Please wait a minute...
浙江大学学报(工学版)
J4  2009, Vol. 43 Issue (7): 1182-1186    DOI: 10.3785/j.issn.1008-973X.2009.
自动化技术、计算机技术     
基于特征的显微图像全自动拼接
范翔, 夏顺仁
（浙江大学 生物医学工程系 教育部国家重点实验室, 浙江 杭州 310027）
Feature based automatic stitching of microscopic images
FAN Xiang, XIA Shun-ren
(Key Laboratory of Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, Zhejiang University, Hangzhou 310027, China)
 全文: PDF(1058 KB)   HTML
摘要：

为了解决高倍镜下显微镜视野减小,无法完全捕获目标的问题,需要设计快速高效的方法对一系列显微图像进行拼接.在照片全景图重建的研究基础上,提出了一种基于特征的全自动显微图像拼接算法.该算法利用尺度不变特征变换（SIFT）提取图像中的特征,将获得的全部特征构建全局kd-Tree,使用优化的最优节点优先（BBF）算法搜索潜在的匹配图像对.采用随机抽样一致性算法（RANSAC）对找到的匹配图像对进行检验.根据最小生成树（MST）算法获得图像序列的连通分量,得到图像对之间的变换矩阵并将图像映射到拼接平面.对一系列显微图像的实验结果表明,该方法对图像中的背景噪声和亮度差异都有较好的鲁棒性,对相互之间只有少量重叠区域的图像序列也能获得可靠和精确的结果.
Abstract:

To solve the problem of  being unable to capture the whole specimen when the field of the view of the microscope is decreased using highly magnifying lens, it is necessary to mosaic series of microscopic images effectively and efficiently. Based on the related research on panorama reconstruction for photography, a feature based automatic mosaicing method was presented. The scale invariant feature transform (SIFT) was applied to extract features from the images, and by the optimized implementation of the best-bin-first (BBF) algorithm, the global kd-Tree containing all the features was constructed to search for the possible overlapping image pairs. The matches were verified by random sample consensus (RANSAC). The minimum spanning tree (MST) was used to obtain the best connected-component of the image set to recover the transformation between images and project the images into the mosaic frame. Experimental results with series of microscopic images showed that the proposed approach was robust to background noises and illumination changes in the images and it could provide reliable and accurate results even for images of low overlapping or with relatively few features．
出版日期: 2009-07-01
:  TP391.41  
基金资助:

国家安全重大基础研究资助项目(5132103ZZT14B)；国家自然科学基金资助项目(60772092).
通讯作者: 夏顺仁,男,教授.     E-mail: srxia@zju.edu.cn
作者简介: 范翔（1984-）,男,江苏苏州人,硕士生,从事医学图像处理方面的研究工作.
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　

引用本文:

范翔, 夏顺仁. 基于特征的显微图像全自动拼接[J]. J4, 2009, 43(7): 1182-1186.

FAN Xiang, XIA Shun-ren. Feature based automatic stitching of microscopic images. J4, 2009, 43(7): 1182-1186.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2009.    或    http://www.zjujournals.com/eng/CN/Y2009/V43/I7/1182

［1］ PIZARRO O, SINGH H. Towards large area mosaicing for underwater scientific applications ［J］. IEEE Journal of Oceanic Engineering: Special Issue on Underwater Image and Video Processing, 2003, 28(4): 651672．
［2］ BROWN M, LOWE D. Automatic panoramic image stitching using invariant features［J］. International Journal of Computer Vision, 2006, 74(1): 5973．
［3］ BROWN M, LOWE D. Recognizing panoramas［C］ ∥ Proceedings of the 9th International Conference on Computer Vision. Beijing: ［s.n.］, 2003: 12181225．
［4］ BROWN M, SZELISKI R, WINDER S. Multi-image matching using multi-scale oriented patches［C］ ∥ International Conference on Computer Vision and Pattern Recognition. ［S.l.］: IEEE, 2005: 510517．
［5］ FAUQUEUR J, KINGSBURY N, ANDERSON R. Multi-scale keypoint detection using the dual-tree complex wavelet transform［C］ ∥ IEEE International Conference on Image Processing. ［S.l.］: IEEE, 2006: 16251628．
［6］ LOWE D. Distinctive image features from scale-invariant keypoints［J］. International Journal of Computer Vision, 2004, 60(2): 91110．
［7］ BEIS J, LOWE D. Shape indexing using approximate nearest-neighbor search in high-dimensional spaces［C］ ∥ Proceedings of the International Conference on Computer Vision and Pattern Recognition. San Juan, Puerto Rico: IEEE, 1997: 10001006．
［8］ NOWOZIN S. Libsift - scale-invariant feature fransform implementation［EB/OL］. ［2006-12-21］. http:∥user.cs.tu-berlin.de/～nowozin/libsift/
［9］ FISCHLER M, BOLLES R. Random sample consensus: a paradigm for model fitting with application to image analysis and automated cartography［J］. Communications of the ACM, 1981, 24: 381395．
［10］ TRIGGS B. Bundle adjustment - a modern synthesis［C］ ∥ International Workshop on Vision Algorithms. ［S.l.］: ACM, 1999: 298372．
［11］ MATTHEW B. Autostitch : a new dimension in automatic image stitching［EB/OL］. ［2006-12-12］. http:∥www.cs.ubc.ca/～mbrown/autostitch/autostitch.html
[1] 王宣银, 梁冬泰. 基于多元图像分析的表面缺陷检测算法[J]. J4, 2010, 44(3): 448-452.
[2] 孙志海, 孔万增, 朱善安. 视频目标定位的减法聚类改进算法[J]. J4, 2010, 44(3): 458-462.
[3] 张冬梅, 刘利刚. 基于角度滤波的平面图形光顺算法[J]. J4, 2009, 43(6): 1042-1046.
[4] 陈成, 庄越挺, 肖俊. 相机运动条件下的视频前景提取[J]. J4, 2009, 43(6): 975-977.