首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于帧间单映变换的视频序列超分辨率重建
引用本文:韩玉兵,束锋,吴乐南.基于帧间单映变换的视频序列超分辨率重建[J].应用科学学报,2007,25(3):257-262.
作者姓名:韩玉兵  束锋  吴乐南
作者单位:1.南京理工大学 电子工程与光电技术学院,江苏 南京 210094;2.东南大学 无线电工程系,江苏 南京 210096
基金项目:江苏省图像处理与图像通信重点实验室开放研究课题
摘    要:针对帧间单映变换的视频序列提出了一种超分辨率重建算法。首先在最小二乘的基础上给出超分辨率重建的数学模型;其次采用基于模型的等级估计法估计相邻帧的单映变换矩阵;然后采用基于RANSAC的特征点匹配方法对非相邻帧进行带指导的改进估计,使其能够达到要求的亚像素级精度;最后运用共扼梯度算法进行迭代重建,并详细讨论了主要的矩阵向量乘法计算。实验结果表明该算法的重建相当有效。

关 键 词:视频处理  超分辨率重建  单映变换  随机抽样一致性算法  共轭梯度算法  
文章编号:0255-8297(2007)03-0257-06
收稿时间:2006-09-11
修稿时间:2006-09-112007-01-15

Super Resolution Reconstruction of Video Sequence Based on Inter-frame Homography Transform
HAN Yu-bing,SHU Feng,WU Le-nan.Super Resolution Reconstruction of Video Sequence Based on Inter-frame Homography Transform[J].Journal of Applied Sciences,2007,25(3):257-262.
Authors:HAN Yu-bing  SHU Feng  WU Le-nan
Institution:1. School of Electronic Engineering & Optoelectronic Techniques, Nanjing University of Science & Technology, Nanjing, 210094, China;2. Department of Radio Engineering, Southeast University, Nanjing, 210096, China
Abstract:A super resolution(SR) reconstruction algorithm of video sequence based on inter-frame homography transform is presented in this paper.The mathematical model of SR is introduced using the least square principle.Hierarchical model-based motion estimation is adopted to compute the adjacent homography matrix.For non-adjacent frames,the homography matrix is estimated with guided feature registration based on RANSAC algorithm,which gives sub-pixel accuracy.Finally,a conjugate gradient algorithm is used to reconstruct the SR frames,and computation of matrix-vector multiplication is studied in detail.Experimental results demonstrate effectiveness of the proposed method.
Keywords:video processing  super resolution reconstruction  homography transform  RANSAC algorithm  CG algorithm
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《应用科学学报》浏览原始摘要信息
点击此处可从《应用科学学报》下载免费的PDF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