Sparse representation-based image super-resolution reconstruction method

Abstract

The invention relates to a sparse representation-based image super-resolution reconstruction method. The steps include: 1) selecting a part of an input image sequence as a salient region, the remainder being a non-salient region; 2) training a pair of salient dictionaries D'1 and D'h according to the salient area, and performing context sparse decomposition on the salient area to obtain a salient sparse coefficient on the low-resolution salient dictionary; 3) training a pair of general dictionaries D1 and Dh according to the non-salient area, and performing sparse decomposition through the low-resolution general dictionary D1 to obtain a non-salient sparse coefficient; and 4) multiplying the sparse coefficient by the high-resolution salient dictionary D'h or the high-resolution general dictionary Dh to perform ratio reconstruction, thereby obtaining a high-resolution image sequence. On the basis of a traditional sparse representation super-resolution frame, the sparse representation-based image super-resolution reconstruction method emphasizes internal structure information of an image, and uses the internal structure information of the image as a prior model constraint L0-norm problem to solve, and the performance is superior to other methods in subjective and objective effects while complexity equivalent to that of a traditional sparse representation method is maintained.

Description

technical field [0001] The invention relates to an image super-resolution method, in particular to a visually salient-based context sparse decomposition image super-resolution method. The invention can be flexibly applied to the fields of video signal format conversion of high-end multimedia systems, zooming in on video surveillance interest areas, satellites, remote sensing and the like, and belongs to the field of image super-resolution reconstruction. Background technique [0002] Image super-resolution reconstruction is to overcome the limitations of imaging equipment or technology and reconstruct a high-resolution image from a single frame of low-resolution image or a sequence of low-resolution images. One of the most common methods is interpolation. Traditional interpolation algorithms such as nearest neighbor interpolation, bilinear interpolation, bicubic interpolation, and Lanczos interpolation all assume that the image has continuity, so as to predict the pixel val...

AI Technical Summary

Problems solved by technology

However, this method uses the general training set to learn the dictionary, and cannot adaptively reconstruct the local features of the image.

In addition, the a priori model of this type of method assumes that adjacent image blocks are independent, ignoring the interdependence of local image regions.

Images