摘要
以数字图像相关方法为核心设计了一套应用于高速相机序列影像运动目标的跟踪方案,该方案首先在基准窗口周围以局域搜索方法获取目标点的整像素位移值,然后联合使用梯度算法和最小二乘算法估计目标点的亚像素位移值,从而获得目标点的像素坐标,实现对目标点的自动跟踪。本文应用该技术方案处理了构筑物振动台实验的高速相机视频测量影像数据,与商用软件PhotoModeler Scanner相比,跟踪结果具有同等精度,在计算速度上有优势。
In this paper, we propose a scheme based on digital image correlation method for realizing the high- speed camera sequence of images moving targets tracking. The scheme include two key steps:(1) to obtain the integer pixel coordinates of the target points in the image sequence by using local search method around a reference window;(2)to obtain the subpixel displacement values of the target points by joint use of gradient algorithm and least squares algorithm. When the pixel coordinates of the target points are calculated, the tracking of the target points is completed. In this paper, the application of the technical solution to deal with the structure shaking table test of the high-speed camera video measurement image data, compared with commercial software PhotoModeler Scanner, tracking results with the same accuracy, but the advantage is the faster processing speed.
出处
《地理信息世界》
2013年第5期51-58,共8页
Geomatics World
关键词
高速相机
序列影像
数字图像相关
运动目标跟踪
非接触测量
位移
High-speed Camera
Sequence of Images
Digital Image Correlation
Moving Targets Tracking
Non- Contact Measurement
Displacement