The progress of human society cannot be separated from medical development. CT is an important auxiliary instrument of modern medicine, and its imaging precision and stability are essential conditions. To provide a more accurate reference for medicine, data fitting, parameter estimation, interpolation, neural network, continuous discretization, and image processing algorithms are applied in this paper to establish a corresponding model between absorption strength and images through Radon transform and Iradon transform, thereby demonstrating the CT imaging process and CT parameter calibration process. Meanwhile, the CT imaging rules are obtained. As a result, CT parameter calibration precision and stability are improved, as well as diagnosis precision. First, the Excel form is compressed according to the data in Annex 1, and the characteristics of data in Annex 2 are observed. Horizontal illumination and vertical illumination of light are taken to determine the detector unit space. When horizontal light illuminates the medium, the chord passing through the center of the circle is longest. The energy magnitude, physical relation between energy and thickness, and length of the corresponding chord can be found in Annex 2 to establish the matrix equation set and solve the center of rotation of the CT system. The data in Annex 2 corresponds to 180 directions, and the equation set is established to gain every direction.
Submitted on May 5, 2018; Revised on June 19, 2018; Accepted on July 21, 2018