Abstract: Aimed at the problems of blurred edges and unclear textures after reconstruction of MRI images, an image reconstruction model with mixed regular constraints is proposed. The maximum and minimum nonconvex total variation with sparse representation closer to L 0 normal form and the shear wave transform with better texture details were used as constraints to construct the reconstruction model. Next, an extended ADMM algorithm was proposed to solve the new model. The experimental results show that the proposed model can not only preserve the texture information of the image, but also preserve the edge information of the image effectively.