首页 | 本学科首页   官方微博 | 高级检索  
     

基于刚体模型的视频插值去噪算法
引用本文:唐权华,雷金娥,周艳,金炜东.基于刚体模型的视频插值去噪算法[J].计算机工程与设计,2010,31(4).
作者姓名:唐权华  雷金娥  周艳  金炜东
作者单位:1. 江西师范大学软件学院,江西,南昌,330022;西南交通大学信息科学与技术学院,四川,成都,610031
2. 南昌工程学院计算机科学系,江西,南昌,330099
3. 西南交通大学电气工程学院,四川,成都,610031
摘    要:为充分利用视频的时空联系去噪,提出了视频的刚体模型.刚体模型以线性关系组织视频的像素,结合了像素间的时域和空域关联,可以通过主色调区域划分和边缘句法匹配获得视频的刚体分解及其对应关系.基于刚体模型的插值去噪利用像素值的空域邻近性和时域稳定性联合判断插值像素的参考价值.对常用测试视频的去噪实验表明了刚体模型的正确性,基于刚体模型的插值去噪方法在视觉效果和峰值信噪比上都表现出色.

关 键 词:视频处理  视频去噪  刚体模型  插值去噪  时空联合去噪

Algorithm of video interpolation denoising based on rigid model
TANG Quan-hua,LEI Jin-e,ZHOU Yan,JIN Wei-dong.Algorithm of video interpolation denoising based on rigid model[J].Computer Engineering and Design,2010,31(4).
Authors:TANG Quan-hua  LEI Jin-e  ZHOU Yan  JIN Wei-dong
Affiliation:TANG Quan-hua1,3,LEI Jin-e2,ZHOU Yan4,JIN Wei-dong4(1.School of Software,Jiangxi Normal University,Nanchang 330022,China,2.Department of Computer Science,Nanchang Institute of Technology,Nanchang 330099,3.School of Information Science , Technology,SouthWest Jiaotong University,Chengdu 610031,4.School of Electrical Engineering,China)
Abstract:A rigid model of video is proposed to denoise videos via temporal-spatial relations of video.The rigid model(RM) resolves videos into rigid block and organize the pixels with linear relations,combining the special and temporal relations.Rigid blocks and the correspondences is generated by color dividing and syntax matching of edges.In interpolation denoising based on RM the references is judged by pixels' neighborhood in space and stability in time domain.The validity of RM and Interpolation based on RM is ...
Keywords:video processing  video denoising  rigid model  interpolation denoising  spatial-temporal denoising
本文献已被 CNKI 万方数据 等数据库收录!
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