【摘要】 基于颜色的粒子滤波实时跟踪算法主要是利用视频图像的颜色直方图信息,综合考虑运动预测和帧间的相似性来确定目标的位置。针对影响粒子滤波算法性能的关键技术,提出了基于混合高斯模型的粒子滤波算法,并将其用于基于颜色的非刚性目标的实时跟踪相关问题。该算法使用混合高斯模型表示粒子,在每个时刻的修正步骤之后,采用EM算法对粒子进行重新拟合。仿真实验表明,本算法在保证跟踪准确度的同时,可以满足实时跟踪的要求。更多还原

【Abstract】 The color-based particle filter for real-time object tracking determines the location of the object by using color information and incorporating the motion prediction and frame similarity.According to the key technique influencing the particle filter,a Gassian mixture particle filter for non-rigid object tracking is presented.It uses Gassian mixture model to represent particles and adopts EM algorithm to refit particles after correction step at each time.The experimental result indicates that the algorithm can realize the real-time object tracking while ensuring the tracking veracity.更多还原