首页 | 本学科首页   官方微博 | 高级检索  
     

一种局部和全局相结合的光流计算方法
引用本文:张建明,钱东海.一种局部和全局相结合的光流计算方法[J].计算机工程与科学,2005,27(5):33-35.
作者姓名:张建明  钱东海
作者单位:上海大学精密机械系,上海,200072
基金项目:上海市教委青年科学基金资助项目(01QN60)
摘    要:光流场是计算机视觉的一个研究方向,微分法是计算光流场的一个常用方法,它分为全局方法和局部方法,全局方法能够得到100%的致密的光流场,而局部方法大多只能得到稀疏的光流场,但它在噪声情况下具有更好的鲁棒性。本文提出一种局部和全局相结合的方法.首先给出五点光流约束的局部方法,再结合全局方法,计算得到了既致密又 鲁棒的光流场。

关 键 词:计算机视觉  光流计算方法  光流场  图像灰度模式  图像处理  图像分割
文章编号:1007-130X(2005)05-0033-03
修稿时间:2003年10月15

An Optical Flow Computing Method with the Local and Global Methods Combined
ZHANG Jian-ming,QIAN Dong-hai.An Optical Flow Computing Method with the Local and Global Methods Combined[J].Computer Engineering & Science,2005,27(5):33-35.
Authors:ZHANG Jian-ming  QIAN Dong-hai
Abstract:Optical flow is a research direction of computer vision. Differential methods are widely used for computing optical flow. They can generally be divided into global methods and local methods. Global methods can yield 100% dense optical flow fields, while most local methods can only yield sparse optical flow fields. However, they are more robust under noise. In this paper, we present an optical flow computing method with both the local methods and the global methods combined. First the five-point-flow-constraint method is selected as the local method, and is then combined with a global method. As a result, both dense and robust optical flow fields can be obtained.
Keywords:computer vision  optical flow field  differential method
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《计算机工程与科学》浏览原始摘要信息
点击此处可从《计算机工程与科学》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号