关键词: 图像制导, 位姿信息, 模板匹配, 模板更新, 跟踪漂移

Abstract: While the image-guided target tracking system is tracking, image rotation, the changes of the scale of targets and imaging perspectives will cause tracking drift problems. In order to solve these problems, a method of correcting and updating the matching template using the position and orientation information of the camera was proposed. According to the relative position and orientation information of camera and targets, the objective perspective imaging model was established as well as the template correction equation of different camera position and orientation; the correction equation was decomposed into two parts which were the rotation of real-time image and the affine transformation of template to reduce the changes in error; the update strategy relative to the correlation coefficient and the distance information was designed to adapt to the changes while target tracking system was gradually close to the target. The algorithm had been validated using the flight simulation video generated by the visual-based simulation software Vega Prime. The experimental results show that the proposed algorithm can effectively adapt to the changes of the target scale, imaging perspective as well as image rotation while the image-guided tracking system is tracking, and it reduces the tracking drift.

Key words: image-guided, position and orientation information, template marching, template updating, tracking drift