摘要
图像处理或在工业控制中经常要用到最小二乘直线拟合,对于有奇异点的直线拟合,传统的最小二乘法拟合误差较大,难以满足较高精度的要求。卡尔曼滤波算法具有最小无偏方差性,能够去除测量系统中的随机误差,将卡尔曼滤波算法与传统最小二乘法结合,建立了一种基于卡尔曼滤波预处理的最小二乘估计的新方法,获得了比传统最小二乘法效果更好的估计结果。试验证明了该方法的有效性和高精度性。
Image processing or regular use in industrial control least square fitting for a singular point of a straight line, the traditional method of least squares fitting error is large, it is difficult to meet the requirements of high accuracy. Kalman filtering algorithm with minimum unbiased variance can remove the random error in the measurement system, the Kalman filter algorithm combined with the traditional method of least squares, a Kalman filter pretreatment least squares estimation of new method to obtain a better effect than the conventional method of least squares estimation result. The tests proved the effectiveness and precision of the method.
出处
《自动化与仪器仪表》
2013年第3期140-142,共3页
Automation & Instrumentation
关键词
卡尔曼滤波
最小二乘法
直线拟合
Kalman filter
least squares method
fitting a straight line
