视频微小运动放大的加速方法

计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (24): 195-200.

视频微小运动放大的加速方法

李乐鹏，雷林，孙水发，尹辉，董方敏

1.三峡大学湖北省水电工程智能视觉监测重点实验室，湖北宜昌 443002
2.三峡大学三峡地区地质灾害与生态环境湖北省协同创新中心，湖北宜昌 443002

出版日期:2015-12-15 发布日期:2015-12-30

Improved video small motion magnification processing

LI Lepeng, LEI Lin, SUN Shuifa, YIN Hui, DONG Fangmin

1.Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering, China Three Gorges University, Yichang, Hubei 443002, China
2.Collaborative Innovation Center for Geo-Hazards and Eco-Environment in Three Gorges Area, Hubei Province, China Three Gorges University, Yichang, Hubei 443002, China

Online:2015-12-15 Published:2015-12-30

摘要/Abstract

摘要： 通过放大视频中的微小运动，可以展现一些肉眼不可见的运动信息。原始欧拉视频放大技术（EVM方法）将输入的视频由RGB颜色空间转换为YIQ颜色空间，对视频的YIQ三个通道进行运动放大处理，这样的处理过程十分耗时。在原方法的基础上，分别提出了基于RGB颜色空间和基于Y通道的运动放大加速方法。将Y通道方法、RGB方法和EVM方法进行实验对比，结果表明：RGB方法相对EVM方法在处理速度方面有所提升，但是运动放大效果差别较大；Y通道方法相对于EVM方法在处理速度方面得到了明显的提升，而且在放大效果方面基本保持一致。

关键词: 微小运动, 欧拉视频放大, RGB颜色空间, YIQ颜色空间, 运动放大

Abstract: Motion magnification can be used to reveal subtle motions in video which are invisible to the naked eyes. The conventional Eulerian Video Magnification（EVM） method converts the input video signal from RGB color space to YIQ color space then magnifies the motion by processing three channels independently in the YIQ color space, but it is time-consuming. In this paper, the RGB space based and the Y channel only based motion magnification methods are proposed separately. The experimental results of three methods（the Y channel based method, the RGB space based method and the original EVM methods） show that：in comparison with the conventional EVM method, although the processing speed of RGB based method has been improved, there are lots of defects in magnified video. However, in the Y channel based method, the processing speed is increased obviously and the magnification effect is consistent with that of the original EVM.

Key words: subtle motion, Eulerian Video Magnification（EVM）, RGB color space, YIQ color space, motion magnification

李乐鹏，雷林，孙水发，尹辉，董方敏. 视频微小运动放大的加速方法[J]. 计算机工程与应用, 2015, 51(24): 195-200.

LI Lepeng, LEI Lin, SUN Shuifa, YIN Hui, DONG Fangmin. Improved video small motion magnification processing[J]. Computer Engineering and Applications, 2015, 51(24): 195-200.

[1]	程欣，赵景秀. 基于颜色三角形信息量的彩色图像边缘检测[J]. 计算机工程与应用, 2014, 50(10): 152-155.
[2]	蒋海波，张云伟，郭志伟，张伟. 室外机器人视觉导航的路径结构识别[J]. 计算机工程与应用, 2010, 46(7): 245-248.