• Improved Cohen-Sutherland algorithm for TGS transmission imaging

    分类: 核科学技术 >> 核科学技术其他学科 提交时间: 2023-05-31

    摘要: Laminar -scanning (TGS), an advanced -ray nondestructive analysis technique, is capable of locating and analyzing the nuclides in radioactive nuclear waste, and the scanning of TGS falls into two parts (e.g., transmission measurement and emission measurement). To be specific, transmission measurement lays the basis for the accurate measurement of non-uniform radionuclide content in TGS scanning. The scan data were obtained using the Monte Carlo tool Geant4 simulation, and a total of 25 voxels were divided into 5 in length and 5 in width in a square barrel. In this study, the encoding cropping algorithm based on the draped foot vector judgment was adopted to rapidly calculate the voxel trace matrix within the square bucket of nuclear waste, and the transmission images were reconstructed using the Ordered Subsets Expectation-Maximization(OSEM). The results indicated that the cropping speed of the improved coding algorithm was significantly increased compared with the original algorithm, and the relative mean deviation (RMD) and root mean square error(RMSE) between the reconstructed attenuation coefficient and the reference standard value tended to decrease with the increase of the cropped line segments in the voxel, and the Pearson correlation coefficient(PCC) tended to converge to be 1.0. The image quality evaluation parameters of high media density materials were better than those of low media density materials in the above three indexes. The reconstruction effect was relatively poor under the more complex filling material. When there were more than 10 cropped line segments in the voxel, the reconstruction data generally tended to be stable. The graphical trimming algorithm is capable of rapidly calculating the trace matrix of the scanned voxels, it shows the advantages of speed and efficiency, and it can serve as a novel method to solve the trace matrix of TGS nuclear waste transmission scans.