Method for improving video resolution through convolutional neural network

Abstract

The invention discloses a method for improving video resolution through a convolutional neural network. According to the method, a video super-resolution reconstruction model based on the convolutional neural network is constructed according to a video image sequence, and in the model construction process, super parameters, including convolution kernel size, the number of neural network layers, etc. of the convolutional neural network are set based on the characteristics of the video image sequence; and then a single-image super-resolution method is utilized to generate image training weightsused for initializing weight parameters of the video model, redundant information among video frames is fully utilized, multiple frame video images are used as input of the convolutional neural network model, and finally a high-resolution video is obtained through an incremental iteration method. Through the method, the video super-resolution model is high in training speed and high in predictionprecision; and a comparison test verification result indicates that compared with other super-resolution algorithms, the peak signal-to-noise ratio and structural similarity of the images reconstructed through the method have comprehensive optimal results.

Description

technical field [0001] The invention relates to a video processing method, in particular to a method for improving video resolution by using a convolutional neural network. Background technique [0002] In order to obtain high-resolution image sequences or videos, video super-resolution (Video Super Resolution, VSR) technology is used to further post-amplify the original low-resolution video, which is conducive to improving the utilization of video resources. It is applied to various fields such as video surveillance, satellite images, and medical image processing. The application conditions of super-resolution technology are simple, as long as different low-resolution (Low Resolution, LR) images of the same scene can be obtained. Under the condition of not increasing the cost of hardware, it uses the existing imaging equipment to improve the resolution of the image by using algorithms, breaking through the limit of the spatial resolution of the image obtained by the imagin...

AI Technical Summary

Problems solved by technology

At present, research on super-resolution methods using Convolutional Neural Networks (CNN) has been carried out. Chao Dong et al. , 2014) used a deep convolutional neural network to construct an end-to-end mapping relationship between LR images and HR images, which first verified the feasibility of convolutional neural network applications in the field of image super-resolution, but the algorithm calculates The amount is complex and prone to overfitting, so it is not suitable for direct processing of multi-frame video

Armin Kappeler et al. based on the convolutional neural network in "Video Super-Resolution With Convolutional Neural Networks" ("IEEE Transactions on Computational Imaging", 2016, 2(2):109-122), first using the pre-training parameters of a single frame image, After multiple hidden layers, it is converted into the parameters of the video sequence image model, but it does not make full use of the redundant information between video frames, and the generated high-resolution video effect is not ideal.

Images