CT与MR医学图像的三种融合方法对比研究  

Contrastive Study on Three Fusion Methods of CT and MR Medical Image

作　　者：罗火灵[1] 许永忠[1] 冉洋[2] 陈世仲[1] 

机构地区：[1]中国矿业大学资源学院生物医学系,徐州221008 [2]安徽省中医院,合肥230031

出　　处：《中国医疗器械信息》2009年第1期56-58,67,共4页China Medical Device Information

摘　　要：医学图像融合技术作为图像处理主要的技术已逐渐成为图像处理研究的热点,它的研究将会对未来医学影像技术进步带来深远的影响。本文对目前比较成熟和应用广泛的三种典型的图像融合方法进行研究和分析,包括像素加权平均融合法、插入像素融合法和像素灰度值选大选小融合法。其中像素加权平均融合法在医学图像融合中有着非常重要的地位,融合的范围比较广泛。插入像素法则是建立在源图像达到完全配准的基础之上,适用范围相对较窄。像素灰度值选大融合方法在某些应用场合中不仅可以获得较好的融合效果,而且其实现也很方便、速度也快。As the major image processing technology, medical image fusion has gradually become the focus of study on image processing, which will exert a profound influence upon the advancement of medical image technology in the future. The paper has studied and analyzed three typical methods of image fusion, which are by far the comparatively mature and widely applied ones, including pixel weighted average method, insert pixel method, and plus-or-minus grey scale values of pixel. Among these methods, pixel weighted average method plays a more significant role in medical image fusion, and therefore has a wider range of fusion. The insert pixel method is based on the complete registration of the original image, and therefore has a narrower range of application. The method of plus grey scale values of pixel can not only achieve a better fusion effect in some situations of application, but is also convenient and fast to be done.

关 键 词：医学图像 图像融合 CT MR 

分 类 号：TP392[自动化与计算机技术—计算机应用技术]