作　　者： ; ; ;

机构地区： 深圳信息职业技术学院

出　　处： 《深圳信息职业技术学院学报》 2013年第3期56-60,共5页

摘　　要： 激光光斑中心的精确定位是机器视觉检测领域的关键技术。本文就双摄像头获取的十字激光图像进行光斑中心检测。首先,根据激光光斑亮度高的特点,以最大类间方差法对激光图像进行二值化处理,既可以简化后续处理、提高实时性,又可以排除背景中的各种干扰因素。然后分别检测两条激光中心线点群作为直线拟合的样本点,并采用最小二乘算法拟合两条激光线得到激光光斑中心的亚像素级定位。最后,对得到的激光光斑中心进行分析,去除由于物体表面不光滑等因素产生的干扰线对检测结果的影响。实验结果表明该算法能够满足实时性需求,鲁棒性好,定位精度高。 High-precision detection of laser spot center is the critical technique in computer vision detection field. In this paper, we detect the center of cross laser images obtained from two cameras. To simplify the following processing and eliminate the interference in the background, the image is segmented to binary image with maximum variance between clusters. Then, the center dots on cross laser lines are detected as the samples, and processed to get the sub-pixel center of cross laser with least square fitting. Finally, we analyze the result and get rid of the centers generated because of arris on the object surface. Experiment results show the processing speed of our method described in this paper can meet the need of real-time system, which has good robust and high location precision.

关 键 词： 最小二乘拟合 激光光斑 最大类间方差法

领　　域： [自动化与计算机技术] [自动化与计算机技术]