摘要
针对传统算法无法准确提取线结构光条中曲率变化较大区域中心线的缺陷,提出了一种基于改进灰度重心法的线结构光中心线提取算法。基于阈值法提取感兴趣区域和开闭运算滤除图像中的噪声点,获取质量相对较高的待处理图像;利用灰度重心法提取光条中心点,得到初始坐标并设置一个矩形计算区域,根据最小二乘法计算每个初始点的方向向量和法向量;在设定的矩形区域内计算初始点在其法线方向上的偏移量,得到精确的中心点坐标。实验结果表明:提出的算法能在线结构光中心中曲率变化较大区域精确提取中心线,提取精度均值为0.096 pixels,精度比率为44.2%,平均提取速度为0.082 s。
Aiming at the defect that the traditional gray gravity method cannot accurately extract the center line of the region with large curvature change in the line structured light strip,an algorithm for extracting the center line of line structured light based on improved gray barycenter method is proposed.The region of interest is extracted based on the threshold method,and the noise points in images are filtered through the open and close operation to obtain the image to be processed with relatively high quality.Center points of the light strip through the gray gravity method.And a rectangular calculation area is set up.The direction vector and normal vector of the initial points are extracted according to the least square method.Finally,the offset of the initial point in the normal direction is recalculated in the rectangular area to obtain the exact coordinates of the center point.Experimental results show that the proposed algorithm can accurately extract the center line from the region with large curvature change in the center of line structured light,achieving a mean precision of 0.096 pixels,a mean precision rate of 44.2%,and an average extraction speed of 0.082 seconds.
作者
夏鑫
付生鹏
夏仁波
赵吉宾
侯维广
XIA Xin;FU Shengpeng;XIA Renbo;ZHAO Jibin;HOU Weiguang(University of Chinese Academy of Sciences,Beijing 100049,China;State Key Laboratory of Robotics,Shenyang Institute of Automation,Chinese Academy of Sciences,Shenyang Liaoning 110016,China;Institutes for Robotics and Intelligent Manufacturing,Chinese Academy of Sciences,Shenyang Liaoning 110169,China)
出处
《激光杂志》
CAS
北大核心
2024年第1期75-79,共5页
Laser Journal
基金
国家自然科学基金(No.51805527、52075532、91948203)。
关键词
线结构光中心线
改进灰度重心法
区域法向量
直线拟合
centerline of line structured light
improved gray gravity method
region normal vector
linear fitting
