Abstract: For classical image deblurring problem, this paper proposes an effective image deblurring algorithm, where an l₀-norm regularization is minimized under the constraint that the solution explains the observations sufficiently well. To effectively solve the optimizing problem, this paper introduces two auxiliary variables and divides the original problem into two sub-problems which are solved by the alternating direction method. The experimental results indicate that with different size blurry kernels, the proposed algorithm can recover image effectively and steadily for the Gaussian blur and motion blur.