Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Robust image watermarking method and system based on deep learning and attention network

An image watermarking and deep learning technology, applied in image data processing, image data processing, biological neural network models, etc., can solve the problems of imperceptibility and unsatisfactory robustness, and achieve resistance to various attacks and imperceptibility The effect of sex and robustness

Active Publication Date: 2022-01-28
SHANDONG UNIV
View PDF11 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the imperceptibility and robustness of existing methods are still not ideal due to the different perception capabilities of the human visual system to different regions of the image and the different anti-attack capabilities of different pixels in the image.

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
  • Robust image watermarking method and system based on deep learning and attention network
  • Robust image watermarking method and system based on deep learning and attention network
  • Robust image watermarking method and system based on deep learning and attention network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] Embodiment 1 of the present invention introduces a robust image watermarking method based on deep learning and attention network.

[0047] Such as figure 1 A robust image watermarking method based on deep learning and attention network is shown, including three stages: watermark embedding stage, simulated attack stage and watermark extraction stage, specifically including the following steps:

[0048] Step S01: Perform data reorganization on the original image data, the input size is The original image of , restructured as The tensor, in this embodiment due to consideration of the follow-up DCT transform, h = M / 8, w = N / 8, c =64.

[0049] Step S02: build as figure 2 The attention model shown, the size generated by step S01 is The raw image tensor input figure 2 In the network, the attention weights are inferred sequentially along the two dimensions (channel and space), multiplied by the original tensor and adaptively adjusted to generate attention fe...

Embodiment 2

[0094] Embodiment 2 of the present invention introduces a robust image watermarking system based on deep learning and attention network.

[0095] Such as Figure 6 A robust image watermarking system based on deep learning and attention networks is shown, including:

[0096] The watermark embedding module is configured to obtain the original image tensor, obtain the attention image tensor according to the obtained original image tensor and the attention model, and generate the watermarked image tensor based on the obtained attention image tensor and the watermark embedding model image;

[0097] The simulated attack module is configured to generate an attacked image according to the generated image containing the watermark and the constructed attack network model;

[0098] The watermark extraction model is configured to extract the image watermark according to the attacked image, the attention model and the deep learning model.

[0099] The detailed steps are the same as the ...

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 watermarking, and provides a robust image watermarking method and system based on deep learning and an attention network, and the method comprises the following steps: obtaining an original image tensor; obtaining an attention image tensor according to the obtained original image tensor and an attention model; generating a watermark-containing image based on the obtained attention image tensor and a watermark embedding model; generating an attacked image according to the generated watermark-containing image and the constructed attack network model; and extracting an image watermark according to the attacked image, the attention model and the deep learning model. Attention weights are deduced along two dimensions of a channel and a space by utilizing the perception capability of a human visual system to different regions and the anti-attack capability of different pixel points, and regions which have small influence on human visual perception and are good in robustness are searched for watermark embedding.

Description

technical field [0001] The invention belongs to the technical field of digital image watermarking, and in particular relates to a robust image watermarking method and system based on deep learning and attention networks. Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] With the popularization of network communication and the rapid development of digital multimedia, the manufacture and dissemination of digital information is becoming more and more convenient. However, while it is convenient for people's lives, the problems of digital product copyright disputes emerge in an endless stream, and copyright protection is urgent. As an effective means of image copyright protection, image watermarking technology embeds watermarks into images in an invisible form for copyright identification. However, the current image watermarking methods are sti...

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): G06T1/00G06N3/04G06F21/16
CPCG06T1/0021G06F21/16G06T2201/0052G06T2201/0065G06N3/045
Inventor 王成优赵怡梦周晓
Owner SHANDONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap 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