radar target detection algorithm based on clustering is proposed.The algorithm extracts three feature quantities,that is,Relative Amplitude Variance(RAV),Relative Average Power(RAP)and Relative Vector Entropy(RVE).The k-Nearest Neighbor(kNN)algorithm in the clustering algorithm is modified to complete a kNN detector with controllable false alarm.In this feature space,the kNN detector is used to separate the target from clutters.The experimental results of measured radar data show that when the observation time is 0.512 s and 1.024 s,the average detection probability of this algorithm is 56.2%and 58.3%higher than that of the fractal based detector respectively(in HH polarization mode),and 29.2%and 31.3%higher than that of the three-feature-based detector respectively.It can be concluded that this algorithm can realize the sea radar target detection under complex sea conditions,and the detection effect is obviously better than that of the detector algorithm based on three features.