Blind watermark decryption implementation method based on screenshot information protection

By using Photoshop's RGB channel and Canvas technology to decrypt blind watermarks in screenshot information, the problem of easy watermark removal and high code complexity in existing traceability solutions is solved, and effective traceability and sensitive information protection of screenshot information is achieved.

CN118917992APending Publication Date: 2024-11-08JIANGSU SUNING BANK CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410921517.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-10
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

In the existing traceability solution, document watermarks are easily discovered and removed by users, and cannot be effectively traced in new leak scenarios. The frequency domain transformation method has a high code complexity and high development difficulty, which limits large-scale applications.

Method used

Using a blind watermark decryption method based on screenshots, the screenshots are processed using Photoshop's RGB channel and contrast curve tool, pixel redrawing is performed through Canvas technology, and RGB channel configuration and contrast curve are fitted with segmented functions to achieve inverse display of blind watermarks and Decrypt.

Benefits of technology

It realizes effective traceability of screenshot information, and can identify blind watermark information on PC, mini-programs and client pages. It is suitable for sensitive information protection and infringement traceability positioning, reducing development complexity and time.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118917992A_ABST
    Figure CN118917992A_ABST
Patent Text Reader

Abstract

The invention discloses a blind watermark decryption implementation method based on screenshot information protection. The method comprises the following steps: acquiring a page screenshot encrypted by using a blind watermark; processing the page screenshot by using an RGB (Red, Green, Blue) channel and a contrast curve tool in Photoshop so as to obtain a watermark inverse display image of the page screenshot; using a piecewise function to fit RGB channel configuration and a contrast curve of the definition of the watermark inverted image within a set range; performing pixel point redrawing on the original image by using a Canvas technology according to the fitted RGB channel configuration and the contrast curve so as to obtain blind watermark inverted display images under different RGB channels and contrast curves; and decrypting the blind watermark information displayed in the blind watermark reverse display image according to a preset encryption scheme to obtain watermark traceability information. According to the invention, blind watermark inversion of the RGB channel of the image is realized based on the Canvas technology, so that traceability and troubleshooting of leaked information are realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of information protection and data processing, and in particular to a method for realizing blind watermark decryption based on screenshot information protection. Background Art

[0002] With the rapid development of computer hardware and software technology and network technology, higher requirements have been put forward for the data security of information systems. Traditional ways of leakage are mainly through printing, copying and copying of electronic documents, online transmission through the network, etc. In view of these leakage ways, physical means such as limiting data copying, limiting document transmission, and limiting network access can be used to prevent sensitive data leakage.

[0003] With the advent of the 5G era and the popularization of smart handheld devices and smart wearable device hardware, new information leakage channels have been added, such as taking photos and screenshots. In order to deal with the above new information leakage, a blind watermark function can be added to the page or system. Once the sensitive information of a company or customer is leaked, the source of the leak can be traced back based on the leaked image, and the source of the leak can be traced and blocked.

[0004] Blind watermarking can make it unaware of users and leakers, and can more effectively obtain the source of leaking users.

[0005] At present, the following defects or deficiencies are common in the industry in achieving data traceability:

[0006] (1) The existing traceability solution uses a document watermark, which is easy for users to see. The embedded watermark can be removed with one click using advanced Photoshop technology or other photo beautification software, making it impossible for the watermark to perform its original traceability function.

[0007] (2) The existing method uses Fourier transform to perform frequency domain transformation on the image file and watermark file respectively, obtains the frequency domain information of the image and the frequency domain information of the watermark, superimposes the two information in the frequency domain, and then converts them into time domain images through inverse Fourier transform. This method can effectively deal with the reverse decoding and tracing of operations such as transferring, forwarding, cropping, and modifying images. The limitations of this solution are: 1. The original image processed by the user must be obtained before the frequency domain transformation can be performed to reversely find the watermark information; for the new leakage scenario mentioned above, this solution cannot complete the traceability. 2. The page content needs to contain the image before the watermark operation can be performed on the included image, which has great limitations; 3. For all images with watermarks, a time domain-frequency domain-time domain conversion must be performed, which has high code complexity, high development difficulty, and long conversion time, which is not conducive to large-scale application in production environments. Summary of the invention

[0008] The purpose of the present invention is to provide a method for implementing blind watermark decryption based on screenshot information protection in view of the deficiencies in the prior art.

[0009] To achieve the above object, the present invention provides a method for implementing blind watermark decryption based on screenshot information protection, comprising:

[0010] Get a screenshot of the page encrypted with a blind watermark;

[0011] Processing the page screenshot using the RGB channel and contrast curve tool in Photoshop to obtain a watermark-reflected image of the page screenshot;

[0012] Using piecewise functions to fit multiple groups of RGB channel configurations and contrast curves of the watermark reversed image within a set range of clarity;

[0013] When blind watermark decryption is performed on other page screenshots obtained later, pixels of other page screenshots are redrawn using Canvas technology according to the fitted multiple groups of RGB channel configurations and contrast curves to obtain blind watermark reversed images under different RGB channels and contrasts;

[0014] According to the preset encryption scheme, the blind watermark information displayed in the blind watermark reversed image under different RGB channels and contrast curves is decrypted to obtain the watermark traceability information.

[0015] Furthermore, the method for obtaining the page screenshot encrypted with the blind watermark includes uploading a local image, identifying an online image, and identifying a specified online page through the URL address of the page.

[0016] Furthermore, the decryption method includes symmetric or asymmetric decryption, and the number of decryption times includes single decryption and multiple decryptions.

[0017] Beneficial effects: The present invention realizes the blind watermark reverse display of the RGB channel of the image based on Canvas technology, thereby realizing the tracing and investigation of leaked information; the present invention can effectively identify blind watermark information on PC pages, mini-program pages and client native pages, and can be widely used in company asset systems, sensitive information page systems and personal intellectual property information systems, such as: customer service complaint systems, information review systems, business approval systems, etc., to realize sensitive information protection and infringement tracing and positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a screenshot of the page encrypted with a blind watermark;

[0019] Figure 2 It is a watermark reversed image obtained by a blind watermark decryption implementation method based on screenshot information protection;

[0020] Figure 3 It is the second watermark reversed image obtained by the blind watermark decryption implementation method based on screenshot information protection;

[0021] Figure 4 It is the third watermark reversed image obtained by the blind watermark decryption implementation method based on screenshot information protection. DETAILED DESCRIPTION

[0022] The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments. The embodiments are implemented based on the technical solution of the present invention. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0023] like Figures 1 to 4 As shown, an embodiment of the present invention provides a method for implementing blind watermark decryption based on screenshot information protection, including:

[0024] Obtain a screenshot of the page encrypted with a blind watermark. Specifically, the method of obtaining the screenshot of the page encrypted with a blind watermark includes uploading a local image, identifying an online image, and identifying a specified online page through the URL address of the page.

[0025] The page screenshot is processed using the RGB channel and contrast curve tools in Photoshop to obtain a watermark-reverse image of the page screenshot.

[0026] A piecewise function is used to fit multiple sets of RGB channel configurations and contrast curves whose clarity of the watermark reversed image is within a set range. Specifically, under the configuration of the above multiple sets of RGB channel configurations and contrast curves, the blind watermark in the watermark reversed image should be clearly visible.

[0027] When blind watermark decryption is performed on other page screenshots obtained later, the pixels of other page screenshots are redrawn using Canvas technology according to the fitted multiple sets of RGB channel configurations and contrast curves to obtain blind watermark reverse images under different RGB channels and contrasts. In the Javascript programming language, Canvas can process images at the pixel level. Through simple algorithms, operations such as pixel inversion, black and white effects, brightness, contrast, and retro effects can be achieved on images. Canvas provides the following methods:

[0028] cxt.getImageData(x,y,width,height)

[0029] x, y, width, height represent the horizontal coordinate, vertical coordinate, width and height of the selected area respectively. This method returns a canvasPixelArray object, which has a data attribute, which stores all the pixel data of an image.

[0030] const imageData=cxt.getImageData(x,y,width,height)

[0031] const data = imageData.data

[0032] The data format in the data attribute is as follows:

[0033] [r1, g1, b1, a1, r2, g2, b2, a2, r3, g3, b3, a3...] It can be seen that the array stores RGBA colors, and every four elements are a group of RGBA, which is all the information of a pixel.

[0034] The total number of pixels of an image can be obtained through cxt.getImageData(x,y,width,height).data.length.

[0035] Canvas also provides the following methods:

[0036] cxt.putImageData(image,x,y)

[0037] This method is used to output the pixels of an image to display an image. The image, x, and y of this method represent the canvasPixelArray object, the horizontal coordinate of the redrawn image, and the vertical coordinate of the redrawn image, respectively. This method is usually used in conjunction with getImageData().

[0038] According to the preset encryption scheme, the blind watermark information displayed in the blind watermark reversed image under different RGB channels and contrast curves is decrypted to obtain the watermark traceability information. The above decryption method includes symmetric or asymmetric decryption, and the number of decryption includes one decryption and multiple decryptions, which depends on the encryption method.

[0039] Taking a local picture as an example, the watermark inversion of this solution is demonstrated. The code snippet of the main watermark inversion function is as follows:

[0040] let canvas=document.getElementById('mycanvas');

[0041] let cxt=canvas.getContext("2d");

[0042] let filter=document.getElementById('filter');

[0043] let filter_cxt=filter.getContext("2d");

[0044] document.querySelector("#File").onchange=function(){

[0045] var Reader=new FileReader();

[0046] Reader.readAsDataURL(this.files[0]);

[0047] Reader.onload = function () {

[0048] let img = new Image();

[0049] img.src = Reader.result;

[0050] img.onload = function() {

[0051] cxt.drawImage(img,0,0,canvas.width,canvas.height);

[0052]

[0053] The color value of each RGB channel is 0-255. In the above example, when the color value of the red channel and the green channel is less than 220, the value is 0. When the value is greater than 220, the value is a function of 8*data[i]–1785, and the piecewise function is used to approximate the original contrast curve. The above data[i] refers to the pixel data of the image. The blue channel value is the weighted value of the RGB channel:

[0054] data[i]*0.01+data[i+1]*0.01+data[i+2]*0.98

[0055] Figure 1 This is a screenshot of the page encrypted with a blind watermark. Finally, the watermark reverse effect is obtained according to the above scheme. Figure 2The present invention can display the page blind watermark more clearly and has a better positioning effect on leak tracing.

[0056] You can also keep the blue channel (B channel) unchanged and only change the other two channels (RG channels). The main part of the blind watermark reverse function is adjusted as follows:

[0057] data[i]=data[i]<240?0:16.42*data[i]-3947

[0058] data[i+1]=data[i+1]<240?0:16.42*data[i]-3947

[0059] Finally, the watermark reverse display effect is obtained according to the above scheme. Figure 3 As shown in the figure. When the blue channel (B channel) remains unchanged and only the other two channels (RG channels) are changed, the blind watermark effect can also be identified under the condition that the piecewise fitting function is selected to be relatively "smooth". However, the blind watermark reverse display effect is poor at this time, so it can be used as some backup solutions.

[0060] While keeping the red channel (R channel) unchanged and only changing the other two channels (GB channels), the main part of the blind watermark reverse function is adjusted as follows:

[0061] data[i+1]=data[i+1]<250?0:51*data[i+1]-12750

[0062] data[i+2]=data[i+2]<250?0:51*data[i+2]–12750

[0063] Finally, the watermark reverse display effect is obtained according to the above scheme. Figure 4 As shown in the figure, the red channel (R channel) remains unchanged, only the other two channels (GB channels) are changed, and the piecewise function is selected to be more "steep", then the blind watermark reverse display effect is better.

[0064] It can be seen that the watermark recognition effect of the blind watermark inversion scheme under the channel weighted hybrid mode is better than that of the dual-channel watermark inversion scheme; by comparing and analyzing the latter two experimental schemes, it is found that the steeper the piecewise fitting function, the more obvious the recognition effect of the blind watermark.

[0065] The above is only a preferred embodiment of the present invention. It should be noted that for ordinary technicians in this technical field, other parts not specifically described belong to the prior art or common knowledge. Several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered as the protection scope of the present invention.

Claims

1. A method for implementing blind watermark decryption based on screenshot information protection, characterized in that: include: Get a screenshot of the page encrypted with a blind watermark; Processing the page screenshot using the RGB channel and contrast curve tool in Photoshop to obtain a watermark-reflected image of the page screenshot; Using piecewise functions to fit multiple groups of RGB channel configurations and contrast curves of the watermark reversed image within a set range of clarity; When blind watermark decryption is performed on other page screenshots obtained later, pixels of other page screenshots are redrawn using Canvas technology according to the fitted multiple groups of RGB channel configurations and contrast curves to obtain blind watermark reversed images under different RGB channels and contrasts; According to the preset encryption scheme, the blind watermark information displayed in the blind watermark reversed image under different RGB channels and contrast curves is decrypted to obtain the watermark traceability information.

2. According to claim 1, a method for implementing blind watermark decryption based on screenshot information protection is characterized in that: The method for obtaining the page screenshot encrypted with the blind watermark includes uploading a local picture, identifying an online picture, and identifying a designated online page through the URL address of the page.

3. According to the method for implementing blind watermark decryption based on screenshot information protection according to claim 1, it is characterized in that: The decryption method includes symmetric or asymmetric decryption, and the number of decryption times includes single decryption and multiple decryptions.