一种基于Radon逆变换的半覆盖螺旋锥束CT重建算法

半覆盖螺旋锥束计算机断层成像能够扩展传统螺旋锥束计算机断层成像的成像视野, 实现小面板探测器成像超视野物体. 但是, 各角度下的投影都存在数据截断, 会使重建结果中产生截断伪影, 降低图像质量. 本文提出了一种基于Radon逆变换的半覆盖螺旋锥束重建算法, 该算法在滤波时先使用局部算子, 再使用全局算子. 局部算子不受数据截断的影响, 并且降低了运算后的值由于数据截断所造成的不连续性, 因此, 减少了随后全局运算产生的截断误差. 仿真和实际实验结果均验证了本文算法的有效性. 和现有算法的对比也表明, 本文算法针对半覆盖螺旋锥束投影具有更强的截断伪影抑制能力, 能够有效提高重建图像的质量.

A half-covered helical cone-beam reconstruction algorithm based on the Radon inversion transformation

Compared with the traditional helical cone-beam computed tomography (CT), the field-of-view (FOV) half-covered cone-beam CT can almost double the FOV and thus image the large object by using a smaller panel detector. However, the projections are transversely truncated, resulting in truncation errors in reconstructed images if no correction measures are taken. In this paper, a half-covered cone-beam reconstruction algorithm based on the Radon inversion transformation is developed, in which the data filtering is performed in two steps. The first step is a local operation and can be carried out correctly even when the data is truncated. This performance of local operation makes the original data closer to zero, so the continuity of data is improved. And this also can restrain the truncation errors caused by the following global operation. Numerical simulations and experimental results are presented to demonstrate the algorithm and to compare it with existing algorithms. Preliminary results indicate that the proposed algorithm can well restrain the truncation errors and improves reconstruction quality.