红外弱小运动目标实时检测的规整化滤波方法

NOVEL REGULARIZING FILTERING METHOD FOR REALTIME DETECTING IR DIM SMALL MOVING TARGET

低信噪比小目标检测能力决定着系统的探测灵敏度和作用距离,是反映红外低可观测目标识别能力至关重要的一项核心技术.自适应杂波背景抑制技术是实现这一目标的有效途径.本文将杂波背景抑制滤波归纳为逆问题求解的优化问题;建立了新的红外弱小目标/背景模型,在此基础上发展了一种基于规整化技术的滤波框架;并提出了"去杂波-保目标"规整化的自适应各向异性滤波新算法.详细的理论分析和试验结果表明:该算法能在单步处理中消除杂波背景、同时增强弱小目标信号,运算量小;对低信噪比的强杂波背景表现出良好的滤波性能和适应能力,且结构简单、利于硬件实时实现.

The performances of detecting small targets at low signal-to- noise ratio （SNR） decide the detection sensitivity and effective ranging of a system. It is a leading key technique to indicate the ability of recognizing low-observable target in infrared （IR） imagery. Adaptive background estimation method is an efficient avenue to complete this task. In this study the clutter background prediction method was reduced to the optimization problems of inverse problem. New models of target/background in IR images were established, based on which a new filtering framework using regnlarization technology was presented, and then a novel anisotropic filtering method with the clutter-removal target-preserving＇ regnlarization was proposed. Detailed theoretical analyses and experimental results show that this method can remove the clutter background and simultaneously enhance the signal of interest in one processing step, and its computing complexity is very little; and it can also provide good filtering results and adaptability to IR targets with strong clutter background; moreover, its logical structure is simple to be implemented in real-time system.