基于二维卡尔曼滤波与最小二乘法的目标定位算法  

Circle Positioning Algorithm Based on 2D Kalman Filtering and Least Square Method

作　　者：马羽[1] 高蕾[1] 

机构地区：[1]平顶山教育学院计算机系,平顶山467000

出　　处：《计算机与数字工程》2017年第8期1652-1655,1669,共5页Computer & Digital Engineering

摘　　要：为了解决当前圆目标检测算法在噪声干扰与目标特征微弱的情况下,使其无法精确检测自动检测工件上的圆目标问题,论文提出了基于二维卡尔曼滤波与最小二乘法的工件圆目标定位检测算法。首先,基于一维卡尔曼滤波,利用图像像素灰度特性代替其一维信号模型,推导用于二维图像的卡尔曼滤波机制,有效滤除图像噪声,提取圆目标轮廓。引入最小二乘法,建立回归圆方程,拟合出目标圆轮廓,计算出圆心位置与半径,完成圆目标检测。实验测试结果显示:与当前工件图像圆目标检测算法相比,在面对较多噪声干扰而目标特征微弱时,论文算法拥有更高定位检测准确度与鲁棒性,能够较好地抵御噪声的影响。In order to solve the current find circle algorithm in dealing with noise and weak target feature,there can be no ac-curately find the target circle,a circle algorithm for prospecting based on Kalman filtering and least squares method is presented in this paper. First of all,based on one-dimensional Kalman filter,and the two-dimensional image of Kalman filter,the image noise is filtered out,the contour extraction of the target circle is completed. Then the least square method is let into,regression equation of a circle is established,target circle contour is fitted out,the position of the center of a circle with radius circle target search is cal-culated. Finally,the programming of the Kalman image filters and least squares fitting are integrated into the circle algorithm. The experimental results show that compared with the traditional algorithm,the algorithm has a higher accuracy of circle finding in the face of more noise interference and the feature of the target circle is not obvious.

关 键 词：卡尔曼滤波 最小二乘法 圆目标定位 图像噪声 回归圆方程 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]