Reference image guided super-resolution method based on dense matching and adaptive fusion

Abstract

The invention belongs to the field of computer vision, relates to an image super-resolution algorithm guided by a reference image, and aims to greatly improve the running speed and the visual result compared with an existing algorithm. The invention discloses a reference image guided super-resolution method based on dense matching and adaptive fusion. The method comprises the following steps: establishing a training data set; aligning the reference image with the low-resolution image; inputting the low-resolution image and the aligned reference image into a convolutional neural network for fusion; setting the learning rate of the network and the weight of the loss function of each part, training the convolutional neural network by using a deep neural network framework PyTorch until the loss converges, and generating a training model; and performing image super-resolution by utilizing the generated training model. The method is mainly applied to computer image processing occasions.

Description

technical field [0001] The invention belongs to the field of computer vision and relates to an image super-resolution algorithm guided by a reference image. Specifically, feature points are extracted through the Scale-invariant feature transform (SIFT) algorithm combined with Random Sample Consensus (RANSAC) to calculate the homography matrix to perform rigid transformation on the reference image, and then through optical flow matching Realize non-rigid transformation, obtain a reference image that is aligned with the low-resolution image as much as possible, and reconstruct the low-resolution image through the encoding and decoding network and similar shape fusion module. Background technique [0002] Super-resolution is a technology that improves the resolution of the original image by means of hardware or software. One or more low-resolution images can be used to obtain a high-resolution image. The current single-image super-resolution has achieved a high PSNR, but there...

AI Technical Summary

Problems solved by technology

This algorithm can search for the most similar image block in the reference image to restore the details of the low-resolution image, but the disadvantage is that the calculation complexity is high, and it cannot realize non-rigid transformation, so that the information of the reference image cannot be fully utilized.

[0004] Another way to achieve non-rigid transformation is to use dense optical flow, but usually this type of algorithm is very sensitive to displacement and difficult to learn, the existing optical flow matching algorithm can only be used for small displacement changes such as light field diagram, video etc., or data with strong prior information for specific images such as face images

Images