结合稳像补帧与VIBE算法的抖动视频前景目标提取方法

doi:10.11896/jsjkx.210800195

计算机科学 ›› 2022, Vol. 49 ›› Issue (11A): 210800195-8.doi: 10.11896/jsjkx.210800195

• 图像处理&多媒体技术 • 上一篇下一篇

结合稳像补帧与VIBE算法的抖动视频前景目标提取方法

刘新富, 蒋慕蓉, 黄亚群, 张占伟, 周莹瑛

云南大学信息学院昆明 650091

出版日期:2022-11-10 发布日期:2022-11-21
通讯作者: 蒋慕蓉(jiangmr@ynu.edu.cn)
作者简介:(1101097590@qq.com)
基金资助:
云南省高校科技创新团队支持项目(IRTSTYN19N07)

Using Image Stabilization and VIBE Algorithm to Extract Foreground Target from Jitter Video

LIU Xin-fu, JIANG Mu-rong, HUANG Ya-qun, ZHANG Zhan-wei, ZHOU Ying-ying

School of Information Science & Engineering,Yunnan University,Kunming 650091,China

Online:2022-11-10 Published:2022-11-21
About author:LIU Xin-fu,born in 1995,postgraduate.His main research interests include image proces-sing,artificial intelligence and machine learning.
JIANG Mu-rong,born in 1963,professor.Her main research interests include mathematical method of image proces-sing and its intelligent calculation.
Supported by:
University Science and Technology Innovation Team Support Project of Yunnan Province,China(IRTSTYN19N07).

摘要/Abstract

摘要： 在动态场景下进行目标检测与跟踪处理时,提取的前景目标往往会因受到背景抖动干扰而出现目标误判、轮廓不完整等问题,为此提出结合稳像与补帧、VIBE算法与连通域修正的抖动视频前景目标提取方法。首先利用ORB对相邻两帧图像进行特征匹配,使用RANSAC剔除误匹配点,以运动目标为中心调整图像边距,裁剪与修补不重合的部分,通过统计法求相邻两帧的偏移均值,完成视频稳像处理;其次利用稠密光流法计算相邻两帧图像的位移,通过位置重映射生成中间帧,实现视频补帧;然后将VIBE算法增加形态学处理与连通域修正,结合Canny算子进行边缘检测,增加目标轮廓的完整性;最后使用视频实例进行测试,与其他视频目标提取算法进行比较分析,精度、召回率至少提高10%,漏检率、错分率至少降低15%。实验结果表明,所提方法能有效去除抖动带来的影响,提取的目标完整性较高。

关键词: 目标提取, 抖动视频, 稳像, 补帧, VIBE算法, 连通域修正

Abstract: In target detection and tracking processing area,the extracted foreground target is often disturbed by background jittering,resulting the problems of target misjudgment,incomplete contour and so on.In this paper,we present a method to improve the integrity of foreground target extraction by combining image stabilization and frame supplement,and then using VIBE algorithm and connected domain correction.Firstly,the image of two adjacent frames is matched by ORB feature,and the wrong ma-tching points are eliminated by RANSAC method.Then,the image margin is adjusted with the moving target as the center,and the non-coincident parts are cropped and repaired.The offset mean value of two adjacent frames is calculated by statistical method to achieve the purpose of video image stabilization.Secondly,the dense optical flow method is used to calculate the displacement of two adjacent frames,and then the intermediate frame is generated by position remapping and supplemented to make the video sequence more smooth and stable.Thirdly,based on VIBE algorithm,morphological processing and connected domain correction are added,and Canny operator is combined for edge detection to increase the integrity of target contour.Finally,a video example is used to test and compare with the commonly used video target extraction algorithms.The precision and recall rates increase by at least 10%.The rates of FNR and PWC reduce by at least 15%.Experimental results show that the proposed method can effectively remove the influence of jitter in most dynamic backgrounds,and the integrity of the extracted target is high.

Key words: Target extraction, Jitter video, Image stabilization, Frame compensation, VIBE algorithm, Connected domain modification

中图分类号:

TP391

刘新富, 蒋慕蓉, 黄亚群, 张占伟, 周莹瑛. 结合稳像补帧与VIBE算法的抖动视频前景目标提取方法[J]. 计算机科学, 2022, 49(11A): 210800195-8. https://doi.org/10.11896/jsjkx.210800195

LIU Xin-fu, JIANG Mu-rong, HUANG Ya-qun, ZHANG Zhan-wei, ZHOU Ying-ying. Using Image Stabilization and VIBE Algorithm to Extract Foreground Target from Jitter Video[J]. Computer Science, 2022, 49(11A): 210800195-8. https://doi.org/10.11896/jsjkx.210800195

参考文献

[1]XIN Y X,SHI P F,XUE RY.Moving Object Detection Based on Region Extraction and Improved LBP Features[J].Computer Science,2021,48(7):233-237.
[2]RAJ R,RAJIV P,KUMAR P,et.al.Feature based video stabilization based on boosted HAAR Cascade and representative point matching algorithm[J].Image and Vision Computing,2020,101(1):1-8.
[3]WU H C,XIAO L,WEI Z H.Simultaneous Video Stabilization and Rolling Shutter Removal[J].IEEE transactions on image processing:a publication of the IEEE Signal Processing Society,2021,30:4637-4652.
[4]LIN S S,LE T N H,WU P Y,et al.Content-and-disparity-aware stereoscopic video stabilization[J].Multimedia Tools and Applications,2020,80(1):1545-1564.
[5]MOHAMMED A,FARUP I,YILDIRIM S,et al.Variational approach for capsule video frame interpolation[J].EURASIP Journal on Image and Video Processing,2018,30(1):1-13.
[6]ZHANG Q,JIANG F.Video interpolation algorithm based on deep learning[J].Intelligent Computer and Applications,2019,9(4):252-257,262.
[7]LIN C J,DENG W,TONG T,et al.Fuzzy video frame interpolation method based on deep voxel stream[J].Journal of Compu-ter Applications,2020,40(3):819-824.
[8]WANG T,WANG W,CUI Y H,et al.Improved ViBe algo-rithm based on color layout descriptor[J].Journal of Computer Applications,2020,40(3):812-818.
[9]HE L F,LIU Y L,ZHONG Y,et al.Video foreground targetextraction algorithm under complex background[J].Laser & Optoelectronics Progress,2020,57(16):160-166.
[10]JIN M,ZHANG J,HUANG T F,et al.Parameter adaptive de-shading foreground target detection algorithm[J].Journal of Huazhong University of Science and Technology(Natural Science Edition),2021,49(1):73-79.
[11]BARNICH O,VAN DROOGENBROECK M.ViBe:a powerful random technique to estimate the background in video sequences[C]//2009 IEEE International Conference on Acoustics,Speech and Signal Processing.IEEE,2009:945-948.
[12]XU H Y,ZHANG B,YANG S.An infrared image registration algorithm combined with FAST algorithm[J].Journal of Shen-yang Institute of Aeronautical Engineering,2017,34(5):57-62.
[13]YANG X Y,CHEN B.Feature matching optimization algorithm based on ORB detection[J].Journal of Computer Applications,2019,39(S2):81-84.
[14]DAI W H,LIU S C,ZHAO S,et al.Improved random sampling consensus algorithm using local pixel matching[J].Journal of National University of Defense Technology,2021,43(4):38-43.
[15]ZHANG L.Design and implementation of facial expression re-cognition system based on sparse optical flow method and HMM[D].Xiamen:Xiamen University,2019.
[16]LIN C Z,ZHANG W,WANG X,et al.Smoke detection algo-rithm based on dense optical flow and edge features[J].Computer Engineering and Science,2018,40(7):1213-1220.
[17]WANG X R,WANG X G.OpenCV and Visual Studio image recognition application development[M].Beijing:Posts& Telecom Press,2017:124-127.
[18]QIN X P.MATLAB image processing and interface programming[M].Beijing:Publishing House of Electronics Industry,2009:310-327,445-447.
[19]ZHOU Z H.Machine learning[M].Beijing:Tsinghua University Press,2016:30-32.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

结合稳像补帧与VIBE算法的抖动视频前景目标提取方法

Using Image Stabilization and VIBE Algorithm to Extract Foreground Target from Jitter Video

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

Metrics

本文评价

推荐阅读 0

[1]	张文雅,徐华中,罗杰. 基于ViBe的复杂背景下的运动目标检测 Moving Objects Detection under Complex Background Based on ViBe 计算机科学, 2017, 44(9): 304-307. https://doi.org/10.11896/j.issn.1002-137X.2017.09.057
[2]	易盟,楚岩. 基于仿射不变约束与快速EKF滤波的航拍图像稳像 Aerial Video Image Stabilization Based on Affine Invariant Constraint and Fast EKF Adaptive Filter 计算机科学, 2016, 43(8): 313-317. https://doi.org/10.11896/j.issn.1002-137X.2016.08.064
[3]	姬莉霞,李学相. 基于相邻帧补偿的高速运动目标图像稳像算法及仿真 Algorithm and Simulation of Image Stabilization for High Speed Moving Target Images Based on Adjacent Frames Compensation 计算机科学, 2014, 41(7): 310-312. https://doi.org/10.11896/j.issn.1002-137X.2014.07.064
[4]	李鹏杰杨树元. 运动目标自动特征提取与跟踪计算机科学, 2002, 29(z2): 174-178.