首页 | 本学科首页   官方微博 | 高级检索  
     

大场景的视频全景图拼接
引用本文:徐则中,庄燕滨.大场景的视频全景图拼接[J].计算机与数字工程,2010,38(11):141-144.
作者姓名:徐则中  庄燕滨
作者单位:[1]常州工学院计算机信息工程学院,常州213002 [2]常州市软件技术研究与应用重点实验室,常州213002
基金项目:江苏高校"青蓝工程",江苏教育自然科学基金
摘    要:为了解决大场景全景图拼接过程中的误差累积问题,提高图像拼接的一致性,提出了一个新的全景图拼接算法。把全局图像配准看作是一个统计估计问题,运用增广的卡尔曼滤波对图像的全局变换参数进行递归地增广和估计。实验结果表明,得到的变换参数是全局一致的,生成的全景图不存在拼接裂缝。拼接算法考虑了各种不确定性,适用于噪声大场景的全景图拼接。

关 键 词:全景图  图像拼接  图像配准  图形融合

Video Panoramic Mosaicing for Large-scale Scene
Xu Zezhong,Zhuang Yanbin.Video Panoramic Mosaicing for Large-scale Scene[J].Computer and Digital Engineering,2010,38(11):141-144.
Authors:Xu Zezhong  Zhuang Yanbin
Affiliation:),)(Dept.of Computer Information and Engineering,Changzhou Institute of Technology1),Changzhou 213002)(Changzhou Key Laboratory of Software Technology and Application2),Changzhou 213002)
Abstract:In order to deal with the problem of registration error accumulation and to improve the global consistency of image mosaics,a method of video panoramic mosaicing is proposed.The problem of global image alignment is considered as a stochastic estimation problem.The global transformation parameters of video images are augmented and estimated recursively with augmented Kalman filter.Experimental results show that the transformation parameters are globally consistent and the panoramic mosaic is free of mosaicing gap.The proposed video panoramic mosaicing method takes account of various uncertainties and is suitable for large-scale noisy scene.
Keywords:panorama  image mosaicing  image registration  image blending
本文献已被 维普 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