基于卡尔曼预测算法的云台三维空间目标跟踪

TARGET TRACKING FOR TILT IN 3D-SPACE BASED ON KALMAN PREDICTION ALGORITHM

基于卡尔曼滤波预测算法建立了云台三维空间目标跟踪模型,提出了目标运动状态运算的具体步骤.该模型充分利用统计信息和包含噪声的已观测信息,对目标下一个时刻的出现坐标给出了最优预测,可实时控制摄像云台的预置角度,为云台去跟踪目标提供了技术保证,便于完成目标图像数据的采集、分析与识别.三维空间仿真结果表明了该跟踪模型的正确性和有效性,具有良好的应用前景.

Target tracking model for tilt in 3D-Space based on Kalman prediction algorithm is established in this paper. Meanwhile, the steps for how to calculate the object movement state are supplied in details. The proposal model uses both the statistics information and noise-contained observed information efficiently, and provides the optimal next interval pre- diction coordinates value which can control the pre-setting angel of shooting tilt in real time. This guarantees the techniques for tilt to track the target and accomplishments for the sampling, analysis and recognition of image data. Simulation results in 3D-space show the correctness and validity of this tracking model which has promising application.