Using the nanophosphors, X-ray luminescence computed tomography (XLCT) is proposed as a new molecular imaging modality that provides functional and molecular imaging capability. The reconstruction for the nanophosphors sample distribution is an ill-posed problem due to the strong scattering of photons in biological tissues. So accurate and stable reconstruction of nanophosphors distribution remains a challenging problem. In order to solve the problem, the diffusion approximation model is chosen to describe the photon transfer process, and the split augmented Lagrangian shrinkage algorithm based on L1 regularization is used for reconstruction. In the numerical simulations and physical experiments, the new method exhibits better performance in imaging quality and convergence rate compared with primal augmented Lagrangian method.