Aiming at promoting execution efficiency and registration accuracy of dense scale invariant feature transform (SIFT) flow based image registration algorithm,in this paper,a new i mproved dense local self-similarity (ILSS) flow based registration algorithm is proposed.By separating color and lu minance information of color images through the color space conversion,the improved image registration algor ithm only extracts dense self-similar feature on the channel of luminance to improve efficiency of image feature extraction stage substantially.With constraints of maintaining the characteristics of the flow f ield smoothness,we then employ pyramid multi-resolution iterative method to improve the computational efficien cy of the estimation of the self-similar flow field stage.The basic idea of the multi-resolution iterativ e method is to roughly estimate the dense ILSS flow at a coarse level of image grid,then gradually propagate and re fine the dense flow from coarse to fine.A large number of experimental results show that compared with dense SIFT f low based image registration algorithm, the dense ILSS flow based one allows robust matching across different scene appe arances and has higher computational efficiency and registration accuracy.The success on these experim ents suggests that the image registration using dense ILSS flow can be a useful tool for various applications in computer vision and computer graphics.