CN112579994A - Digital product content protection system and method based on artificial intelligence - Google Patents
Digital product content protection system and method based on artificial intelligence Download PDFInfo
- Publication number
- CN112579994A CN112579994A CN202011543482.7A CN202011543482A CN112579994A CN 112579994 A CN112579994 A CN 112579994A CN 202011543482 A CN202011543482 A CN 202011543482A CN 112579994 A CN112579994 A CN 112579994A
- Authority
- CN
- China
- Prior art keywords
- digital
- watermark
- information
- image
- signature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 8
- 238000005516 engineering process Methods 0.000 claims description 20
- 238000013475 authorization Methods 0.000 claims description 15
- 239000013598 vector Substances 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims 1
- 239000003550 marker Substances 0.000 claims 1
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000005096 rolling process Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000010422 painting Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/10—Protecting distributed programs or content, e.g. vending or licensing of copyrighted material ; Digital rights management [DRM]
- G06F21/16—Program or content traceability, e.g. by watermarking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Technology Law (AREA)
- Computer Hardware Design (AREA)
- Computer Security & Cryptography (AREA)
- Editing Of Facsimile Originals (AREA)
- Image Processing (AREA)
Abstract
The application relates to a digital product content protection system based on artificial intelligence, which comprises an encoding device, a decoding device, a convolution block execution nuclear reactor and a model parameter memory; the coding device and the decoding device are correspondingly arranged, are connected with the convolution block execution nuclear reactor and are used for executing data calling from the convolution block execution nuclear reactor; the convolution block execution kernel heap is connected with the model parameter memory, and the convolution block execution kernel heap loads the parameters of the specified addresses in the model parameter memory according to the register parameters in the execution process. The method and the device can realize the anti-counterfeiting of the digital product content and confirm that the provider of the digital product content is a real and reliable information provider.
Description
Technical Field
The application relates to a digital product content protection system and method based on artificial intelligence, which are applicable to the technical field of information.
Background
The development of the internet makes the copying, circulation and dissemination of information in the virtual network world easier. However, the internet also dilutes the labor value of content creators in their willingness to spread often while attempting to eliminate information access boundaries. Digital works are naturally reproducible, tamperable, and non-exclusive. In addition, consumers have poor copyright awareness and the phenomena of theft and abuse of digital works are very common. Meanwhile, the acceleration of online information circulation and the increasing complexity of social networks are difficult to provide evidence for right maintenance, the right maintenance cost is too high, and related rights and interests are often difficult to effectively protect.
Digital blind watermarks are marks that are covertly embedded in noise-tolerant signals such as video or image data. It is commonly used to identify copyright ownership of such signals. "watermarking" is the process of hiding digital information in a carrier signal. Digital blind watermarks can be used to verify the authenticity or integrity of a carrier signal or to reveal the identity of its owner, primarily for tracking copyright infringements. Like traditional physical watermarks, digital blind watermarks are typically only perceptible under certain conditions, such as using a particular extraction algorithm. Traditional watermarks are often applied to visible entities such as art paintings and banknotes, while in digital blind watermarks the signals are usually electronic pictures and videos. A signal may carry several different watermarks at the same time. Unlike metadata added to the carrier signal, the digital blind watermark does not alter the size of the carrier signal. One application of digital blind watermarking is source tracking. The watermark is embedded in the digital signal at each distribution point. If a copy of the work is later found, the watermark can be retrieved from the copy and the distribution source known.
The traditional digital resource copyright system usually adopts a centralized storage mode to carry out centralized storage on multimedia data, but the method is easy to be attacked in a centralized way from the outside, so that the stored data is lost and falsified, the authenticity and the credibility of information cannot be ensured, and sensitive data storage is not facilitated. This brings great trouble to the handling of infringement disputes and becomes a pain point for copyright protection and digital resource infringement tracking.
With the rise of bitcoin, the underlying blockchain technology has attracted extensive attention in recent years, and research on application of blockchain technology has emerged in various application fields. The block chain technology is born by the basic technology and the basic architecture of the bitcoin, has the characteristics of decentralization, distrust, collective maintenance and reliable storage, and is essentially a decentralization distributed ledger database. The point-to-point network distributed system is constructed on the basis of a cryptography technology, and a technology with the characteristics of data non-falsification, information disclosure transparency and the like is created through consensus. The digital signature mechanism, the existence certification and the non-falsification of the block chain are particularly suitable for the management of digital copyright, and the benefits of content producers can be effectively guaranteed.
Disclosure of Invention
The purpose of the present application is to design a system and a method for protecting digital product content based on artificial intelligence, which can solve the above-mentioned defects of the existing digital resource protection and provide a true and credible basis for dealing with infringement disputes.
The application relates to a digital product content protection system based on artificial intelligence, which comprises an encoding device, a decoding device, a convolution block execution nuclear reactor and a model parameter memory; the coding device and the decoding device are correspondingly arranged, are connected with the convolution block execution nuclear reactor and are used for executing data calling from the convolution block execution nuclear reactor; the convolution block execution kernel heap is connected with the model parameter memory, and the convolution block execution kernel heap loads the parameters of the specified addresses in the model parameter memory according to the register parameters in the execution process.
The encoding device can be used for extracting the content characteristics of an original image, fusing a watermark input sequence and generating a marked image by signature; and the decoding device restores the initial watermark signature from the marked image by adopting a feature extraction model.
Preferably, the system further comprises an anti-interference device, the anti-interference device is arranged between the encoding device and the decoding device, and the anti-interference device maps the low-dimensional three-channel characteristic dimension of the marked image to a high-dimensional space to increase image redundancy.
Preferably, the encoding device comprises a plurality of sequential convolution block execution streams, low-dimensional to high-dimensional watermark data features are extracted from the input watermarks in sequence, a feature vector set which is finally output is recorded as a feature space, content splicing is directly performed on vectors of the feature space and an original image, and then the watermark features and image contents are fused on a convolution block network of pre-training model parameters.
The application also relates to a method for protecting the content of a digital product, which comprises the following steps:
(1) generating a digital blind watermark according to the timestamp, the equipment code and the equipment signature;
(2) adding blind watermarks to the digital resources by using a digital image watermarking technology;
(3) the device submits the transaction with the digital blind watermark and the digital signature thereof to the block chain, the block chain confirms the transaction and the signature, packs the transaction information into the block and records the transaction information in the block chain according to the time sequence;
(4) when an infringing copy of the digital resource appears on the market, infringing tracking is performed through transaction records of the digital blind watermark in the infringing copy and the digital blind watermark on the blockchain system.
In the step (4), the method for performing infringement tracking includes: and extracting digital blind watermark information in the infringement copy, and inquiring the timestamp and the authorized equipment code of the digital blind watermark on the chain of the block chain system according to the code of the digital blind watermark to realize infringement tracking.
Wherein, in the step (4), the step of checking whether the digital blind watermark in the infringing copy is tampered or not is included; the checking step comprises the following steps: inquiring the distribution record of the digital resources on the block chain, inquiring the equipment authorization information, the identity authorization information, the distribution timestamp information and the digital signature information, comparing the equipment authorization information, the identity authorization information, the distribution timestamp information and the digital signature information with the digital watermark information, and checking whether the digital watermark is tampered; if the digital watermark has been tampered with, it is tracked again by an on-chain authorization record for the digital resource.
The watermark in the application is added by a resource owner, the fault-tolerant capability is improved by combining an anti-interference device, and the watermark has certain robustness and attack resistance, can refuse unauthorized access and tampering, so that the digital blind watermark technology can deeply combine the data recording and tracing capability of a block chain technology; the method and the device can realize the anti-counterfeiting of the digital product content and confirm that the provider of the digital product content is a real and reliable information provider; according to the method, the digital blind watermark is combined with the block chain through a digital signature mechanism, existence certification and non-falsification characteristics of the block chain, so that the digital watermark information is not chained to be falsified, infringement tracking information is guaranteed to be real and credible, and effective basis and powerful support are provided for supervision; digital content subject to copyright protection or other security requirements can be traced back to the source of the content leakage if the leakage occurs in a certain channel.
Drawings
Fig. 1 shows a schematic composition diagram of a digital blind watermark protection apparatus of the present application.
Fig. 2 shows a schematic diagram of an encoding apparatus of the present application.
Fig. 3 shows a schematic diagram of the interference rejection unit of the present application.
Detailed Description
To make the objects, technical solutions and advantages of the present application more apparent, embodiments of the present application will be described in detail below with reference to the accompanying drawings. It should be noted that the embodiments and features of the embodiments in the present application may be arbitrarily combined with each other without conflict.
The application provides a method and a system for protecting digital product content, which are based on a digital image blind watermark technology and a block chain technology and comprise two modules or processes of digital blind watermark adding of the digital product content and chain watermark generation, recording and tracing based on the block chain. The digital image blind watermarking technology is based on a deep neural network model and comprises a hardware image coding and decoding device and a matching method. More specifically, the method and the device use an AI model built in hardware to extract the content characteristics of the original image and fuse the watermark signatures to generate a marked image, and use a characteristic extraction model to extract a digital blind watermark signature from the marked image for matching and checking a watermark database on a chain.
Further, the AI model comprises an encoding module and a decoding module, wherein an additional anti-interference module is embedded between the encoding module and the decoding module and is used for improving the resistance of the marked image to attack in the transmission process and the fault-tolerant capability of watermark identification. In particular, the module also uses deep neural network configuration, namely the device adopts a uniform AI model mechanism to realize the digital blind watermarking technology. Preferably, the anti-interference module adopts a simple full-connection network to map the low-dimensional three-channel characteristic dimension of the marked image to a high-dimensional space. In particular, the dimension N is a configurable parameter (N is more than or equal to 3), and redundant information of the marked image is increased. The larger the value of N is, the higher the fault tolerance of the digital blind watermarking technology is represented, so that the robustness and the attack resistance of the watermarking technology are enhanced.
Furthermore, the encoding module and the decoding module are symmetrical structures and respectively comprise a network encoding structure and a network decoding structure. More specifically, the encoding and decoding structure is mainly divided into two steps:
(1) inputting the mapping of a digital blind watermark sequence space and a watermark characteristic space, and mapping an original watermark signature to a high-dimensional characteristic space by adopting a full convolution network through a network coding structure; the network decoding structure is opposite to the network decoding structure, and the high-dimensional feature space of the watermark is mapped to the original signature of the watermark for searching the chained digital blind watermark database.
(2) The network coding structure adopts a full convolution network to fuse the digital blind watermark characteristic vector and an original image to be marked to generate a marked image; the network decoding structure needs to analyze the corresponding digital blind watermark feature vector from the marked image.
In the present application, the complete flow of the hardware-based digital image watermarking processing method is as follows: the encoding network maps the input digital blind watermark to a high-dimensional feature space; the encoding network fuses the watermark characteristics with the original image to generate a marked image; optionally, the anti-interference layer maps the marked image to an N-dimensional space, so that the image redundancy is increased; extracting watermark characteristic vectors from the redundant marked images by a decoding network; and the decoding network restores the initial watermark signature from the watermark feature vector.
The protection system of digital product content according to the present application may be a digital blind watermark protection apparatus 100 as shown in fig. 1. As shown in fig. 1-3, the digital blind watermark protection apparatus 100 includes an encoding apparatus 101, a decoding apparatus 102, a convolution block execution kernel heap 103, and a model parameter memory 105. The encoding device 101 and the decoding device 102 are correspondingly arranged and are both connected with the convolution block execution nuclear heap 103 and used for executing data calling from the convolution block execution nuclear heap 103. The convolution block execution kernel heap 103 is connected with the model parameter memory 105, and the convolution block execution kernel heap 103 loads the parameters of the specified addresses in the model parameter memory according to the register parameters in the execution process, and performs calculation and outputs the result.
Specifically, the encoding device 101 is configured to encode an input image and a watermark input sequence, and generate a watermark image by fusion; the decoding device 102 restores an initial watermark signature from the noisy watermark image for the block chain management system to perform matching verification; the convolution block execution kernel heap 103 comprises a plurality of convolution block processors, each of which can independently configure parameters such as convolution kernel size and the like, read given input storage parameters, and execute data calculation in the actual neural network execution process in the encoding and decoding device; the memory 105 is a built-in integrated data memory of the embodiment of the watermark protection device, and is used for storing model parameters of a neural network. Optionally, the system further includes an anti-interference device 104, which is disposed between the encoding device 101 and the decoding device 102, and is capable of performing high-dimensional image redundancy processing on the watermark image, so as to increase robustness of the watermark system.
Specifically, the encoding apparatus 101 comprises 4 sequential rolling block execution streams, sequentially extracts the watermark data features from the low dimension to the high dimension from the input watermark, and finally outputs the feature space corresponding to the feature vector denoted as Wf. The vector of the feature space and the original image are directly subjected to content splicing (Concatanate operator), and then the watermark feature and the image content are fused on a convolution block network of the pre-training model parameters, so that the generated watermark image content is similar to the original image, and meanwhile, the implicit image watermark feature is added.
Specifically, the decoding apparatus 102 uses a sequential rolling block network to extract the watermark feature space WfThe vector of (2). In addition, in order to restore the dimension increasing operation of the anti-interference device, an additional convolution kernel channel dimension reducing operation is required to be carried out, and the feature vector is mapped to the feature space W matched with the original 3-dimensional imagef'; finally, the initial water is reversely restored by utilizing the operation of the rolling block network in the coding deviceAnd (6) printing a signature.
In particular, in the present embodiment, the rolling block execution core stack 103 is deployed on an FPGA or a neural network computing chip, and actually runs the encoding and decoding process. The convolution block execution core pile number is set to be N-4, parallel execution of encoding output and decoding output can be supported, and the processing performance of the device is improved under a single task program. And the convolution block loads the parameters of the designated address in the model parameter memory according to the register parameters in the execution process, and runs calculation and outputs the result.
In particular, the antijam device 104 generates a high-dimensional redundant watermark feature vector using a pre-trained fully-concatenated network. After the original marked watermark image is subjected to characteristic disturbance through network or physical propagation, part of blind watermark signature information can be lost. Therefore, the low-dimensional features are mapped to the high-dimensional space, the redundancy of the watermark information in the image can be increased, higher watermark fault tolerance is increased for the subsequent decoding device 102, and the attack resistance of the digital watermark is improved.
According to the method for generating, recording and tracing the digital blind watermark chain based on the block chain technology, a block chain system for digital blind watermark management is constructed by using the block chain technology, and watermark author information, generation time information and context information are recorded through block transaction, so that the method is beneficial to tracking the infringement behavior of digital resources.
More specifically, the present application relates to a method for protecting digital product content based on a block chain, which utilizes the above-mentioned digital blind watermark protection apparatus, and includes the following steps:
(1) generating a digital blind watermark according to the timestamp, the equipment code and the equipment signature;
(2) for digital resources, adding watermarks by using a digital image watermarking technology;
(3) the device submits the transaction with the digital blind watermark and the digital signature thereof to the block chain, the block chain confirms the transaction and the signature, packs the transaction information into the block and records the transaction information in the block chain according to the time sequence;
(4) when an infringing copy of a digital resource appears on the market, infringement tracking is performed through transaction records of a digital blind watermark in the infringing copy and a digital blind watermark on a blockchain system. If the infringement digital resource copy appears in the market, the copy inevitably has the embedded digital watermark, the digital blind watermark information in the copy is extracted, the timestamp and the authorized equipment code of the digital blind watermark are inquired on the chain of the block chain system according to the coding of the digital blind watermark, and the correlation analysis is carried out, so that the infringement tracking is realized. The user is a divulger of digital resources from which infringing resources flow. If the watermark image has been attacked, the proposed method is able to extract the watermark in most cases, unless the attack completely distorts the original digital resource.
In order to ensure the correctness of the digital blind watermark information, whether the digital blind watermark information is tampered needs to be verified, and the method comprises the following steps:
step 1: extracting a digital blind watermark from the infringed digital resource copy;
step 2: inquiring the distribution record of the digital resources on the block chain, inquiring the equipment authorization information, the identity authorization information, the distribution timestamp information and the digital signature information, comparing the equipment authorization information, the identity authorization information, the distribution timestamp information and the digital signature information with the digital watermark information, and checking whether the digital watermark is tampered; the digital watermark information and the distribution record can be combined to track the infringement behavior of the digital resources, and the transaction record of the digital resources on the block chain can help track the propagation of the digital resources;
and step 3: if the digital watermark is not tampered, the infringement tracking information is real and effective, and if the digital watermark is tampered, the digital watermark is tracked again through the chain authorization record of the digital resource.
The method and the system for protecting the digital product content based on artificial intelligence can solve the problem of pain point of current digital resource copyright protection and provide a real and credible basis for handling infringement disputes; the technology can realize the anti-counterfeiting of the digital product content and confirm that the provider of the digital product content is a real and reliable information provider; the digital blind watermark is combined with the block chain through the digital signature mechanism, the existence certification and the non-tampering characteristics of the block chain, so that the digital blind watermark information is not chained to be tampered, the infringement tracking information is guaranteed to be real and credible, and effective basis and powerful support are provided for supervision.
Although the embodiments disclosed in the present application are described above, the descriptions are only for the convenience of understanding the present application, and are not intended to limit the present application. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the disclosure as defined by the appended claims.
Claims (9)
1. A digital product content protection system based on artificial intelligence is characterized by comprising an encoding device, a decoding device, a convolution block execution kernel heap and a model parameter memory; the coding device and the decoding device are correspondingly arranged, are connected with the convolution block execution nuclear reactor and are used for executing data calling from the convolution block execution nuclear reactor; the convolution block execution kernel heap is connected with the model parameter memory, and the convolution block execution kernel heap loads the parameters of the specified addresses in the model parameter memory according to the register parameters in the execution process.
2. The system of claim 1, wherein the encoding device is configured to extract content features of an original image and fuse a watermark input sequence, and the signature generates a marked image; and the decoding device restores the initial watermark signature from the marked image by adopting a feature extraction model.
3. The digital production content protection system according to claim 1 or 2, further comprising a tamper resistant device disposed between the encoding device and the decoding device, the tamper resistant device mapping a low-dimensional three-channel feature dimension of the marker image to a high-dimensional space to increase image redundancy.
4. The system according to any one of claims 1-3, wherein the encoding means comprises a plurality of sequential convolutional block execution streams, sequentially extracting low-dimensional to high-dimensional watermark data features from the input watermark, recording the final output feature vector set as a feature space, directly performing content stitching on the vectors of the feature space and the original image, and then fusing the watermark features and the image content on a convolutional block network of pre-trained model parameters.
5. A method for protecting digital product contents using a digital product contents protection system according to any one of claims 1 to 4, comprising the steps of:
(1) generating a digital blind watermark according to the timestamp, the equipment code and the equipment signature;
(2) adding blind watermarks to the digital resources by using a digital image watermarking technology;
(3) the device submits the transaction with the digital blind watermark and the digital signature thereof to the block chain, the block chain confirms the transaction and the signature, packs the transaction information into the block and records the transaction information in the block chain according to the time sequence;
(4) when an infringing copy of the digital resource appears on the market, infringing tracking is performed through transaction records of the digital blind watermark in the infringing copy and the digital blind watermark on the blockchain system.
6. The protection method according to claim 5, wherein in the step (4), the infringement tracing is performed by: and extracting digital blind watermark information in the infringement copy, and inquiring the timestamp and the authorized equipment code of the digital blind watermark on the chain of the block chain system according to the code of the digital blind watermark to realize infringement tracking.
7. Protection method according to claim 5 or 6, characterized in that in step (4) it comprises the step of checking whether the digital blind watermark in the infringing copy has been tampered with.
8. The protection method according to claim 7, wherein the step of verifying comprises: inquiring the distribution record of the digital resources on the block chain, inquiring the equipment authorization information, the identity authorization information, the distribution timestamp information and the digital signature information, comparing the equipment authorization information, the identity authorization information, the distribution timestamp information and the digital signature information with the digital watermark information, and checking whether the digital watermark is tampered.
9. Protection method in accordance with claim 8 characterized in that if the digital watermark has been tampered with, it is tracked again by an on-chain authorization record of the digital resource.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011543482.7A CN112579994A (en) | 2020-12-23 | 2020-12-23 | Digital product content protection system and method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011543482.7A CN112579994A (en) | 2020-12-23 | 2020-12-23 | Digital product content protection system and method based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112579994A true CN112579994A (en) | 2021-03-30 |
Family
ID=75139169
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011543482.7A Pending CN112579994A (en) | 2020-12-23 | 2020-12-23 | Digital product content protection system and method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112579994A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116956244A (en) * | 2023-09-18 | 2023-10-27 | 翼方健数(北京)信息科技有限公司 | Copyright protection method and system for AIGC text authored content |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1999001980A1 (en) * | 1997-07-02 | 1999-01-14 | Kowa Co., Ltd. | Electronic watermarking method and method of extracting watermark information |
| US5889868A (en) * | 1996-07-02 | 1999-03-30 | The Dice Company | Optimization methods for the insertion, protection, and detection of digital watermarks in digitized data |
| CN101093576A (en) * | 2007-06-06 | 2007-12-26 | 北京理工大学 | Digital watermark method of serial multiple robustnesses based on multiple copyright authentications |
| CN101661610A (en) * | 2009-09-17 | 2010-03-03 | 上海交通大学 | Digital image anti-fake detecting system with active and passive identification functions |
| CN101901470A (en) * | 2010-02-10 | 2010-12-01 | 桂林电子科技大学 | Image-tampering detection and recovery method based on energy-domain semi-fragile watermarking |
| CN103310403A (en) * | 2013-05-14 | 2013-09-18 | 深圳市威富多媒体有限公司 | Handwriting characteristic and digital document visible watermark fusion method |
| CN103795891A (en) * | 2014-03-04 | 2014-05-14 | 山东科技大学 | Method for coding, embedding and decoding of watermark resistant to range zooming attack |
| CN108449527A (en) * | 2018-03-28 | 2018-08-24 | 南京吉印信息科技有限公司 | A kind of digital watermark method and system for taking the photograph attack for the anti-screen of remote sensing images |
| CN110211016A (en) * | 2018-02-28 | 2019-09-06 | 佛山科学技术学院 | A kind of watermark embedding method based on convolution feature |
| US20190318067A1 (en) * | 2016-12-07 | 2019-10-17 | Lamark | Method for recording a multimedia content, method for detecting a watermark within a multimedia content, corresponding devices and computer programs |
| CN111090837A (en) * | 2019-12-06 | 2020-05-01 | 上海科乐宜信息科技有限公司 | Intelligent tracking method based on block chain technology |
| CN111145069A (en) * | 2019-12-03 | 2020-05-12 | 支付宝(杭州)信息技术有限公司 | Image watermarking processing method and device based on block chain |
| CN111223034A (en) * | 2019-11-14 | 2020-06-02 | 中山大学 | High-capacity printing/shooting resistant blind watermark system and method based on deep learning |
| CN111491170A (en) * | 2019-01-26 | 2020-08-04 | 华为技术有限公司 | Method for embedding watermark and watermark embedding device |
| CN111784556A (en) * | 2020-06-23 | 2020-10-16 | 中国平安人寿保险股份有限公司 | Method, device, terminal and storage medium for adding digital watermark in image |
| CN111798360A (en) * | 2020-06-30 | 2020-10-20 | 百度在线网络技术（北京）有限公司 | Watermark detection method, watermark detection device, electronic equipment and storage medium |
| CN112003888A (en) * | 2020-07-09 | 2020-11-27 | 深圳市雄帝科技股份有限公司 | Block chain-based certificate management method, device, equipment and readable medium |
| CN114065150A (en) * | 2021-11-22 | 2022-02-18 | 安徽大学 | Picture copyright protection method |
-
2020
- 2020-12-23 CN CN202011543482.7A patent/CN112579994A/en active Pending
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5889868A (en) * | 1996-07-02 | 1999-03-30 | The Dice Company | Optimization methods for the insertion, protection, and detection of digital watermarks in digitized data |
| WO1999001980A1 (en) * | 1997-07-02 | 1999-01-14 | Kowa Co., Ltd. | Electronic watermarking method and method of extracting watermark information |
| CN101093576A (en) * | 2007-06-06 | 2007-12-26 | 北京理工大学 | Digital watermark method of serial multiple robustnesses based on multiple copyright authentications |
| CN101661610A (en) * | 2009-09-17 | 2010-03-03 | 上海交通大学 | Digital image anti-fake detecting system with active and passive identification functions |
| CN101901470A (en) * | 2010-02-10 | 2010-12-01 | 桂林电子科技大学 | Image-tampering detection and recovery method based on energy-domain semi-fragile watermarking |
| CN103310403A (en) * | 2013-05-14 | 2013-09-18 | 深圳市威富多媒体有限公司 | Handwriting characteristic and digital document visible watermark fusion method |
| CN103795891A (en) * | 2014-03-04 | 2014-05-14 | 山东科技大学 | Method for coding, embedding and decoding of watermark resistant to range zooming attack |
| US20190318067A1 (en) * | 2016-12-07 | 2019-10-17 | Lamark | Method for recording a multimedia content, method for detecting a watermark within a multimedia content, corresponding devices and computer programs |
| CN110211016A (en) * | 2018-02-28 | 2019-09-06 | 佛山科学技术学院 | A kind of watermark embedding method based on convolution feature |
| CN108449527A (en) * | 2018-03-28 | 2018-08-24 | 南京吉印信息科技有限公司 | A kind of digital watermark method and system for taking the photograph attack for the anti-screen of remote sensing images |
| CN111491170A (en) * | 2019-01-26 | 2020-08-04 | 华为技术有限公司 | Method for embedding watermark and watermark embedding device |
| CN111223034A (en) * | 2019-11-14 | 2020-06-02 | 中山大学 | High-capacity printing/shooting resistant blind watermark system and method based on deep learning |
| CN111145069A (en) * | 2019-12-03 | 2020-05-12 | 支付宝(杭州)信息技术有限公司 | Image watermarking processing method and device based on block chain |
| CN111090837A (en) * | 2019-12-06 | 2020-05-01 | 上海科乐宜信息科技有限公司 | Intelligent tracking method based on block chain technology |
| CN111784556A (en) * | 2020-06-23 | 2020-10-16 | 中国平安人寿保险股份有限公司 | Method, device, terminal and storage medium for adding digital watermark in image |
| CN111798360A (en) * | 2020-06-30 | 2020-10-20 | 百度在线网络技术（北京）有限公司 | Watermark detection method, watermark detection device, electronic equipment and storage medium |
| CN112003888A (en) * | 2020-07-09 | 2020-11-27 | 深圳市雄帝科技股份有限公司 | Block chain-based certificate management method, device, equipment and readable medium |
| CN114065150A (en) * | 2021-11-22 | 2022-02-18 | 安徽大学 | Picture copyright protection method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116956244A (en) * | 2023-09-18 | 2023-10-27 | 翼方健数(北京)信息科技有限公司 | Copyright protection method and system for AIGC text authored content |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110110500B (en) | Decentralized image copyright protection system and method with immediate infringement detection function | |
| Xue et al. | Intellectual property protection for deep learning models: Taxonomy, methods, attacks, and evaluations | |
| Bartolini et al. | Image authentication techniques for surveillance applications | |
| Zhu et al. | Secure neural network watermarking protocol against forging attack | |
| Wang et al. | Image copyright protection based on blockchain and zero-watermark | |
| Al-Qershi et al. | Evaluation of copy-move forgery detection: datasets and evaluation metrics | |
| Jaiswal et al. | Multimedia semantic integrity assessment using joint embedding of images and text | |
| Nabi et al. | A comprehensive survey of image and video forgery techniques: variants, challenges, and future directions | |
| Asghar et al. | Edge–texture feature-based image forgery detection with cross-dataset evaluation | |
| CN109785222A (en) | A method of the information for webpage is quickly embedded in and extracts | |
| Wu et al. | A novel zero-watermarking scheme based on NSCT-SVD and blockchain for video copyright | |
| CN109040760A (en) | The guard method of network image copyright information, device and storage medium | |
| Wang et al. | A blockchain-based system for secure image protection using zero-watermark | |
| Lederer et al. | Identifying appropriate intellectual property protection mechanisms for machine learning models: A systematization of watermarking, fingerprinting, model access, and attacks | |
| CN112579994A (en) | Digital product content protection system and method based on artificial intelligence | |
| Verma et al. | Survey on image copy-move forgery detection | |
| Yao et al. | A secure image evidence management framework using multi-bits watermark and blockchain in IoT environments | |
| Maji et al. | Image tampering issues in social media with proper detection | |
| Meng et al. | Traceable and authenticable image tagging for fake news detection | |
| Li | Handbook of Research on Computational Forensics, Digital Crime, and Investigation: Methods and Solutions: Methods and Solutions | |
| CN103177219A (en) | Medical biochemical detection report critical data tamper-proofing method | |
| Wen et al. | On Function-Coupled Watermarks for Deep Neural Networks | |
| Huang et al. | Convolutional neural networks tamper detection and location based on fragile watermarking | |
| Liu et al. | BiFPro: A Bidirectional Facial-data Protection Framework against DeepFake | |
| Tyagi et al. | ForensicNet: Modern convolutional neural network‐based image forgery detection network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |