远程会诊中病理切片图象的拼接算法研究  

Stitching Algorithm for Pathologic Picture in Remote Consultation

作　　者：韩丽萍[1] 张乃珍[2] 陈礼民[3] 

机构地区：[1]山西大学电子系,山西太原030006 [2]山西省人民医院,山西太原030012 [3]山西大学计算中心,山西太原030006

出　　处：《小型微型计算机系统》2003年第12期2207-2209,共3页Journal of Chinese Computer Systems

摘　　要：针对现有图象配准方法存在的问题 ,提出了一种新的基于特征的图象对准算法 ,利用差值数据中最大包与次大包偏差的极大值来匹配相邻图象 ,研究结果表明该算法在获得良好拼接的基础上 ,大大提高了拼接速度 ,是一种快速实用的拼接算法 .In this paper, the problems existing in current image registration are firstly analyzed, then a new image registration algorithm based on feature is presented, in which overlapping images are stitched through deviation between maximum envelope and submaximum envelope of difference. The results show that the speed of mosaic is increased largely and the effect of mosaic is excellent with the algorithm, it is a rapid and applied algorithm of image mosaic.

关 键 词：远程会诊 图象拼接 差值数据 最大包 次大包 病理切片图象 图象对准 

分 类 号：R445[医药卫生—影像医学与核医学] TP391.41[医药卫生—诊断学]