中文摘要：

研究了在雷达／红外复合制导中如何将雷达的距离信息应用于红外图像序列的目标跟踪。针对目标渐进的图像序列，通过分析图像放大对相关性的影响，制定了一种根据弹目距离确定模板更新周期的模板更新策略。将距离更新、定时更新和基于目标模板缓冲区的模板更新相结合，提出了一种融合距离信息的自适应模板更新算法。以一座大楼为目标，验证了该方法的有效性。结果表明，该方法对目标渐进的图像序列有较强的自适应性，比传统方法大大提高了跟踪精度和跟踪稳定性，对实验中各类图像的跟踪正确率达90％以上。

英文摘要：

This paper mainly researches how to apply the distance information of millimeter wave radars to the object-tracking of infrared image sequences. In connection with the image sequences whose objects are gradually magnified, a template updating strategy in which the template updating cycle is confirmed by the distance between target and missile is proposed after analyzing the effect of image magnification on the correlation characteristics. In combination of the timing updating, template buffer updating with the distance updating, an adaptive template updating algorithm for fusing distance information is presented. By taking a building as experimental target, the effectiveness of the method is verified. It demonstrates that the method is very suitable for the proposed image sequences. As compared with the traditional tracking methods, it has improved the tracking accuracy and tracking stability greatly and its tracking accurateness is more than 90% for different images.