基于像素匹配和卡尔曼滤波的运动估计在浮选泡沫流速特征提取上的应用

Pixel Matching and Discrete Kalman Estimator based on Flotation Froth Motion Velocity Extraction and Analysis

为了解决传统特征匹配算法难以准确提取浮选泡沫流速特征的问题,本文提出一种基于像素匹配和卡尔曼滤波相结合的算法。该算法首先综合考虑前后两帧图像上面像素点的流动,在基于理想情况下的灰度值不变、梯度时空不变以及运动连续性假设,提出使用欧拉-拉格朗日方程与计算数学最小化约束能量方程,得到的结果作为卡尔曼预测的观测值参与测量更新,从而在状态更新中得到最优的泡沫流速特征向量。实验结果表明,该算法计算准确,平稳性高,实时性好,可以很好地作为浮选机优化控制的原始参数。

In order to solve the problem that the traditional feature matching algorithm can not accurately extract the velocity characteristics of flotation froth, a new algorithm based on pixel matching and Kalman filter is proposed. The algorithm firstly considers the flow of the pixels between two adjacent frames, assumes Grey value, Gradient Constancy and Motion Smoothness, uses the Euler-Lagrange equation and computational mathematics methods to minimize the energy equation. And then the obtained results are taken as the measurements of the Kalman prediction, thus the best estimation result is obtained. The experimental results show that the algorithm is accurate, stable and highly real-time which can be used as the original parameters of flotation optimization control.