In order to improve robustness of the optical flow method in the treatment of the illumination change and large displacement. The optical flow estimation method combining structure texture decomposition preprocessing and weighted median filter is proposed. The idea to obtain the data term is to use the combination of gray and gradient constancy assumption

the combination of local and global constraint. Meanwhile

the use of texture decomposition

the weighted median filter and the pyramid structure is further enhance the accuracy and practicability of the optical flow algorithm. The proposed method is evaluated by using both the Middlebury optical flow database images and real scene images. The experimental results show improved optical flow estimation method in the treatment of the illumination change performance is good

not only to obtain dense optical flow field

and improves the detection precision optical flow field on the edge of the target. Compared with traditional optical flow method

the proposed method under the condition of illumination change can obtain a more ideal result and reduce the interference of the actual light changes in a scene. The optical flow method can better apply to actual scenario.