摘要
在基于改进Snake模型的基础上,结合Kalman滤波进行运动预测,提出了跟踪运动物体轮廓的一种方法.在物体轮廓的提取方面,将MINIMAX准则应用于参数的自动选取,对霍夫变换进行改进使其能应用于Snake初始轮廓的自动选取,对动态规划算法进行改进使其应用于Snake模型,使其有更好的收敛效果.并采用Kalman滤波器预测物体的运动轨迹,实现在视频摄像中运动物体的有效跟踪.
An object contour tracking algorithm based on improved Snake model and Kalman filter is proposed in this paper. For object contour extracting, MINIMAX criterion is used to solve the automatic choice of parameters for the model, and an improved Hough transform is used to select the initial contour automatically. An improved dynamic programming algorithm is also used in the Snake model in order to have a better convergence result. In this paper,the Kalman filter is applied to predict the possible trajectories of moving target, and track the moving object in the video frames effectively.
出处
《浙江工业大学学报》
CAS
北大核心
2010年第2期168-172,共5页
Journal of Zhejiang University of Technology
基金
浙江省科技计划重点项目(2007C21012)
