Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Video compression method based on super-resolution reconstruction

A technology of super-resolution reconstruction and video compression, which is applied in digital video signal modification, electrical components, image communication, etc. It can solve the problems of low video peak signal-to-noise ratio, loss of information, and difficulty in use, so as to reduce the degree of compression, Improve the peak signal-to-noise ratio and reduce the effect of information loss

Active Publication Date: 2019-08-23
XIDIAN UNIV
View PDF7 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] At present, most of the traditional video compression methods use the correlation of video data in space and time for video compression. Among them, the widely used video compression method is the H.264 video compression method, which mainly uses intra-frame prediction. Video compression, inter-frame predictive compression, and data quantization coding are used to achieve video compression, but this method does not fully apply the prior information in video big data, and will lose more in some scenarios that require low bit rate compression of data information, resulting in a low peak signal-to-noise ratio of the reconstructed video, which brings difficulties to subsequent use

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Video compression method based on super-resolution reconstruction
  • Video compression method based on super-resolution reconstruction
  • Video compression method based on super-resolution reconstruction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The embodiments and effects of the present invention will be further described below in conjunction with the accompanying drawings.

[0036] refer to figure 1 , the specific implementation steps of this example are as follows:

[0037] Step 1, get training samples.

[0038] The high-definition video in the present embodiment comprises 542 video sequences, and each video sequence is made up of 32 continuous frames, mainly is the high-definition video sequence collected from high-definition documentaries, relatively real, and there are forest, snow, desert, urban life in the data set and other scenes, where most of the video frames have a resolution of 1280*720, the steps to obtain training samples from these video sequences are as follows:

[0039] (1a) Store 2 copies of the above-mentioned 542 video sequence backups, one as the original sample set X={X 1 ,X 2 ,...,X i ,...,X N}, where X i Indicates the i-th video, N indicates the total number of videos, X i ={X ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a video compression method based on super-resolution reconstruction, which mainly solves the problem of low peak signal-to-noise ratio of a reconstructed video caused by loss of more information under the condition of low code rate compression in the existing method, and comprises the following implementation steps: 1, obtaining a training sample containing a compressed sample set and an input video set; 2, constructing a deep convolutional neural network model based on a TensorFlow architecture; 3, training the constructed deep convolutional neural network model by using the obtained training sample; and 4, preprocessing the video to be compressed, inputting the preprocessed video into the trained deep convolutional neural network model, and compressing and recovering the video to obtain a final recovered video. According to the method, information loss under the condition of low code rate compression is reduced, the peak signal to noise ratio of the recoveredvideo is improved, and the method can be applied to video storage, video transmission and video communication occasions.

Description

technical field [0001] The invention belongs to the technical field of video compression, in particular to a video compression method, which can be applied to video storage, video transmission and video communication occasions. Background technique [0002] With the continuous development of the video industry chain and the continuous breakthrough of computer technology, the information dissemination method using video as the carrier has been widely used. Compared with ordinary text and pictures, the amount of data contained in video is relatively large, and with the rapid development of imaging equipment, in some scenarios where ultra-high-definition video is used, the amount of data contained in ultra-high-definition video is very large. When storing or transmitting, video is often required to be compressed due to the limitation of memory capacity and network bandwidth, and the data will be damaged during the compression process, which will bring difficulties to subsequent...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): H04N19/132H04N19/172H04N19/42
CPCH04N19/132H04N19/172H04N19/42
Inventor 董伟生范兴宣毋芳芳石光明
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com