Ultrasparse View X-ray Computed Tomography for 4D Imaging

• DOI: 10.1021/acsami.3c06291
• Publication Date: 2023-07-13T05:15:35Z
• Authors: Yanjie Zheng; Kelsey B. Hatzell
• Keywords: Similarity (geometry)

X-ray computed tomography (CT) is a noninvasive, nondestructive approach to imaging materials, material systems, and engineered components in two three dimensions. Acquisition of three-dimensional (3D) images requires the collection hundreds or thousands through-thickness radiographic from different angles. Such 3D data acquisition strategies commonly involve suboptimal temporal sampling for situ operando studies (4D imaging). Herein, we introduce sparse-view approach, Tomo-NeRF, which capable reconstructing high-fidelity <10 two-dimensional images. Experimental 2D were used test reconstruction capability two-view, four-view, six-view scenarios. Tomo-NeRF with structural similarity 0.9971-0.9975 voxel-wise accuracy 81.83-89.59% experimentally obtained The less than synthetic structures. Experimentally demonstrate 0.9973-0.9984 84.31-95.77%.