基于HS约束与轮廓条件的光流场计算  

Optical flow computation method based on HS constraint and outline condition

作　　者：谢剑斌[1] 王晖[1] 程江华[1] 刘通[1] 

机构地区：[1]国防科技大学电子科学与工程学院,湖南长沙410073

出　　处：《系统工程与电子技术》2009年第4期761-763,共3页Systems Engineering and Electronics

基　　金："十一五"国防重点预研项目(102010201)资助课题

摘　　要：针对视频图像中灰度变化与速度场可能存在的非连续区域,提出了一种基于HS(Horn-Schunck)约束与轮廓条件的光流场计算新方法。该方法首先采用高斯滤波器减少噪声干扰;然后采用改进的Horn-Schunck算法计算时空梯度,结合两种不同的约束条件求取光流速度;最后的仿真实验结果表明,该方法能够准确地计算视频图像非连续区域的光流场,并有更好的鲁棒性。This paper presents a new optical flow computation method based on HS constraint and outline condition for the discrete region of both gray change and velocity flow in video image. Firstly, a Gaussian filter is used to smooth the image so as to reduce noise. Secondly, an improved Horn-Schunck method is used to calculate spatiotemporal gradient, and the velocity of optical flow is calculated on the basis of combining the two constraints. Simulation results show that the method can calculate optical flow in discontinuous regions and is robust.

关 键 词：非连续区域 光流 时空梯度 高斯滤波器 

分 类 号：TP751.2[自动化与计算机技术—检测技术与自动化装置]