对红外背景、低信噪比条件下的小目标检测技术研究

低信噪比条件下小目标检测技术的研究是红外信息处理中的关键性问题．灰度形态滤波是一种结合信号形状的非线性运算，基于灰度形态滤波的预处理可有效地抑制图像中的背景干扰和噪声．对于多帧连续运动图像，经过前端的滤波预处理后，图像之间存在着一定的关联性，利用相关性结合神经网络算法可以对运动目标轨迹进行有效的刻画．

Study of small target detection in IR background and low SNR

The study of small target detection in low SNR is the key problem about processing of infrared information.A grayscale morphology filter is a kind of nonlinear operation with signal shape.The pretreatment based on a grayscale morphology filter can effectively restrain the background disturb and yawp in the picture. For many frames of moving pictures,there exist relativity after front pretreatment of a filter.The trace of moving target can be depicted effectively by using relativity and a neural network algorithm.