Research on tracking algorithm of moving target based on computer vision

被引：0

作者：

Fu, Ronghui ^{[1
]}

机构：

[1] College of Computer Science, Neijiang Normal University, Neijiang, SiChuan,641110, China

来源：

Boletin Tecnico/Technical Bulletin | 2017年 / 55卷 / 19期

关键词：

Bayesian filtering - Filtering algorithm - Kalman filtering algorithms - Kalman-filtering - Moving target tracking - Prediction accuracy - Simulation studies - Tracking algorithm;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

KF algorithm would produce larger prediction error, and sometimes may lead to filter divergence. In this paper, further research and improvement have been made aiming at the disadvantage that tracking effect of filtering algorithm based on KF is not ideal when the mutation of moving target state occurs. This paper deeply studies the moving target tracking theory including the Bayesian filtering theory and the Kalman filtering algorithm and proposes Kalman filtering algorithm based on multi-innovation theory. A simulation study is carried out and shows that the improved MI - KF algorithm is effective and has a higher prediction accuracy than KF algorithm.

引用

页码：425 / 434