摘要
针对复杂海杂波环境下雷达目标跟踪受到强杂波干扰,跟踪难度增加的问题,提出优化多核相关滤波的弱小目标检测前跟踪算法.通过引入分数阶傅里叶变换,将弱小目标从特征不明显的时域变换到幅值变化比较明显的分数阶域,并进行滤波以提高对弱小目标的检测率.针对多核相关滤波(MKCF)方法中模板提取不鲁棒的问题,优化模板提取方法,结合卡尔曼滤波进行目标匹配,根据目标类型采用不同的模板提取方法,采用最大似然方法融合预测结果,以增强目标的跟踪精度.结合检测前多帧跟踪算法,综合多帧信息,选取最佳轨迹估计.实验结果表明,提出算法能够适应复杂的海杂波环境,对低信噪比、杂波干扰强的多目标进行有效跟踪,与传统方法相比具有较好的精度.
An optimized track-before-detect algorithm for weak and small target detection based on improved multikernel correlation filtering was proposed aiming at the problem that radar target tracking under complex sea clutter environments was subject to strong clutter interference and increased tracking difficulty.Weak and small targets were transformed from the time domain with indistinct features to the fractional Fourier domain where amplitude variations are more pronounced by introducing fractional Fourier transform,followed by filtering to enhance detection rates.The template extraction approach was optimized by incorporating Kalman filtering for target matching in order to address the non-robust template extraction issue in multi-kernel correlation filtering(MKCF)methods.Different template extraction methods were adopted according to target types,and prediction results were fused by using maximum likelihood method in order to improve tracking accuracy.Comprehensive multi-frame information was utilized to select optimal trajectory estimation combined with multi-frame track-before-detect(TBD)algorithm.The experimental results demonstrate that the proposed algorithm effectively adapts to complex sea clutter environments,achieves efficient multi-target tracking under low signal-to-noise ratio and strong clutter interference,and demonstrates superior accuracy compared with traditional methods.
作者
吴晓佳
杨金龙
赵豪豪
WU Xiaojia;YANG Jinlong;ZHAO Haohao(School of Artificial Intelligence and Computer Science,Jiangnan University,Wuxi 214122,China;Engineering Research Center of Integration and Application of Digital Learning Technology,Ministry of Education,Beijing 100039,China)
出处
《浙江大学学报(工学版)》
北大核心
2025年第5期947-955,972,共10页
Journal of Zhejiang University:Engineering Science
基金
数字化学习技术集成与应用教育部工程研究中心2023年创新基金:资助项目(1311013)
江苏省自然科学基金:资助项目(BK20181340)。
作者简介
吴晓佳(1998-),女,硕士生,从事目标跟踪的研究.orcid.org/0009-0007-8027-160X.E-mail:6213113030@stu.jiangnan.edu.cn;通信联系人:杨金龙,男,副教授.orcid.org/0000-0001-9548-4236.E-mail:jlyang@jiangnan.edu.cn。
