摘要
针对机械臂各个关节的角度值,提出了一种利用RGB-D图像的计算方法。首先通过Kinect获取彩色图像和深度图像并做对齐处理;接着利用阈值分割的方法提取机械臂目标,用细化算法提取出机械臂轮廓骨架;然后结合机械臂各刚体部件的位置关系对其关节点定位;最后利用向量的性质计算机械臂关节角度。该方法能够有效地计算出机械臂各关节角度,具有较高的准确性和较强的鲁棒性。
For the angle value of each joint of manipulator arm,a calculation method is proposed,using RGB-D image.Firstly,the RGB color and depth image are obtained,using Kinect and alignedto segment the manipulator arm from the depth image by the threshold segmentation.Secondly,the manipulator arm outline frames are extracted according to the thinning algorithm.By combining the position relation of the rigid body parts of the manipulator arm,the joint points are located and the joint angle of the manipulator is calculated based on the properties of the vector.The method can locate joint point and calculate the joint angle of the manipulator with high accuracy and robustness.
作者
刘海丰
陈鹏
Liu Haifeng Chen Peng(Hubei Key Laboratory of intelligent robot, Wuhan Institute of Technology, Wuhan , Hubei 430205, China School of Computer Science and Engineering, Wuhan Institute of Technology, Wuhan , Hubei 430205, China)
出处
《湖北工程学院学报》
2016年第6期97-101,共5页
Journal of Hubei Engineering University
基金
湖北省教育厅青年人才项目(Q20111504)
武汉工程大学科学研究基金项目(12116021)
武汉工程大学研究生教育创新基金项目(CX201274)
关键词
机械臂
深度图像
细化算法
关节点定位
关节角度
manipulator arm
depth image
thinning algorithm
joint point calibrated
joint angle