The spectral CT system independently developed by the laboratory can not only generate high-resolution slice images, but also accurately determine the types of different components in the sample by using the reconstructed spectral information of up to 128 spectral lines per pixel. In the current interface, the plane position where the imaging slice is located is marked by the laser line scan, which is convenient for the user to intuitively understand the imaging structure. The multi-spectral imaging capability of the system makes it excellent in the non-destructive detection of the internal structure and composition of ores, jewelry, dangerous goods and other objects, and can also be extended to the biomedical field to achieve non-invasive diagnosis and quantitative analysis of live pathology.

Taking nine ore samples as an example, the system collects multi-view data through 360-degree projection scanning, and combines it with an efficient reconstruction algorithm to generate continuous tomography slice images. Subsequently, the laboratory extracted the spectral information from the nine target regions in the slice to accurately identify and classify different mineral components according to the spectral line characteristics. The whole process is fully automated, which not only ensures the detection accuracy, but also greatly improves the analysis efficiency, providing a one-stop and intelligent solution for geological exploration, quality inspection research and medical and engineering cross-application. In the future, the laboratory will continue to improve the on-site adaptation capability and real-time data processing module of the equipment, and help the industry and scientific research institutions to carry out more in-depth exploration of spectral CT applications in multiple fields.