Multi-task super-resolution image reconstruction method based on KSVD dictionary learning

Abstract

The invention discloses a multi-task super-resolution image reconstruction method based on KSVD dictionary learning, which mainly solves the problem of relatively serious quality reduction of the reconstructed image under high amplification factors in the existing method. The method mainly comprises the following steps: firstly, inputting a training image, and filtering the training image to extract features; extracting image blocks to construct a matrix M, and dividing the matrix M into K classes to acquire K pairs of initial dictionaries H1, H2...Hk and L1, L2...Lk; then, training the K pairs of initial dictionaries H1, H2...Hk and L1, L2...Lk into K pairs of new dictionaries Dh1, Dh2...Dhk and Dl1, Dl2...Dlk by utilizing a KSVD method; and finally, carrying out super-resolution reconstruction on the input low-resolution image by utilizing a multi-task algorithm and the dictionaries Dh1, Dh2...Dhk and Dl1, Dl2...Dlk to acquire a final reconstructed image. The invention can reconstruct various natural images containing non-texture images such as animals, plants, people and the like and images with stronger texture features such as buildings and the like, and can effectively improve the quality of the reconstructed image under high amplification factors.

Description

technical field [0001] The invention belongs to the technical field of image processing, and relates to an image super-resolution reconstruction method, that is, by learning a relationship model between a low-resolution image and a high-resolution image, it is predicted from a single input low-resolution image The corresponding high-resolution images can be used for super-resolution reconstruction of various natural images. Background technique [0002] Super-resolution image reconstruction can be regarded as an inverse problem of recovering a high-resolution image from one or more low-resolution images. In order to solve this problem, some traditional model-based methods have been proposed: such as MAP (maximum a-posteriori) method, maximum likelihood estimation method, convex set projection method (POCS), etc. However, these traditional methods will produce over-smoothing and jagged effects, and the quality of the reconstructed image will be seriously degraded under the c...

AI Technical Summary

Problems solved by technology

Sun et al. have extended this method, mainly by using some original contour prior knowledge to remove the smoothing of boundaries and details, but this method still requires a large number of low-resolution images and high-resolution images block to ensure the sufficiency of prior contour detail information, the amount of calculation is huge, and the image reconstruction time is long, resulting in low efficiency

In addition, the above algorithms are all based on single-task reconstruction algorithms. Compared with multi-task algorithms, information sharing between multiple tasks cannot be performed, resulting in deviations in image reconstruction effects.

Images