Image reconstruction method based on two-dimensional analysis sparse model and training dictionaries of two-dimensional analysis sparse model

Abstract

The invention discloses an image reconstruction method based on a two-dimensional analysis sparse model and training dictionaries of the two-dimensional analysis sparse model. The image reconstruction method can reflect image spatial characteristic, is good in reconstruction effect and includes steps of: (1) constructing a training sample set; (2) constructing and training a dictionary in the first direction; (3) constructing and training a dictionary in the second direction; (4) solving a dictionary omega 0 in the original sparse solution so as to perform one-dimensional analysis sparse reconstruction; (5) solving a reconstruction value by aid of the omega 0 and a one-dimensional analysis sparse reconstruction method; and (6) performing reverse operation for the reconstruction value to obtain a reconstruction value corresponding to N image blocks and further to obtain a reconstruction image.

Description

technical field [0001] The invention belongs to the technical field of signal modeling, and in particular relates to an image reconstruction method based on a two-dimensional analysis sparse model and a training dictionary thereof. Background technique [0002] Signal models play an important role in dealing with many problems, such as compression, sampling, reconstruction, and so on. At present, a very important method for signal modeling is the sparse representation method based on the synthetic model. The synthetic model looks like this: Dα=x, where x∈R d , D ∈ R dxn . Signal x ∈ R d is considered to be some sparse basis α∈R under a given dictionary n linear combination of . This model has attracted extensive attention in the past few years. In particular, it is widely used in the fields of image compression, super-resolution reconstruction, and image denoising. For the sparse representation method of this synthetic model, the research focus is on obtaining the di...

AI Technical Summary

Problems solved by technology

However, in reality, the correlation of image blocks is directional. If the image blocks are directly rearranged by column or row, unnecessary correlation will be introduced, such as the last pixel of the first column signal and the first pixel The correlation of the first pixel of the two columns is relatively weak in the image, but when the image blocks are rearranged into a one-dimensional signal by column, during the training process, it is bound to be considered that the two pixels are strongly correlated

The introduction of this unnecessary correlation generally exists in traditional dictionary training, which cannot reflect the temporal and spatial characteristics of the image, and the reconstruction effect is poor.

Images