针对暗环境等复杂背景下的弱运动小目标识别的难题,提出了轨迹跟踪预测与模糊理论相结合的小目标跟踪算法。首先利用高通滤波和帧相减技术将背景和目标初步分离,然后利用轨迹跟踪法对目标点的位置进行预测,在动态搜索范围内寻找目标点进行图像序列分析,并结合模糊理论进行记忆更新来确定目标的可信度,最终通过多帧图像的轨迹跟踪找出真实的目标。实验结果表明:该算法能够获取热像仪拍摄的小目标弹丸落地前的运动轨迹曲线图,验证了该算法能够对低信噪比的运动小目标进行可靠检测。

Aiming at the problem of how to detect the SNR small moving target in the complex background, a new tracking algorithm based on moving target tracking forecasting and fuzzy algorithm was proposed. A space high pass filtering method is presented to improve the image quality for restraining the background noise and enhancing the dim target, then moving target tracking theory is used to forecast the location of target point, and then the memory was updated by using fuzzy theory, and the matching target point was searched for analyzing image sequence within a dynamic search rang. The experiment result shows that, the proposed arithmetic can calculate the small projectile trajectory before landing shoot by thermal imager, which verifies that the arithmetic can reliably detect small moving target of lower SNR.