第2期
第18卷 2010年2月
文章编号
1004-924X(2010)02-0470-07
光学精密工程
Optics and Precision Engineering
图像自动拼接融合的优化设计
冯宇平1,2，戴明1，孙立悦3，张威1,2
（1.中国科学院长春光学精密机械与物理研究所，吉林长春130033；
Vol.18No. 2
Feb.2010
2.中国科学院研究生院，北京100039;3.吉林大学通信工程学院，吉林长春130012）
摘要：为了扩大摄像机拍摄图片的视场范围，提出了--种结合额域和空域进行图像序列自动拼接融合的优化算法。基于颗域相位相关方法改进排序并确定重叠区域，使用改进的Haris算子在空间域提取图像角点（无需人工设定阔值），并通过双向最大互相关系数匹配获得初始特征点对；然后，用RANSAC法实现精确匹配；最后，利用图像均值结合线性加权函数实现图像的整体亮度调整和融合拼接，实验结果表明，该优化算法排序过程简单有效，特征提取匹配过程不仅提高了拼接成功率，其单次拼接时间较现有算法也提高了40%左右；另外，该算法较好地解决了图像间对应点难以确定的
问题，对光照变化图像的拼接融合取得了满意的效果。关键调：图像接；自动排序，角点检测；图像融合
中图分类号：TP391
文献标识码：A
Optimizeddesignofautomaticimagemosaic FENG Yu-ping"-2,DAI Ming',SUN Li-yue”,ZHANG Weil-(1.Changchun Instituteof Optics,FineMechanics andPhysics, ChineseAcademyof Sciences，Changchun130033,China;
2.GraduateUniversity of ChineseAcademyofSciences,Beijing100039,China 3.SchoolofCommunicationEngineering,JilinUniversity,Changchun130012,China
Abstract; In order to improve the visual field of photos, an optimized algorithm of automatic image mosaic is presented based on frequency and time domains. The frequency phase correlation is adopted to sort the unorder image sequence and to estimate the overlapping area, and an improved Harris oper ator is adopted to extract corners. Then, the bidirectional greatest correlative coefficient is used to ob tain the initial feature point pairs, and the false feature point pairs are rejected by RANSAC. Finally, the image mean combined with the linear weight function is used to implement image mosaic. Experi-ment results show that this algorithm can offer a excellent sorting process and show good nomosaic re-sults. The maching speed by proposed algorithm has improved by 4o% as compared with those of tra-ditional methods,which efficiently solves the difficulty in confirming corresponding points and achives
the desired visual effect in mosaic images with notable illumination difference. Key words: image mosaic; auto-sorting; corner detection; image fusion
收稿日期：2009-01-06：修订日期：2009-03-09.
基金项目：国防科技预研基金资助项目（No.1040603）