一种有效的激光光斑中心检测算法研究

An Efficient Detection Algorithm for Laser Spot Center

激光光斑中心的准确定位对光学检测精度起着决定性作用。传统圆拟合算法消噪能力不强,导致预测失准。通过图像处理和数据分析进行消噪,引入圆拟合率,排除误差较大噪声点干扰。提出基于二维向量空间圆拟合算法,将原始光斑图像映射到二维向量空间,根据圆光斑的几何性质,组建超定方程组求解光斑中心坐标和半径。实验结果表明,该方法在满足实时性前提下,检测精度优于传统重心法与Hough变换法,适合各种复杂环境下的光斑中心检测。

The precise location of the laser spot center plays a decisive role in the precision detection optical system. For the traditional circle fitting algorithm, the noise cancellation capability is not effective, so it＇s easy to cause the final forecast is incorrect. In order to effectively eliminate the error noise interference, this paper uses an eliminating noise method based on image processing and data analysis and introduces the concept of circle fitting rate. This paper presents a circle fitting algorithm based on two-dimensional vector space. In this algorithm, the original image is mapped into a two-dimensional vector space. According to the geometrical properties of circular spot, over-determined equations are listed and solved for obtaining the spot center coordinates and radius. Experimental results show that to meet the premises of real-time measurement, the accuracy of this algorithm is superior to the traditional algorithms such as gravity model and Hough transform. This algorithm is suitable for spot center detection in all kinds of complicated environment.