红外成象型空空导弹目标识别与跟踪算法研究

根据红外成象制导型空空导弹图象传感器在目标识别与跟踪过程中获得的图象序列的特点。提出了一种基于红外图像序列的弹道终端红外图象跟踪时的跟踪算法．实现了目标特征点的自动识别与跟踪。文中以最大距离法对目标图象进行分割．以机头部位为局部图象跟踪时的跟踪部位，分析了导引头投影成象模型以及弹目距离、相对速度对目标图象变化情况的影响．研究了中段形心跟踪模式与末端局部图象跟踪模式的转换时机．定义了确定目标充满视场程度以及阻塞视场与否的两个判据。最后，以实测的飞机红外图象对文中提出的算法进行了仿真，结果表明本文的算法是可行和有效的。

Investigation of Target Recognition and Tracking Algorithm for Infrared Imaging Air-to-Air Missile

In accordance with the characteristics of image sequence obtained by the image detector of infrared imaging guidance air-to-air missile,a trajectory terminal feature point automatic recognition and tracking algorithm based on infrared image sequence is proposed in this paper. The maximum distance method is used to segment the target image from background, the nose cone of aircraft was selected as tracking point during partial image tracking, The model of perspective projection for imaging seeker is provided and the effect of missile-target distance and relative velocity on aircraft imaging is analyzed, the switch occasion from centroid tracking to partial tracking is studied,and we provide the definitions of overflow and jam-up degree for field of view.Finally, we do some experiments with real infrared image of aircraft. The experimental results show that the algorithm provided in this paper is feasible and effective.