CT Reconstruction from Limited-View Cone-Beam Projections Using 3D Image Moments

Limiting the views of cone-beam CT scan can effectively reduce the damage to the human's health brought by X-rays when taking CT examination. However, the CT images reconstructed from the limited-view cone-beam projections will degrade greatly. To improve the quality of the reconstructed images when the cone-beam projections are incomplete, this paper proposes a new method by using 3D image moments for limited-view cone-beam CT reconstruction. It uses 3D geometric moments to establish the relationship between 3D Radon data and image moments. After converting the cone-beam projection acquired in limited views into known 3D Radon data, we use this relationship to estimate unknown 3D Radon data from known ones. In this way, 3D Radon data become complete. Finally, with help of the inverse formula of the first derivative of the 3D Radon transform, we reconstruct the 3D CT images from the complete 3D Radon data. The experimental results show that our method has good performance in limited-view CT reconstruction. The experimental results show that our method can improve the reconstructed cone-beam CT images when only limited-view projections are given.