Offline seal anti-counterfeiting system based on dark watermark technology

Through the dark watermarking technology that embeds document information in the seal image, the existing seal anti-counterfeiting technology relies on network verification and the risk of forgery, which has effectively achieved efficient and secure verification of offline seals, ensuring document authenticity and anti-counterfeiting capabilities. .

CN119089414AInactive Publication Date: 2024-12-06姚乐天
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411112823.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing anti-counterfeiting technology of seals relies on network verification, poses a risk of forgery, and cannot effectively prevent the copying of seals, resulting in frequent forgery incidents.

Method used

The offline seal anti-counterfeiting system based on dark watermark technology is adopted. Through the seal generation and identification system, deep learning and cryptography are used to embed document information in the seal image to generate and verify the authenticity of documents without network connection, ensuring that documents cannot be forged.

Benefits of technology

It realizes the simplicity, speed, economical and safe verification of document authenticity, eliminates the occurrence of seal forgery incidents, and effectively maintains social and economic stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119089414A_ABST
    Figure CN119089414A_ABST
Patent Text Reader

Abstract

The invention discloses an off-line seal anti-counterfeiting system based on a dark watermark technology, and belongs to the field of information security. The system is based on the dark watermark technology, the password use technology, the information coding technology, the deep learning technology and other technologies, network connection and network server establishment are not needed, the use mode of an existing seal is basically not changed, it is technically guaranteed that a document with a seal cannot be forged, and the authenticity verification process is simple and economical.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of information security and relates to an off-line seal anti-counterfeiting system based on dark watermark technology. Background Art

[0002] Many important documents, such as various certificates, licenses, contracts, etc., use seals as proof of the authenticity of the document. However, with the development of information technology, seals currently only have a legal deterrent effect, because almost everyone can master the technology to copy seals. If the benefits of copying seals are large enough, even if copying seals involves criminal offenses, it cannot prevent some people from taking risks. In recent years, a large number of criminal offenses involving forged seals have occurred.

[0003] At present, one type of seal or document anti-counterfeiting technology uses a website to store document information and verifies authenticity by querying it online. This method actually abandons the anti-counterfeiting function of the seal, and the physical document is only used as an index for querying the website. For example, China Higher Education Student Information and Career Center stores real degree certificate information for query and verification. This method needs to ensure the authority of the website information and the consistency of the file information with the website information.

[0004] The present invention is based on dark watermark technology, using technologies such as passwords, information coding and deep learning. It does not require the establishment of network connections and network servers, and basically does not change the existing way of using seals. It technically ensures that documents with seals cannot be forged, and the process of verifying authenticity is simple, economical, fast and convenient, thereby preventing the occurrence of such counterfeit seal incidents. It can be used for the anti-counterfeiting of certificates, contracts, seals, official documents, invoices, identity documents, etc., and is of great significance for maintaining social and economic stability and eliminating economic and fraud crimes. Summary of the invention

[0005] The invention provides a safe off-line seal anti-counterfeiting system based on dark watermark technology, which can determine the authenticity of a document only through the image of the document with a seal.

[0006] The technical solution adopted by the present invention is as follows:

[0007] An off-line seal anti-counterfeiting system, comprising a seal generating system and a seal counterfeiting identifying system;

[0008] The seal generation system can be run on a host or a mobile phone. The seal generation process is as follows:

[0009] Step 1-1: The user scans or takes a photo of the document that needs to be stamped or anti-counterfeited to obtain an image of the document that needs to be stamped or anti-counterfeited.

[0010] Step 1-2: Display the document image on your phone or computer, and the user can select the seal to be stamped and the location of the seal in the document. Try not to cover the text in the document.

[0011] Step 1-3: Obtain a feature vector of a preset dimension related to the content information of the document image. There are two methods for obtaining the feature vector: the first is to directly encode the document image; the second is to first use the OCR technology to obtain the text of the document image, and then encode the text. The encoding method of the above two methods for obtaining feature vectors can use the image and text encoding technology in deep learning, that is, the technology of using a neural network to encode an image or text into a vector of a fixed dimension, and the distance between the vectors encoded by similar images or texts is also small. The image encoding method can also use image hashing technology. If the image is encoded directly, it is necessary to minimize the difference between the image before and after stamping or processing. The method of filling the background color at the stamp position or pre-adding the original seal image can be used. If the method of filling the background color at the stamp position is used, the same processing needs to be done in steps 2-4 of seal recognition. If the method of pre-adding the original seal image is used, in steps 1-5, the generated seal image with embedded watermark will replace the pre-added original seal image.

[0012] Step 1-4: Use dark watermarking technology to embed the vector obtained in step 1-3 into a part of the seal image or document image, and obtain a watermarked seal image or a part of the document image embedded with document information. Dark watermarking technology can use deep learning technology, that is, input the image and watermark information into a deep network. The deep network is divided into two parts: an encoding network and a decoding network. The encoding network fuses the image and the watermark information to generate an image with watermark information. The generated image is difficult to see the difference from the input image with the naked eye. The decoding network can correctly decode the input watermark information from the watermarked image.

[0013] Step 1-5: Print the seal or partial document image embedded with the watermark to a specified position in the document to obtain a printed anti-counterfeiting paper document.

[0014] The seal authentication system can be run on a host or mobile phone. The process of identifying the authenticity of a document is as follows:

[0015] Step 2-1: The user scans or takes a photo of the document to be authenticated to obtain an image of the document to be authenticated.

[0016] Step 2-2: Locate the stamp or part of the document image with the dark watermark.

[0017] Step 2-3: Use the dark watermark decoding function to decode the seal or partial document image obtained in step 2-2 to extract the watermark information, that is, a vector of a preset dimension.

[0018] Step 2-4: Obtain the image or text features of the document and encode them. This process is the same as steps 1-3 in the seal generation process, and obtains the encoding vector related to the document content.

[0019] Step 2-5: Compare the similarity of the vectors obtained in step 2-3 and step 2-4. If the similarity is greater than a preset threshold, the document is determined to be true, otherwise the document is determined to be false. The vector similarity measurement can be measured by cosine similarity or Hamming distance.

[0020] After the present invention is encoded, an encryption process can be added, and the encryption can use a symmetric or asymmetric encryption algorithm. Correspondingly, in steps 2-3 of the seal authentication process, the obtained watermark information needs to be decrypted, and the key needs to use the key of the corresponding encryption process. Adding this encryption and decryption process can authenticate or control the user of stamping and authentication, and can also increase the difficulty of cracking the entire system. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Flowchart of stamp generation.

[0022] Figure 2 Flowchart for seal authentication. DETAILED DESCRIPTION

[0023] The specific embodiments of the present invention are described in detail below in combination with the technical solutions and the accompanying drawings.

[0024] Example 1

[0025] As shown in the accompanying drawings, the main steps include:

[0026] 1. When a user wants to stamp an anti-counterfeiting stamp on a document, he first scans the document through a scanner, and the document image is obtained by the stamp generation system.

[0027] 2. The seal generation system displays the document image to the user, and the user selects the seal and the location of the seal using the mouse.

[0028] 3. The seal generation system uses the OCR function to obtain the text information of the document and encodes the text information into a sequence of 0s and 1s.

[0029] 4. The seal generation system uses the private key in the asymmetric key to encrypt the 0, 1 sequence to obtain an encrypted 0, 1 sequence.

[0030] 5. The seal generation system uses the dark watermark function to embed the encrypted 0, 1 sequence into the image of the seal selected by the user, and prints the seal image to the user-specified location.

[0031] Seal identification process:

[0032] 6. The user scans or takes a photo of the document to be authenticated to obtain a document image.

[0033] 7. After obtaining the image, the seal authentication system performs recognition to obtain the seal image.

[0034] 8. The seal authentication system decodes the dark watermark of the seal image to obtain an encrypted 0, 1 sequence, and then uses the public key in the asymmetric key to decrypt it to obtain a 0, 1 sequence representing the text features of the document.

[0035] 9. Use the same method as step 3 to obtain the document’s text feature encoding 0, 1 sequence.

[0036] 10. Compare the similarities of the 0 and 1 sequences obtained in steps 8 and 9. If the similarity is greater than the preset threshold, the document is judged to be true, otherwise the document is judged to be false.

[0037] Embodiment 2:

[0038] 1. If a user wants to perform anti-counterfeiting processing on a document without stamping it, he should first scan or take a photo of the document through a scanner, and the document image will be obtained by the stamp generation system.

[0039] 2. The seal generation system displays the document image to the user, and the user uses the mouse to select the location to add the anti-counterfeiting image. The anti-counterfeiting image can be any image selected by the user, such as a company logo or trademark.

[0040] 3. The seal generation system fills the area that needs to be processed with the background color, and then uses the image hash function to obtain the hash value of the document image, which is usually a sequence of 0, 1.

[0041] 4. The seal generation system uses the private key in the asymmetric key to encrypt the 0, 1 sequence to obtain an encrypted 0, 1 sequence.

[0042] 5. The seal generation system uses the dark watermark function to embed the encrypted 0, 1 sequence into the anti-counterfeiting image selected by the user, prints the anti-counterfeiting image to the user-specified location, and obtains a paper document that has been anti-counterfeited.

[0043] Seal identification process:

[0044] 6. The user scans or takes a photo of the document to be authenticated to obtain a document image.

[0045] 7. After the seal identification system obtains the image, it performs recognition and obtains the location of the anti-counterfeiting image in the document image. The content of the anti-counterfeiting image is generally known in advance, so it is relatively easy to locate.

[0046] 8. The seal identification system decodes the dark watermark of the anti-counterfeiting image to obtain an encrypted sequence of 0s and 1s, and then uses the public key in the asymmetric key to decrypt it to obtain the hash value of the document image.

[0047] 9. The seal identification system fills the image with the background color, and then uses the same method as step 3 to get the hash value of the document image.

[0048] 10. Compare the similarity of the hash values ​​obtained in step 8 and 9. If the similarity is greater than the preset threshold, the document is determined to be true, otherwise the document is determined to be false.

Claims

1. An off-line seal anti-counterfeiting system, characterized in that: Including seal generation system and seal authentication system; The process of stamp generation is as follows: Step 1-1: The user scans or takes a photo of the document to be stamped or anti-counterfeited to obtain an image of the document to be stamped or anti-counterfeited; Step 1-2: Display the document image on the mobile phone or computer, and the user can select the seal to be stamped and the position of the seal in the document; the selected position should not cover the text in the document as much as possible; Step 1-3: Obtain a feature vector of a preset dimension related to the content information of the document image; there are two methods for obtaining the feature vector: The first one is to encode the document image directly; The second method is to first use OCR technology to obtain the text of the document image, and then encode the text; the encoding method of the above two methods of obtaining feature vectors can use the image and text encoding technology in deep learning, that is: the technology of using neural networks to encode images or texts into vectors of fixed dimensions, and the distance between vectors encoded by similar images or texts is also small; the image encoding method can also use image hashing technology; if the image is encoded directly, it is necessary to minimize the difference between the image before and after stamping or processing, and the method of filling the background color at the stamping position or pre-adding the original stamp image can be adopted; If the method of filling the background color at the stamp position is adopted, the same processing needs to be done in steps 2-4 of stamp recognition; if the method of pre-adding the original stamp image is adopted, in steps 1-5, the generated stamp image embedded with watermark will replace the pre-added original stamp image; Step 1-4: Embed the vector obtained in step 1-3 into a part of the seal image or document image using dark watermark technology to obtain a seal image or a part of the document image with a watermark embedded in the document information; dark watermark technology can use deep learning technology, that is, input the image and watermark information into a deep network, the deep network is divided into two parts: an encoding network and a decoding network, the encoding network fuses the image and the watermark information to generate an image with watermark information, the generated image is difficult to see the difference with the input image with the naked eye, and the decoding network can correctly decode the input watermark information from the watermarked image; Step 1-5: Print the seal or partial document image embedded with the watermark to a specified position in the document to obtain a printed anti-counterfeiting paper document; The process of the seal authentication system to identify the authenticity of documents is as follows: Step 2-1: The user scans or takes a photo of the document to be authenticated to obtain an image of the document to be authenticated; Step 2-2: Locate the seal or part of the document image with the dark watermark; Step 2-3: Use the dark watermark decoding function to decode the seal or partial document image obtained in step 2-2 to extract the watermark information, that is, a vector of a preset dimension; Step 2-4: Obtain the image or text features of the document and encode them. This process is the same as steps 1-3 in the seal generation process, and obtain the encoding vector related to the document content; Step 2-5: Compare the similarity of the vectors obtained in step 2-3 and step 2-4; if the similarity is greater than a preset threshold, the document is determined to be true, otherwise the document is determined to be false; the vector similarity measurement can be measured by cosine similarity or Hamming distance.