【摘要】 目前 ,运动目标跟踪是计算机视觉领域中最活跃的研究课题之一。本研究提出了基于模糊熵聚类和Kalman滤波预测的区域跟踪方法 ,用Kalman滤波的预测值作为下一帧图像运动区域的聚类中心 ,从而减少了迭代次数 ,加强了跟踪的实时性。算法有效地利用了计算机视觉技术从图像序列中检测出运动区域 ,并获得了运动目标的轨迹 ,统计了运动目标的数目。更多还原

【Abstract】 Moving targets tracking is currently one of the most active research topics in the domain of computer vision.A method of moving targets tracking based on fuzzy entropy clustering and Kalman filter forecasting is advanced,and the forecasting value of Kalman filter as the center of next frame cluster is used.This method reduces the iterative numbers and increases the real-time tracking.The method effectively applies the teachnology of the computer vision detecting the motion region in image sequences.It acquires with the trajectories of moving targets and counts the number of moving targets.更多还原