基于梯度互信息的CT/PET图像分步配准  

Multistage CT/PET Medical Image Registration Based on Gradient and Mutual Information

作　　者：吴小丹[1] 奚日辉[1] 宋玲玲[1] 

机构地区：[1]上海大学通信与信息工程学院,200072

出　　处：《微计算机信息》2008年第9期292-293,247,共3页Control & Automation

摘　　要：研究了基于互信息的医学图像配准方法,针对其缺乏空间信息而容易导致误配准以及计算量大,耗时长的问题,提出了一种梯度和互信息相结合的分步配准方法,应用于CT和PET图像的配准。该方法利用形态学运算去除PET图像背景伪迹,运用对比度拉伸和直方图均衡化对其增强,获取结构清晰、完整的梯度图像。利用矩和主轴法对梯度图像进行初步配准,对初步配准后的梯度图像进行互信息精配准。实验表明,该方法用于CT和PET图像的配准,在精度和速度上都明显优于传统互信息方法。Medical image registration based on mutual information is researched in this paper. As the absence of space information, it is easy to get into local maximum and results in error registration, meanwhile, it needs long time as the computed quantity of mutual information is large. A multi-stage registration approach based on mutual information (MI) and gradient information is proposed in this paper and applied in the registration of CT/PET. Morphology method is used to reduce noise of the PET image, contrast-...

关 键 词：医学图像配准 互信息 梯度 矩和主轴 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]