基于时空域滤波的红外弱小目标背景抑制

Background Suppression for Infrared Dim Small Target Detection Based on Spatial-Temporal Filter

摘要:

在红外成像探测系统中, 对红外图像背景进行有效的抑制是准确检测出弱小目标的前提条件。基于目标在空域局部灰度稳定和时域运动连续的约束, 提出了一种基于时空域滤波的红外弱小目标背景抑制新方法。首先, 利用引导滤波保存图像细节和时域偏微分方程提取图像中突变区域的优势, 实现对图像空域与时域中平稳和强起伏不同特征复杂背景进行抑制处理; 然后, 将时空域背景抑制结果利用相与操作算子处理完成对高度类似弱小目标信号的剔除; 最后, 为恢复前期抑制结果中丢失的目标信息, 利用时空域融合结果作为引导图像进行进一步优化处理, 得到最终背景抑制结果。仿真实验采用两组低信杂比运动弱小目标红外图像序列进行方法验证, 并将该方法与几种背景抑制方法进行了比较, 实验结果表明: 该方法无论从主观视觉还是客观评价指标上均优于其他几种方法。

Abstract:

In the infrared imaging detection system, the effective suppression of infrared image background is the prerequisite of accurate detection of small and dim targets. In this paper, we propose a new method of infrared image background suppression based on spatial-temporal filtering based on the constraint of the target in the local gray stability in spatial domain and the motion continuity in temporal domain. First, the stable and strong fluctuation complex background suppression is processed in temporal and spatial domain using the superiority of guided image filtering in image detail preservation and partial differential equations in mutation area extraction. Then, the OR-operation is applied to eliminate the clutter similar to the small target. Finally, in order to recover the target information that lost in the early suppression, the result of OR-operation is used as the guided image to further optimize the suppression results. In the simulation experiment, two real low signal-to-clutter ratio (SCR) infrared imagery sequences containing moving dim small target are applied to verify the efficiency of the proposed method. Compared with several conventional methods, the experimental results show that the performance of proposed method is superior to these methods both in aspects of subjective visual and objective evaluation index.