论文部分内容阅读
设计了一种基于机器视觉的天轮偏摆检测系统。利用机器视觉系统的CCD摄像机实时采集天轮偏摆运行状态视频;对每一帧图像进行图像滤波和图像增强,采用Mean Shift算法实现天轮偏摆的跟踪;经坐标变换将跟踪目标的像素偏移换算为实际物理尺寸,从而计算出天轮偏摆。试验结果表明,设计的天轮偏摆检测系统现场安装检测方便且能满足精度要求。
A vision detecting system based on machine vision is designed. Using CCD camera of machine vision system to collect real-time video of heave yaw of runner; carrying out image filtering and image enhancement for each frame of image, adopting Mean Shift algorithm to track yawing of heave wheel; tracking pixel shift of target by coordinate transformation The displacement is calculated as the actual physical size to calculate the deflection of the day wheel. The test results show that the design of the yaw-deflection detection system on-site installation and testing convenience and can meet the accuracy requirements.