Unlock instant, AI-driven research and patent intelligence for your innovation.

Image compression method and device based on generative adversarial network

An image compression and network technology, applied in image communication, digital video signal modification, electrical components, etc., can solve problems such as unproposed solutions

Inactive Publication Date: 2021-09-21
核芯互联科技(青岛)有限公司
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] For the above problems, no effective solution has been proposed

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
  • Image compression method and device based on generative adversarial network
  • Image compression method and device based on generative adversarial network
  • Image compression method and device based on generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027] Figure 4 is a flow chart of an image compression method based on a generative confrontation network according to an embodiment of the present invention, such as Figure 4 As shown, the method includes the following steps:

[0028] Step S102, acquiring original image data.

[0029] Step S104, performing DCT transformation on the original image data through a generative adversarial network to obtain an image transformation result.

[0030] Step S106, performing interpolation with an 8*8 quantization table according to the image transformation result to obtain an interpolation quantization table.

[0031] Step S108, using the interpolation quantization table to perform quantization and entropy coding on the image transformation result to obtain target image data.

[0032] Optionally, the generator of the GAN uses an enhanced U-shaped network to extract features.

[0033] Optionally, the enhanced U-shaped network uses EfficientnetB0 as a basic network for extracting fe...

Embodiment 2

[0047] Figure 5 is a structural block diagram of an image compression device based on a generative confrontation network according to an embodiment of the present invention, such as Figure 5 As shown, the device includes:

[0048] An acquisition module 20, configured to acquire original image data.

[0049] The transformation module 22 is configured to perform DCT transformation on the original image data through a generative adversarial network to obtain an image transformation result.

[0050] The interpolation module 24 is configured to perform interpolation with an 8*8 quantization table according to the image transformation result to obtain an interpolation quantization table.

[0051] The processing module 26 is configured to perform quantization and entropy coding on the image transformation result by using the interpolation quantization table to obtain target image data.

[0052] Optionally, the generator of the GAN uses an enhanced U-shaped network to extract fea...

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 an image compression method and device based on a generative adversarial network. The method comprises the following steps: acquiring original image data; performing DCT on the original image data through a generative adversarial network to obtain an image transformation result; performing interpolation on a 8*8 quantization table according to the image transformation result to obtain an interpolation quantization table; and performing quantization and entropy coding on the image transformation result by using the interpolation quantization table to obtain target image data. According to the method, the technical problems are solved that an image compression algorithm completely based on deep learning is long in compression time and requires a deep learning model for decoding, while the deep learning model for decoding has a very low decoding speed if not operating in a high-performance GPU or a dedicated chip.

Description

technical field [0001] The present invention relates to the field of image compression coding, in particular to an image compression method and device based on a generative confrontation network. Background technique [0002] With the continuous development of intelligent technology, more and more intelligent devices are used in people's life, work and study. The use of intelligent technology has improved the quality of people's life and increased the efficiency of people's study and work. [0003] Due to the relatively large amount of image data, in the actual storage and transmission process, the image needs to be compressed and encoded. The existing widely used image compression encoding JPEG adopts DCT transformation, quantization and entropy encoding. Due to the limitation of computing resources, the block size of DCT transformation should not be too large. In the standard JPEG scheme, the size of DCT is 8, but for the field of image compression, the larger the size of ...

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): H04N19/625H04N19/124H04N19/91
CPCH04N19/625H04N19/124H04N19/91
Inventor 黄震坤徐洋尹海滨
Owner 核芯互联科技(青岛)有限公司