一种基于变形模板的椭圆跟踪算法

Ellipse Tracking Algorithms Based on Deformable Templates

给出一种新的基于活动轮廓与光流约束的椭圆跟踪算法。椭圆变形模板在引入对目标模型严格的全局约束的同时,降低了计算复杂度;基于模型的光流技术被用于提取目标的运动信息并引导轮廓的变形;借助扩展Kalman(EKF)滤波器有机组合目标的形状与运动信息;最后利用光流测量方程给出的误差测度及EKF给出的估计方差对虚假样本点作出判断与舍弃,从而保证对图像噪声、遮挡及伪边缘具有较强的克服能力。算法除可用于跟踪刚体运动之外,也可用于非刚体(例如人头)运动的跟踪。计算实例表明了算法的有效性。

A new ellipse-tracking algorithm based on the active contour and the optical flow technique was proposed. The ellipse-special deformable template brings in a strict global constraint on the contour and decreases the computational burden at the same time. A model-based optical flow is used to extract the motion features of object and to guide the deformation of contour. By using of the Extended Kalman Filter, the measurements of shape and motion are naturally integrated. Besides, an optical-flow based measurement error and the estimated covariance given by the EKF filter are used to detect and reject the contour sample points that correspond to noise, occlusions or spurious edge. The resulting algorithm can be used to tracking the motion of both rigid and nonrigid objects （such as human head）. Experiments with all this two application are presented to validate the algorithms.