强噪声背景红外微弱动目标集成检测

Integration detection algorithm of infrared small dim moving target based on strong complex noise

微弱运动目标检测一直是红外成像识别、跟踪领域中的重要课题。我们提出了一种新的基于强复杂噪声背景的红外微弱运动目标集成检测算法。针对复杂噪声,提出一种新的基于邻域梯度累积平方误差最小原则的自适应滤波算法。该算法,对各种加性、乘性噪声的滤波性能优异且均衡,适合对复杂噪声进行滤波。针对微小目标检测,我们提出一种基于运动平台的微小目标运动场集成检测算法。该算法,改变传统微小目标检测算法,要求提前对运动背景进行精确全域运动估计及补偿的模式,而只对图像运动场进行估计及补偿,提高了算法的可实现性。

Detection of small dim moving target is the important task in infrared image recognition and tracking. A new adapt filtering algorithm based on minimum cumulation-squared error （MCSE） of gradients in neighborhood is proposed for infrared small dim moving target in strong complex noise entironment. The performance of the algorithm is good and balanced for various plus noise, multiplicative noise, or strong complex noise. For detecting small objects, an integration detection algorithm of small moving target on moving camera is presented. The traditional detection algorithms need accurate global motion estimation and compensation, and the new algorithm only processes motion field and its application field is improved.