| 本文已被：浏览 134次 下载 0次 |
|
|
|
| 基于双极化联合群稀疏TQWT的雷达海面目标检测方法 |
|
张帆, 夏正欢, 金世超, 刘新, 闫嘉益, 赵志龙, 张闯, 刑康, 刘宗强, 崔志颖, 薛长虎, 张涛
|
|
空间信息体系与融合应用全国重点实验室
|
|
| 摘要: |
| 针对低信杂噪比（Signal-to-Clutter-plus-Noise Ratio，SCNR）下海面微弱目标检测困难，以及单通道稀疏重构易引发虚警且丢失极化信息等问题，本文提出了一种基于双极化联合群稀疏可调Q因子小波变换（Tunable Q-factor Wavelet Transform，TQWT）的雷达海面目标检测方法。该方法首先利用目标与背景在时频域的形态差异，构建双Q因子过完备字典以实现信号成分的有效分离；其次，引入加权混合 范数构建联合群稀疏优化模型，约束双极化通道共享非零支撑集，协同抑制随机噪声与孤立杂波残差；最后，提取高Q因子稀疏系数，经逆TQWT变换重构分离后的目标信号，并结合韦布尔分布统计模型计算自适应阈值完成检测。仿真与实测数据均验证了该方法的有效性。实测结果表明，在低SCNR场景下，相较于单通道独立重构，该方法使VH和VV通道的信杂噪比分别提升约13dB，检测概率分别由58.8%和57.6%均提升至97.2%，且有效滤除了杂波残差。该方法在有效提升检测性能的同时，较好地保留了目标的极化散射特征。 |
| 关键词: 海面目标检测 可调Q因子小波变换（TQWT） 双极化 形态成分分析 |
| DOI： |
| 分类号:TN951 |
| 基金项目:北京市科技新星计划资助项目（20240484615） |
|
| Radar Sea-Surface Target Detection Method Based on Dual-Polarization Joint Group Sparse TQWT |
|
金世超, 刘新, 闫嘉益, 赵志龙, 张闯, 刑康, 刘宗强, 崔志颖, 薛长虎, 张涛
|
|
|
| Abstract: |
| To address the difficulty of detecting weak sea-surface targets under low signal-to-clutter-plus-noise ratio (SCNR) conditions, as well as the issues of high false alarm rates and the loss of polarization information caused by single-channel sparse reconstruction, this paper proposes a radar sea-surface target detection method based on dual-polarization joint group sparse Tunable Q-factor Wavelet Transform (TQWT). First, by exploiting the morphological differences between the target and the background in the time-frequency domain, this method constructs a dual Q-factor overcomplete dictionary to achieve effective separation of signal components. Second, a weighted mixed norm is introduced to formulate a joint group sparse optimization model, which constrains the dual-polarization channels to share a common non-zero support set, thereby jointly suppressing random noise and isolated clutter residuals. Finally, the high Q-factor sparse coefficients are extracted, and the separated target signal is reconstructed via inverse TQWT. Then, an adaptive threshold is calculated based on a Weibull distribution statistical model to accomplish target detection. Both simulated and measured data verify the effectiveness of the proposed method. Measured results demonstrate that in low SCNR scenarios, compared with single-channel independent reconstruction, the proposed method improves the SCNR of both VH and VV channels by approximately 13 dB, and increases their detection probabilities from 58.8% and 57.6% to 97.2%, respectively, while effectively suppressing clutter residuals. While effectively improving detection performance, this method also effectively preserves the polarimetric scattering characteristics of the targets. |
| Key words: sea-surface target detection Tunable Q-factor wavelet transform (TQWT) dual-polarization morphological component analysis (MCA) |