一种快速的四方向全变分图像去噪方法

A Fast Image Denoising Method by Quaternion Total Variation

图像去噪是图像处理领域的重要研究方向之一.在众多图像去噪算法中,全变分去噪方法由于其良好的数学解释引起学者们的广泛关注.传统全变分方法挖掘了图像横向和纵向的梯度信息,因此可通过增加方向数量进一步提高去噪性能.其导致的结果是由于去噪模型中约束条件增加将产生更大的计算量.本文首先通过快速傅里叶变换和卷积理论将空间域的四方向全变分约束问题转换成频率域,然后结合分裂Bregman迭代算法进行快速图像去噪.经与业内其他先进方法进行对比,本算法的快速性与有效性得到了验证.

hnage denoising is an important problem in image processing. Among many approaches, the total variation has attracted great attention because of its nice mathematical interporetation. Traditional total variation explores the vertical and horizontal directions and the number of directions can be increased to further improve denoising performance. The resulting challenge is higher computation since muhiple constraints are introduced in denoising model. This work first transforms the quaternion total variation constraints problem in the spatial domain into a problem in the frequency domain by using the fast Fourier transform and the convolution theorem. Then, it incoqoorates the split Bregman iterations to enable fast image denoising. The eftectiveness and the computation efficiency are verified by the comparisons with other methods including state-of-the-art methods.