基于约束序贯M估计的时空域融合红外杂波抑制

Spatial-temporal Fused Filtering for Infrared Clutter Suppression Based on Restricted Sequential M-estimation

针对红外弱小目标检测中的强背景杂波干扰抑制问题进行研究,提出了一种基于参数约束序贯M估计的时空域融合自适应杂波抑制算法。该算法首先在分析序列图像帧间失配的基础上建立了一种改进的时空域融合背景预测模型,结合二维离散傅里叶快速变换图像配准和双线性插值方法进行灰度值估计;然后,基于约束序贯M估计方法进行模型参数的自适应估计,利用双阈值方法提取参数估计样本,并引入遗忘因子和控制因子提高算法的稳健性。仿真实验表明:所提算法不仅能够在一般的杂波干扰环境下实现背景的滤除,而且能够在强杂波干扰环境下较好地实现背景的抑制与目标信号的保持;相对现有算法具有较明显的性能优势和更强的环境适应能力。

Research is conducted on the intense clutter background removal so as to detect infrared small dim target.An adaptive spatial-temporal fused filtering algorithm for infrared clutter suppression is proposed based on restricted parametric sequential M-estimation.Firstly,an improved spatial-temporal fused prediction model for clutter background is presented on the analysis of the mismatch between the sequential images.And the image intensity is estimated by means of two-dimensional fast discrete Fourie transform（2DFT） and bilinear interpolation.Secondly,the model parameters are adaptively reckoned on restricted sequential M-estimation as the data sampled to double thresholds.Meanwhile,the forget factor and control factor are introduced to improve the robustness.The experiments show that the proposed algorithm not only perfectly removes the background under common circumstance but also effectively filters the intense clutter and preserves the targets signal,which holds fairly better performance and robust accommodation to environment than the existing ones.