首页 | 本学科首页   官方微博 | 高级检索  
     

Practical Algorithm for Real-Time Panorama Mosaic
作者姓名:彭启民  贾云得  柳阳
作者单位:School of Information Science and Technology, Beijing Institute of Technology, Beljing 100081, China
基金项目:国家科技攻关项目,中国科学院资助项目
摘    要:A real-time algorithm for cylindrical panorama mosaic from un-calibrated images is presented, A novel scheme based on feature matching is proposed to align the images, Feature candidates are extracted with Plessy's corner detector and then are matehed between eonsecutive frames based on FFT, Overlays between consecutive frames are detected with an improved two-step feature matching. Finally, the images are stitched by fusing in accordance with the found seam linked matched features. This method can be implemented efficiently and show robustness to varying illumination and perspective distortion.

关 键 词:实用算法  覆面  非校准图像  特征匹配
文章编号:1004-0579(2005)04-0365-04
收稿时间:2004-05-14

Practical Algorithm for Real-Time Panorama Mosaic
PENG Qi-min,JIA Yun-de and LIU Yang.Practical Algorithm for Real-Time Panorama Mosaic[J].Journal of Beijing Institute of Technology,2005,14(4):365-368.
Authors:PENG Qi-min  JIA Yun-de and LIU Yang
Affiliation:School of Information Science and Technology, Beijing Institute of Technology, Beijing 100081, China
Abstract:A real-time algorithm for cylindrical panorama mosaic from un-calibrated images is presented. A novel scheme based on feature matching is proposed to align the images. Feature candidates are extracted with Pl essy's corner detector and then are matched between consecutive frames based on FFT. Overlays between consecutive frames are detected with an improved two-step feature matching. Finally, the images are stitched by fusing in accordance with the found seam linked matched features. This method can be implemented efficiently and show robustness to varying illumination and perspective distortion.
Keywords:panorama  stitch  feature  match  overlay  real-time
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《北京理工大学学报(英文版)》浏览原始摘要信息
点击此处可从《北京理工大学学报(英文版)》下载全文
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