基于平行PI线段的平行束CT重建

Parallel PI-line segment-based parallel beam CT reconstruction

为了研究基于PI线段的反投影滤波重建方法在CT(computed tomography)系统中的应用,提出了一种基于平行PI线段的简化的平行射束重建算法。该算法首先对平行X射线束产生的投影数据进行微分,然后把这些微分后的数据反投影到一系列的平行PI线段上,最后沿着PI线段的方向对反投影数据进行Hilbert滤波,即得到重建图像。利用Shepp-Logan模型进行计算机数值模拟实验,得到了该算法和传统的滤波反投影算法的对比结果。利用微焦点X射线CT的投影数据进行重建,得到了实际数据下的重建图像。实验结果表明,该简化算法能够进行精确重建,可以应用于实际的CT系统。

A simplified algorithm for parallel-beam computed tomography （CT） reconstruction was developed using a PI-line segment-based backprojection algorithm in CT system. The algorithm first differentates the parallel beam projection data which is then backprojected onto a series of parallel PI-line segments. The data on the PI-line segments are then filtered using a Hilbert transform to get the reconstructed image. Numerical simulations using the Shepp-Logan model demonstrates that the algorithm can accurately reconstruct images. Analysis of real images from a micro-focus X-ray CT system also illustrates the systems effectiveness.