Multi-scale residual attention network image super-resolution reconstruction method based on attention

Abstract

The invention belongs to the technical field of image super-resolution reconstruction, and discloses a multi-scale residual attention network image super-resolution reconstruction method based on attention. The method comprises the steps of selecting a common image data set as a to-be-experimented image set, dividing the to-be-experimented image set into an image training set and an image test set, and performing image preprocessing, designing a multi-scale residual structure unit module, introducing a channel attention mechanism, and building a multi-scale residual attention neural network model based on channel attention, inputting the preprocessed image training set into a multi-scale residual attention neural network model based on channel attention for model training, and inputting the preprocessed image test set into the trained model for testing to obtain a finally reconstructed high-resolution image. According to the method, a basic unit is enabled to focus on extraction of high-frequency information, important feature map information in a channel is better highlighted, important information in an image is better extracted, and reconstruction errors are reduced.

Description

technical field [0001] The invention belongs to the technical field of image super-resolution reconstruction, in particular to an attention-based multi-scale residual attention network image super-resolution reconstruction method. Background technique [0002] Image super-resolution reconstruction is a technique that utilizes the input image as a low-resolution image to generate a high-resolution output image. The application field of image super-resolution reconstruction involves the field of image processing, and has important application prospects in military, computer vision, medical diagnosis, public security and satellite images. [0003] At present, the algorithms used in image super-resolution reconstruction can be mainly divided into: methods based on interpolation, methods based on reconstruction and methods based on learning. [0004] Based on the interpolation method, common interpolation methods are nearest neighbor interpolation method, bilinear interpolation ...

AI Technical Summary

Problems solved by technology

However, deepening the number of network layers and simply stacking the number of network layers will increase the computational burden, and the network still has shortcomings such as difficulty in convergence.

At the same time, the introduction of a single residual structure cannot extract features of different scales. On this issue, the network treats the feature information of each channel equally, and performs equal processing on channels rich in high-frequency information and a large amount of low-frequency information, which is wasteful. Limited network computing resources, resulting in the loss of high-frequency information

Images