Abstract: In order to remove the blur of uniform linear motion caused by the camera shake or the motion of the object itself, on the basis of the theory of Wiener filter image restoration and genetic algorithm, an adaptive Wiener filter based on genetic algorithm for deblurring is proposed. The genetic algorithm is used to adaptively optimize the k value of the regularization term, which could achieve the best image restoration effect from the fitness function. The experimental results show that this method could select the appropriate adaptive-degree-function, remove the motion blur of the real image as much as possible, and the effect of image restoration is better than that of Wiener filtering algorithm, which has some practical significance.

Key words: Wiener filtering, genetic algorithm, image deblurring