非合作双基地雷达弱目标KCF⁃TBD方法

Non⁃Cooperative Bistatic Radar Weak Target KCF⁃TBD Method

非合作双基地雷达因其特殊的探测方式,致使回波中目标信噪比较低,雷达扫描周期的帧与帧之间探测并不稳定,因此传统的先检测后跟踪(Detect⁃Before⁃Track,DBT)在这种场景下并不适用。本文利用动态规划检测前跟踪(Dynamic Programming Track⁃Before⁃Detect,DP⁃TBD)值函数累积的思想,结合现用于视觉跟踪中的核相关滤波器(Kernelized Correlation Filters,KCF),将峰值旁瓣比(Peak Sidelobe Ratio,PSR)作为值函数,提出了KCF⁃TBD,并采用深度学习技术对跟踪框进行修正,以解决实际目标跟踪中出现的漂移、丢失等问题,从而提高跟踪过程的稳定性和适应性。在实际的海面目标数据上的验证结果显示,所提算法具有较高的跟踪稳定性和适应性。

Non⁃cooperative bistatic radar has a special detection mode that results in a low signal⁃to⁃noise ratio(SNR)of the target in the echo,which makes the detection between frames in the radar scanning cycle unstable.The traditional detect⁃before⁃track(DBT)method is not suitable for this scenario.To address this issue,this paper proposes a novel approach named KCF⁃TBD that combines dynamic programming track⁃before⁃detect(DP⁃TBD)value function accumulation and kernelized correlation filters(KCF)used in visual tracking.The peak sidelobe ratio(PSR)is used as the value function,and the tracking box is modified using deep learning technology to tackle drift and loss issues in tar⁃get tracking.The experimental results on the actual sea surface target data demonstrate that the proposed algorithm has high tracking stability and adaptability.