一种基于FDK的锥束CT重建的改进方法  

An Improved Technique for Cone-beam CT Based on FDK Method

作　　者：刘亮[1] 潘晋孝[1] 任变青[1] 

机构地区：[1]中北大学信息与通信工程学院,太原030051

出　　处：《光散射学报》2007年第1期91-95,共5页The Journal of Light Scattering

基　　金：国家自然科学基金资助项目(60532080)

摘　　要：在实际的锥束CT扫描过程中,由于探测器的离散化,对每个探测器单元来说都有一定大小.那么相对于被扫描物体的每个体素点,在某一个角度下过它的射线打在探测器上是有一定面积的,如果取体素截面大小和探测器单元大小相近,过它的射线是分布在相邻的几个探测器单元上的,在重建过程中,如果依据落在各个探测器单元上的射线所占面积的大小考虑到那些探测器单元的信息对被重建体素点的影响,会提高重建图像的边缘清晰度,本论文利用实际的锥束扫描数据,采用传统的FDK方法和改进的FDK方法对物体重建,通过比较得出用新方法重建的图像边缘比传统FDK方法清晰,而且新方法还有抑制噪声的作用。In practical circular cone - beam CT scanning process, because of the dispersing of the detector, the detector cell isn＇t non - area, for each pixel in the scanned object, it will taking some area on the detector in a certain angel, if the pixel＇s sectional area is about as big as the detector cell, the x - radial through the pixel will taking more than one border upon detector cells, if the area of the x - radial taking on each cell be concerned in the back projection process, the reconstructed image＇s quality will be improved in this article. We reconstructed the object with practical date in FDK method and proved FDK method, the quality of the image which taking the new method is better than the traditional one, and the new method also will restrain the yawp in a certain extent.

关 键 词：锥束重建 探测器面积 FDK算法 

分 类 号：TN911.73[电子电信—通信与信息系统]