Many researchers in image and video processing field test the effectiveness of the proposed or existing methods depended on the assumption that the brightness or illumination in scene is static among all sequenced images or frames. So they used synthetic dataset with frames contain approximately static blurriness degrees. This is not practical in the real world. In this paper, a method of generating synthetic blurred video dataset with frames containing different blur variances to solve this problem. The result showed that the proposed algorithm has ability to produce useful blurred dataset with having different blurriness values.