An adaptive image colorization algorithm based on reproducing kernel Hilbert spaces is proposed in this paper aiming at overcoming unsatisfactory and inefficiency in the traditional colorization algorithm.A nonlinear equation is established through minimization of a functional.By choosing adaptive reproducing kernel related to the image properties.A colorized image is obtained by numerically solving the equation with a least-square approach.The simulation results show that the proposed algorithm outperforms previous method in improving the image quality and selecting adaptive parameter.