In medical applications of computed tomography(CT)imaging,due to the excessive X-ray radiation exposure imposed to the patients is harmful,it is of importance to reconstruct high-quality CT images from limited CT data.In this paper,we consider X-ray imaging in limited scanning angular range and limited projection views,which are both effective way to reduce the radiation dose.While the two imaging modalities lead to CT reconstruction from limited projections which do not satisfy the complete conditions for exact reconstruction.In order to solve the reconstruction problem,we develop an image reconstruction model where the l0-norm of the image gradient is taken as the regularization term.To validate the efficiency and the feasibility of the developed reconstruction method,numerical experiments are performed.From the experimental results,it is shown that the developed reconstruction method has advantage of suppressing the streak artifacts and the gradual changed artifacts than classical reconstruction method.