LAUSR

For the hyperspectral image (HSI) denoising problem, a symmetric proximal alternating direction method multiplier (spADMM) is proposed to solve the sparse optimization problem which cannot be solved accurately by traditional ADMM. The proposed method finds a high-quality recovery method using the traditional low-rank Tucker decomposition method, which can fully take into account the overall spatial and spectral correlation between HSI bands by using the Tucker decomposition. By choosing appropriate steps to update the Lagrange multipliers twice, it makes the selection and use of variables more flexible and better for solving sparsity problems. To maintain stability, we also add appropriate proximity terms to solve the problem during the computation. Experiments have shown that the spADMM has better results than the traditional ADMM. The final experimental results on the dataset demonstrate the effectiveness of the method.