Abstract: In this paper,the single-objective fitting optimization model of CT parameter calibration and the filtered back-projection image reconstruction model of CT imagining are established by using the attenuation law of X-ray in the medium and the related knowledge of CT image reconstruction.Firstly,based on the attenuation law of attenuation of X-rays in the medium and related theory Radon transform,a single-objective optimization model is established based on the principle of least squares.Secondly,The genetic algorithm is used to solve the grouping of the rotation center point and the detection unit spacing multiple times,and the solution results basically converges to the same value.Then,under the calibrated CT system theorem,the filtered back project image reconstruction model of CT image is established.Based on the received information of the detector unit,the image reconstruction model is solved to obtain the absorption rate information of each position on the square tray,and the position and shape of the medium on the tray are constructed.Finally,the feasibility of the model is verified by image reconstruction and comparison of the original data.