基于分割先验图像多能CT重建算法

M ulti-energy CT reconstruction algorithm based on segmentation prior image

固定电压CT成像系统动态范围有限,照射复杂工件时产生的投影极易出现欠曝光、过曝光现象,从而导致有效投影的缺失。为解决投影不完备问题研究了基于分割先验图像多能CT重建算法。该方法首先采用ART-TV算法对最低能量的投影进行重建,并对重建结果进行分割,截取与高能重建图像相似部分做为先验图像,以先验图像约束的压缩感知算法对邻近高能投影重建,重建结果再次进行分割后截取先验图像,重复上述步骤直至最高能量完成重建。仿真实验表明:和传统的多能重建方法相比,研究的方法在完整重建复杂结构工件基础上有效地减少伪影,提高图像像素的稳定性,在主观视觉感受和客观评价参数均有一定的进步。

In CT reconstruction system based on fixed voltage,the projection data often appears overexposed or underexposed because of the limited dynamic range when the complex workpiece is rayed.As a result,it leads to a lack of useful projection.In order to solve the problem of under-sampled projection,the multi-energy CT reconstruction algorithm based on segmentation prior image is studied.Firstly,ART-TV is used to reconstruct the lowest energy projection,and the reconstruction result is segmented.The compelling part is regarded as the initial image,and utilizing the prior image constrained the compressed sensing algorithm to reconstruct the next higher energy projection.The result is segmented again,and the prior image is intercepted.Repeat the above steps until the highest energy.The simulation results show that compared with the traditional multi-energy reconstruction method,the method proposed in this paper can effectively reduce artefacts and improve the stability of image pixels on the basis of completely reconstructing workpiece with complex structure,and has made some progress in the subjective visual perception and objective evaluation parameters.