一种基于块Krylov迭代的三阶张量低管秩逼近的近似算法

An Approximation Algorithm for Third-order Tensor Low-tubal Rank Approximation Based on Block Krylov Iteration

基于块Krylov迭代和随机计数嵌入矩阵技术,提出了一种新的张量低管秩逼近算法。利用块Krylov迭代和随机计数嵌入矩阵技术,保证了算法精度处于较高水平。与其他几种常见算法相比,彩色图片实验结果表明,所提出的算法具有峰值性噪比值高而运算时间短的特点;人工合成数据实验表明,提出的算法在投影误差和相对误差上同时占优。

A new tensor low-tubal rank approximation algorithm is proposed,which is based on block Krylov iteration and random count embedding matrix techniques.By using block Krylov iteration and random count embedding matrix techniques,the accuracy of the algorithm is guaranteed at a high level.Compared with other prevailing algorithms,the experimental results of color images show that the proposed algorithm has higher PSNR values and less computing time.The synthetic data experiments show that the proposed algorithm is superior to prevailing methods in terms of projection error and relative error.