Sparse Representation Image Reconstruction Method Based on Gaussian Scale Block Grouping

Abstract

The present invention provides a sparse representation image reconstruction method based on Gaussian scale structure block grouping, which includes the following steps: using a non-local self-similar model trained from natural images, mixing non-local similar blocks into the prior model method obtained In the grouping, use the search method to extract the optimal block grouping model; combine the block grouping model and the non-locally extended Gaussian scale mixture model, use the alternating minimization method to perform synchronous sparse coding, and solve the update image block; mix the block grouping model and the Gaussian scale The model is combined into the encoding framework, and the selected training dictionary is used to calculate the image reconstruction update solution obtained by the joint model, and the update solution value is sent back to the block grouping model to perform steps 1 and 2 again, and so on. Iterate until the optimal solution is generated, then output the optimal solution of the reconstructed image. The reconstructed image obtained by this method has better edge, texture and other detail preservation performance and better peak signal-to-noise ratio quality.

Description

technical field [0001] The invention belongs to the technical field of image super-resolution reconstruction, and in particular relates to a sparse representation image reconstruction method based on Gaussian scale structure block grouping. Background technique [0002] With the rapid development of the information age, digital images are widely used due to their good performance, and have become one of the most important carriers for human beings to transmit information. However, in the process of imaging, transmission, conversion, storage, reproduction and display of images, due to the inherent physical limitations of imaging equipment itself and the influence of unfavorable factors such as unpredictable external environments, the acquired images are often degraded images. . [0003] In order to restore useful information in degraded images, image super-resolution restoration technology has become a research hotspot in the fields of computer vision, computer graphics, etc...

AI Technical Summary

Problems solved by technology

However, in the process of imaging, transmission, conversion, storage, reproduction and display of images, due to the inherent physical limitations of imaging equipment itself and the influence of unfavorable factors such as unpredictable external environments, the acquired images are often degraded images.

Images