基于总变分加权低秩矩阵恢复的椒盐噪声去噪

摘要

传统的基于低秩矩阵恢复的椒盐噪声去噪算法易产生条纹失真，中值滤波去除椒盐噪声后边缘易产生移位和块状效应，纹理细节不太清晰，为此提出一种椒盐噪声去噪模型。在原有的低秩矩阵核范数约束基础上引入总变分约束项，为提高低秩矩阵的低秩性和稀疏矩阵的稀疏性，引入加权的方法。实验结果表明，该算法能增加低秩矩阵的低秩性和稀疏矩阵的稀疏性，保证了去噪效果，保留了图像的细节信息，具有更佳的视觉效果，提高了客观评价指标。%The traditional salt-and-pepper denoising based on low-rank matrix recovery algorithm is easy to produce stripes dis-tortion,median filter may lead to edge shift and blocky,and serious detail loss.To solve the problem,a salt-and-pepper denoi-sing algorithm was proposed,in which total variation was added to the low rank restraint model.Inspired by reweighted L1 mini-mization for sparsity enhancement,reweighting singular values were used to enhance low rank of matrix,an efficient iterative re-weighting scheme was proposed for enhancing low rank and sparsity simultaneously.Experimental results show that the pro-posed algorithm can enhance low rank and sparsity of a matrix simultaneously,guarantee visual effect and keep the details.At the same time,the obj ective evaluation indexes are improved.