Comparative Analysis and Improvement of CT Scanning 3D Reconstruction Methods for Coal Samples

To assess the accuracy of commonly used 3D reconstruction techniques for coal samples and provide a foundation for examining the coal’s internal microstructure, as well as its mechanical and seepage properties, this study focuses on samples from the Yuwu and Yuecheng Mines. X-ray CT scanning was employed to acquire CT slices of the coal samples, which were then subjected to 3D reconstruction using Avizo, Mimics, and Matlab. By comparing and analyzing the strengths and limitations of each reconstruction method in terms of image processing quality and reconstruction fidelity, the most effective method was identified. This selected method was further refined using layer interpolation techniques, and its validity was confirmed with mercury intrusion experimental data. The results indicate that the 3D reconstructions achieved with Avizo are the most accurate and closely reflect the actual coal structure. The further improvement revealed that appropriate interpolation could bring the reconstructed coal sample data closer to the mercury intrusion data, thereby enhancing the accuracy of the coal sample’s 3D reconstruction. The improved 3D reconstruction method presented in this study provides more reliable data support for subsequent analyses of the coal’s microstructure.