摘要
提出了基于Kinect摄像机的机器视觉测速方法,用于实时的测量运用目标的速度信息。首先,设计出固定在运动目标上的由五个已知位置关系的圆组成的标识物,利用图像处理算法根据标志物的拓扑关系检测出运动目标标志物在Kinect坐标系下的位置;然后,将Kinect坐标系坐标通过坐标变换转换到世界坐标系;最后,根据采样时间内测出的距离求出速度信息。通过以上3步骤,依次解决运动目标检测和运动目标测速问题。理论分析和试验结果表明,本文方法在移动机器人运动或者静止情况下能够处理运动目标作直线运动、原地转动及其合成的曲线运动等不同运动形式下的速度测量,应用范围广泛,计算实时性和准确性高。
A novel method based on machine vision using Kinect camera is proposed to realize the real- time velocity measurement of moving object. Firstly, a moving object marker fixed on moving object is designed, which consists of five circles with certain position relation, and then the image processing algo- rithm is utilized to detect moving object marker position precisely in the Kinect coordinate system accord- ing to the topological relationship of the object marker. Secondly, the coordinates in Kineet system are transformed to world coordinate system. Finally, the velocity information is calculated with the distance in certain sampling time. Theoretical analysis and experimental results show that the method can be used to measure the velocity of the moving object in the form of linear motion,situ rotation and the synthetic curvilinear motion, no matter the mobile robot is in the motion or stationary situation. It has a wide range of applications with real-time computing and high accuracy.
作者
许烁
张二东
朱小生
江渊广
周可璞
XU Shuo ZHANG Er-dong ZHU Xiao-sheng JIANG Yuan-guang ZHOU Ke-pu(School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200072, China Shanghai Key Laboratory of Intelligent Manufacturing and Robotics, Shanghai 200072 ,China)
出处
《光电子．激光》
EI
CAS
CSCD
北大核心
2017年第3期316-325,共10页
Journal of Optoelectronics·Laser
关键词
动态环境
移动机器人
速度测量
机器视觉
图像处理
dynamic environment
mobile robot
velocity measurement
machine vision
image process-ing
