基于OpenMV的目标跟踪系统设计

Design of Target Tracking System based on OpenMV

为解决传统目标跟踪系统存在的结构复杂、资源利用率低等问题,提出一种基于嵌入式摄像头OpenMV的目标跟踪系统设计方案.在摄像头采集到图像信息并进行图像预处理后,判别出需要跟踪的运动目标以及其形心像素坐标,并在图像中进行标定.依据得到的连续的形心坐标,结合图像捕捉帧率,采用线性拟合的方式对运动目标下一帧的位置进行预测,并运用PID算法控制云台系统跟踪此预测坐标.系统加入对运动目标运动轨迹的判别算法,有效提高了系统的抗干扰能力.试验结果表明此系统可以准确预测并跟踪运动目标的位置,当画面中突然闯入相同干扰目标时,系统仍可以正确跟踪运动目标,具有较强的鲁棒性.

To solve the problem with complex structure and low resource utilization of traditional target tracking system,a design scheme of target tracking system,based on OpenMV,is proposed.The tracking moving target and its centroidal pixel coordinates are identified and calibrated in the image after the camera gathers the image information and performs the necessary image processing.On the basis of continuous centroidal coordinates,the location of the next frame is predicted by using image capture frame rate and the method of linear fitting.The coordinates are then tracked using PID algorithm control pan tilt system to track the coordinates.A discriminant algorithm is added to a moving target movement trajectory,which greatly increases the anti-interference ability of the system.The experimental results show that the system can accurately predict and track the position of the moving target.When the same interference target is suddenly intruded in the picture,the system can track the moving target correctly with strong robustness.