【摘要】 针对天基短波红外图像中弱小目标易被云层、地表杂波淹没，且在低信杂比条件下检测困难的问题，提出一种融合安德森加速的自正则化加权稀疏模型（Self-Regularized Weighted Sparse, SRWS）与相对局部对比度（Relative Local Contrast Measure, RLCM）的改进检测方法。通过引入安德森加速机制，显著降低了背景估计的计算复杂度，利用背景残差图和RLCM实现了多尺度目标检测性。实验结果表明，本文算法在复杂背景下仍保持优异性能，接收者操作曲线下面积（Area Under Curve, AUC）最高达0.950，最低不低于0.842；信杂比增益（Signal-to-Clutter Ratio Gain, SCRG）显著优于红外图像块（Infrared Patch Image, IPI）、局部对比度法（Local Contrast Measure, LCM）等传统方法。本研究有效提升了天基短波红外弱小目标的检测精度与稳定性，为复杂背景下的遥感目标检测提供了可靠的解决方案。更多还原

【Abstract】 To address the challenge of detecting dim and small targets in space-based short-wave infrared(SWIR) imagery-where targets are readily obscured by cloud cover and ground clutter under low signal-toclutter ratio(SCR) conditions-an enhanced detection algorithm is proposed that integrates Anderson-accelerated Self-Regularized Weighted Sparse(SRWS) modeling with the Relative Local Contrast Measure(RLCM). Computational complexity in background estimation is substantially reduced through the incorporation of Anderson acceleration, while multi-scale target detection is achieved via background residual maps combined with RLCM. Experiments conducted on 289 SWIR images spanning seven representative scenarios demonstrate consistently strong performance in complex backgrounds, with the AUC reaching 0. 950 and remaining no lower than 0. 842 under the most challenging conditions. The signal-to-clutter ratio gain(SCRG) is significantly improved relative to conventional methods, including IPI and LCM. Overall, detection accuracy and robustness for dim and small targets in space-based SWIR remote sensing are effectively enhanced, providing a reliable solution for target detection in complex background environments.更多还原