Supercharge Your Innovation With Domain-Expert AI Agents!

An image compression method based on combination of a convolutional neural network and traditional coding

A convolutional neural network and image compression technology, applied in the field of image compression, achieves the effects of reducing image distortion, reducing high-frequency information components, and improving image reconstruction quality

Active Publication Date: 2019-06-18
FUDAN UNIV
View PDF7 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

After entering the information age, the flow of data in the Internet is increasing day by day, which is a very heavy burden on the network bandwidth and storage resources that current hardware technology can provide

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
  • An image compression method based on combination of a convolutional neural network and traditional coding
  • An image compression method based on combination of a convolutional neural network and traditional coding
  • An image compression method based on combination of a convolutional neural network and traditional coding

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0023] The present invention will be further described below through the examples, but the protection scope of the present invention is not limited to the examples.

[0024] use figure 1 The network structure in , the neural network is trained with 400 images of size 481×321.

[0025] The specific implementation method is:

[0026] (1) During training, use the method used in [7] to randomly crop the image to 180×180, and then cut the cropped image into 64 small images with a size of 40×40, and use a step size of 20 when cropping. The initial learning rate is set to 0.01, which decays to 0.0001 after 80 epochs. The loss function is minimized using the Adam stochastic gradient descent method. The batch size is set to 64;

[0027] First perform alternate training: fix the parameters of the decCNN network, minimize the loss function of the enhCNN network, let the network learn image enhancement tasks, then fix the parameters of the enhCNN network, minimize the loss function of...

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 belongs to the technical field of digital image processing, and particularly relates to an image compression method based on combination of a convolutional neural network and traditionalcoding. The method comprises the following steps: reducing a high-frequency information component of an image by using a convolutional neural network decCNN; Compressing the image by using a traditional encoding module to obtain an image code for storage and transmission; Decoding the obtained coded data to obtain a reconstructed image; And the convolutional neural network enhCNN is utilized to enhance the decoded image, so that the reconstruction effect is improved. Experimental results show that better image reconstruction quality can be obtained when a higher compression ratio is achieved,and resources occupied by image data in the processes of storage, transmission and the like are greatly reduced.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, and in particular relates to an image compression method. Background technique [0002] With the continuous development of computer technology and network communication technology, real-time video communication, video surveillance and other fields have attracted more and more attention. After entering the information age, the flow of data in the Internet is increasing day by day, which is a very heavy burden on the network bandwidth and storage resources that current hardware technology can provide. As image data is the most important resource of the Internet, it is undoubtedly very meaningful to effectively compress it. Image compression technology (Image Compression) uses as little data as possible to represent the original image, and at the same time allows the quality of the restored reconstructed image to be distorted to a certain extent, which greatly reduces the pressure ...

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
IPC IPC(8): G06T9/00G06N3/04H04N19/42
CPCY02T10/40
Inventor 颜波容文迅
Owner FUDAN UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com