拉普拉斯与双边滤波的图像去运动模糊算法

运动模糊图像盲复原是图像复原中的关键问题之一,由于模糊信息的复杂性以及图像噪声的影响,导致复原的结果有振铃效应.为解决这些问题,提出基于拉普拉斯先验的改进算法.在原有的模糊核估计过程中,引入拉普拉斯锐化增强,增强图像的边缘信息,达到改善模糊核估计结果的目的.在复原时,针对图像振铃噪声,引入了双边滤波,抑制该噪声对复原结果的影响,进一步改善复原图像的质量.与三种现有图像去模糊算法相比,该方法能更准确地估计模糊信息,对图像复原具有更好的鲁棒性.

Image deblurring algorithm using Laplacian prior and bilateral filtering approach

To restore blurry image is a key problem in blind image deblurring,because of complexity of blurring information and image noise,which will affect restoration result has ring effect.To deal with these problems,propose improved algorithm based on Laplacian prior.In kernel estimation process,introduce Laplaction sharpening to strengthen edges information of image,which improve the result of kernel estimation.To suppress ring noise,adopt bilateral filtering in blur image restoration.Experimental results show that the method is more robust to artifact noise and achieves higher quality results than precious approaches on blurry image.