改进的联合代数重建法及其有限角投影重建

Modified Simultaneous Algebraic Reconstruction Technique and Its Application in CT from Limited Range

提出了一种基于联合代数重建法的改进算法 (MSART)。该算法将穿过像素的射线长度和灰度值综合考虑进误差分配公式 ,有效地抑制了联合迭代重建法 (SART)的边缘效应 ,使得重建精度大大提高。即使在少数投影有限角图像重建中 ,通过选择合适的投影方向 ,也能达到最大误差不大于 5 % ,平均误差不大于 1%的重建效果。

A modified simultaneous algebraic reconstruction technique (MSART) is presented, which takes into account both the density of pixel and the ray length across it in the error distribution. MSART can solve the edge distortion of SART effectively and makes an encouraging improvement in reconstructed image quality with sparse data. Even in the reconstruction of fewer views from limited range, good images can be obtained by choosing proper views. That is to say, the maximum error can be less than 5% and the average error can be less than 1%.