运动模糊图像的维纳滤波复原研究

Study on Wiener filtering for restoration of motion blurried image

根据运动造成图像模糊的特点详细分析了匀速直线运动模糊图像的退化模型和恢复模型,提出直接在运动方向上建立点扩展函数的算法,并利用改进的霍夫变换检测点扩散参数,再利用二次维纳滤波的方法复原图像。通过实验表明在图像先验条件的要求没有增加的情况下,该方法提高图了像复原的抗噪性和稳定性,并且有效的保持图像细节。

According to the characteristics of image blurring caused by motion,analyzed detailedly the degradation model and the recovery model of blurred image causing by uniform linear motion,made up the algorithm of point spread function which is directly set up in the moving direction,and used the improved Hough transform to detect point spread parameters.This paper uses secondary Wiener filtering method to recover blurring image.Experiments show that the means enhance the noise immunity and stability of recovered images, as well as effectively preserved image detail at the request of a priori conditions without adding any circumstances.