光电图像序列运动弱目标实时检测算法

针对光电探测图像序列中的运动弱小目标实时检测问题,提出了一种基于时空域融合滤波的弱目标检测算法。算法在空域上利用形态学Tophat滤波抑制背景增强目标,在时域上通过改进的帧间差分方法增强运动目标,两者融合后经自适应门限分割与航迹关联确认目标。实际录取数据分析结果表明,算法全面考虑运动弱小目标在时域与空域方面的特性,能更有效地从复杂背景中检测低信噪比运动弱小目标,减小了虚警率,抗噪声干扰能力强。

Real-time algorithm for small moving target detection in photoelectric image sequences

For the real-time small target detection of the photoelectric image sequences, a temporal- spatial fusion filtering algorithm is proposed. It suppresses background and enhances targets by morphologic Tophat transform in spatial domain, and detects moving target by improved frame difference method in temporal domain, and then affirms target by adaptive segmentation of the fusion result and track association. The detecting result shows the algorithm can fully take into account the characteristic of the small moving target in temporal and spatial domain, and can be more effective to detect low Signal-to-Noise Ratio (SNR) moving small targets in complex background, reduce false alarm and resist noise disturbance.