首页 | 本学科首页   官方微博 | 高级检索  
     

显微图像拼接的误差分析
引用本文:苗立刚,岳永娟,彭思龙.显微图像拼接的误差分析[J].小型微型计算机系统,2007,28(7):1255-1258.
作者姓名:苗立刚  岳永娟  彭思龙
作者单位:中国科学院,自动化研究所,国家专用集成电路设计工程技术研究中心,北京,100080
摘    要:针对显微图像拼接中的误匹配和误差积累问题,把图像对准误差划分为第一类误差和第二类误差,并提出了一种新的图像拼接方法.首先利用所有重叠图像对的局部对准约束建立全局对准模型,它可以消除第一类误差引起的误差积累;然后根据全局对准误差的分布特性,提出消除第二类误差的最小回路一致性方法.实验表明,该图像拼接方法计算简单、有效,适用于大规模的显微图像拼接.

关 键 词:图像拼接  图像对准  最小回路一致性  误差积累  图像误匹配
文章编号:1000-1220(2007)07-1255-04
修稿时间:2006-04-042006-09-18

Error Analysis of Microscopic Image Mosaicing
MIAO Li-gang,YUE Yong-juan,PENG Si-long.Error Analysis of Microscopic Image Mosaicing[J].Mini-micro Systems,2007,28(7):1255-1258.
Authors:MIAO Li-gang  YUE Yong-juan  PENG Si-long
Affiliation:National ASIC Design Engineering Center, Institute of Automation, Chinese Academy of Sciences, Beijing 100080, China
Abstract:Image alignment error is classified as type I error and type II error in this paper for the error accumulation and image mismatch of microscopic image mosaicing,and a global image alignment model of image mosaicing is proposed to deal with these two types of error.Firstly local alignment constraints of all overlapping images are used to construct global alignment model,thus,to eliminate error accumulation induced by type I error effectively.Then minimum cycle consistency method is proposed to automatically eliminate type II error according to the characteristics of global alignment error distribution.Experimental results show that this method is simple and robust,and it can be used in constructing mosaics of large-scale microscopic images.
Keywords:image mosaicing  image alignment  minimum cycle consistency  error accumulation  image mismatch
本文献已被 CNKI 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