CN116630131A - Coding and decoding system and method for invisible screen watermark - Google Patents
Coding and decoding system and method for invisible screen watermark Download PDFInfo
- Publication number
- CN116630131A CN116630131A CN202310914931.1A CN202310914931A CN116630131A CN 116630131 A CN116630131 A CN 116630131A CN 202310914931 A CN202310914931 A CN 202310914931A CN 116630131 A CN116630131 A CN 116630131A
- Authority
- CN
- China
- Prior art keywords
- watermark
- screen
- image
- invisible
- original
- 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 46
- 238000012549 training Methods 0.000 claims abstract description 36
- 238000013135 deep learning Methods 0.000 claims abstract description 24
- 238000010586 diagram Methods 0.000 claims description 49
- 238000007781 pre-processing Methods 0.000 claims description 17
- 238000010276 construction Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 7
- 238000005070 sampling Methods 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 5
- 238000011176 pooling Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 abstract description 4
- 238000011156 evaluation Methods 0.000 description 9
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/0021—Image watermarking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T9/00—Image coding
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Editing Of Facsimile Originals (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the invention discloses a system and a method for encoding and decoding a screen invisible watermark, wherein the method for encoding the screen invisible watermark comprises the following steps: obtaining a certain number of watermark pictures and original pictures; constructing an encoder model based on the deep learning framework; inputting the watermark image and the original image into the encoder model for training to obtain a trained encoder model; obtaining watermark information to be encoded and an original image to be encoded, and generating a watermark image to be encoded based on the watermark information; and inputting the watermark image to be encoded and the original image to be encoded into the trained encoder model to obtain the invisible watermark image. The method for encoding the screen invisible watermark solves the problems that in the prior art, the transparency of the screen invisible watermark is high, the transparency of the screen invisible watermark experienced by a user in screen use is low, and the success rate of watermark extraction is extremely low.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a system and a method for encoding and decoding invisible watermarks of a screen.
Background
Invisible watermarks, as the name implies, are invisible watermarks, and the biggest difference between invisible watermarks and traditional watermarks is that text contents are hidden. The prior invisible screen watermarking technology mainly comprises the following two steps:
one is to encode watermark information into binary form, regenerate a watermark pattern of pure color equal to the size of the screen, represent 1 by square block of one color, represent 0 by square block of another color, encode watermark information into binary form onto watermark pattern in this way, and then float this watermark pattern on the uppermost layer of the screen with certain transparency. The method has the defects that the transparency of the watermark layer is low, watermark information cannot be extracted, and the screen use experience of a user is seriously affected due to high transparency.
The other is that the image is firstly converted into a frequency domain form based on DCT, wavelet transformation and other algorithms, and then watermark information is embedded in a high-frequency part, and the method has the defects of extremely low success rate of extracting the watermark from the picture shot by the screen and basically no solution.
Disclosure of Invention
The embodiment of the invention aims to provide a system and a method for encoding and decoding a screen invisible watermark, which are used for solving the problems that the transparency of the screen invisible watermark is high, the transparency of the screen invisible watermark experienced by a user is low, and the success rate of extracting the watermark is extremely low in the prior art.
In order to achieve the above object, an embodiment of the present invention provides a method for encoding a invisible watermark on a screen, where the method specifically includes:
obtaining a certain number of watermark pictures and original pictures;
constructing an encoder model based on the deep learning framework;
inputting the watermark image and the original image into the encoder model for training to obtain a trained encoder model;
obtaining watermark information to be encoded and an original image to be encoded, and generating a watermark image to be encoded based on the watermark information;
and inputting the watermark image to be encoded and the original image to be encoded into the trained encoder model to obtain the invisible watermark image.
Based on the technical scheme, the invention can also be improved as follows:
further, the constructing an encoder model based on the deep learning framework includes:
the encoder model comprises a convolution layer, a pooling layer, a BN layer, an up-sampling layer, a down-sampling layer, a jump connection, a stacking splice and an output layer which are sequentially connected, wherein the size of the output layer is consistent with that of an original image.
Further, the watermark image and the original image are input into the encoder model for training, and a trained encoder model is obtained, which comprises the following steps of;
and calculating a VIF value between the output invisible watermark and the input original image, calculating a loss value based on the VIF value, and back-propagating and modifying encoder model parameters until the loss value is close to 0.
A method for decoding a screen invisible watermark, the method comprising:
displaying the invisible watermark on a screen and shooting a screen;
preprocessing the screen image to obtain a rectangular screen area image;
constructing a decoder model based on the deep learning framework;
training the decoder model based on the rectangular screen region diagram to obtain a trained decoder model;
inputting a rectangular screen area diagram to be decoded into a decoder model to obtain a watermark diagram with the same size as the input diagram;
and acquiring original watermark information based on the watermark map.
Further, the preprocessing the screen map to obtain a rectangular screen area map includes:
the screen area of the screen diagram is transformed into a rectangle through perspective transformation, and the area outside the screen is cut off, so that only a complete screen area of the rectangle is left.
Further, the obtaining the original watermark information based on the watermark map includes:
traversing the watermark image obtained in the step five in a rectangular block mode according to a preset step length K, wherein the white color is considered as 1, the black color is considered as 0, so that the information string in a binary form is obtained, and the information string is converted into decimal to obtain the original watermark information.
A system for encoding a screen invisible watermark, comprising:
the first acquisition module is used for acquiring a certain number of watermark pictures and original pictures;
a first construction module for constructing an encoder model based on a deep learning framework;
the first training module is used for inputting the watermark image and the original image into the encoder model for training to obtain a trained encoder model;
the second acquisition module is used for acquiring watermark information to be encoded and original images to be encoded and generating a watermark image to be encoded based on the watermark information;
and the coder model is used for inputting the watermark image to be coded and the original image to be coded into the trained coder model to obtain the invisible watermark image.
A decoding system for a screen invisible watermark, comprising:
the shooting module is used for displaying the invisible watermark image on the screen and shooting a screen image;
the preprocessing module is used for preprocessing the screen image to obtain a rectangular screen area image;
a second construction module for constructing a decoder model based on the deep learning framework;
the second training module is used for training the decoder model based on the rectangular screen area diagram to obtain a trained decoder model;
the decoder model is used for inputting the rectangular screen area diagram to be decoded into the decoder model to obtain a watermark diagram with the same size as the input diagram;
and the third acquisition module is used for acquiring the original watermark information based on the watermark image.
An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method when the computer program is executed.
A non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method.
The embodiment of the invention has the following advantages:
the method for encoding the invisible watermark of the screen acquires a certain number of watermark pictures and original pictures; constructing an encoder model based on the deep learning framework; inputting the watermark image and the original image into the encoder model for training to obtain a trained encoder model; obtaining watermark information to be encoded and an original image to be encoded, and generating a watermark image to be encoded based on the watermark information; inputting the watermark pattern to be encoded and the original pattern to be encoded into the trained encoder model to obtain the invisible watermark pattern, and solving the problems that the transparency of the invisible watermark of the screen is high, the transparency of the invisible watermark of the screen is low, and the success rate of extracting the watermark is extremely low when the screen of a user is used and experienced in the prior art.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It will be apparent to those skilled in the art from this disclosure that the drawings described below are merely exemplary and that other embodiments may be derived from the drawings provided without undue effort.
The structures, proportions, sizes, etc. shown in the present specification are shown only for the purposes of illustration and description, and are not intended to limit the scope of the invention, which is defined by the claims, so that any structural modifications, changes in proportions, or adjustments of sizes, which do not affect the efficacy or the achievement of the present invention, should fall within the scope of the invention.
FIG. 1 is a flow chart of a method of encoding a invisible screen watermark of the present invention;
FIG. 2 is a flow chart of a method of decoding a invisible watermark of the present invention;
FIG. 3 is a block diagram of a system for encoding a invisible watermark according to the present invention;
FIG. 4 is a block diagram of a decoding system for invisible screen watermarking according to the present invention;
FIG. 5 is a schematic diagram of a watermark pattern to be encoded according to the present invention;
FIG. 6 is a schematic diagram of a watermark pattern of the present invention;
fig. 7 is a schematic diagram of an entity structure of an electronic device according to the present invention.
Wherein the reference numerals are as follows:
the system comprises a first acquisition module 10, a first construction module 20, a first training module 30, a second acquisition module 40, an encoder model 50, a shooting module 60, a preprocessing module 70, a second construction module 80, a second training module 90, a decoder model 100, a third acquisition module 110, an electronic device 120, a processor 1201, a memory 1202 and a bus 1203.
Detailed Description
Other advantages and advantages of the present invention will become apparent to those skilled in the art from the following detailed description, which, by way of illustration, is to be read in connection with certain specific embodiments, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Examples
Fig. 1 is a flowchart of an embodiment of a method for encoding a screen invisible watermark according to the present invention, as shown in fig. 1, the method for encoding a screen invisible watermark according to the embodiment of the present invention includes the following steps:
s101, obtaining a certain number of watermark pictures and original pictures;
specifically, a certain number refers to at least 16 pieces, at least 16 pieces of watermark images and at least 16 pieces of original images are obtained.
S102, constructing an encoder model based on a deep learning framework;
specifically, the encoder model comprises a convolution layer, a pooling layer, a BN layer, an up-sampling layer, a down-sampling layer, a jump connection, a stacking splice and an output layer which are sequentially connected, wherein the size of the output layer is consistent with that of an original image.
The encoder model comprises 2 inputs, an original image and a watermark image, and after the original image and the watermark image are input into the encoder model, a watermark-containing image which is very similar to the original image in visual sense is obtained, and the image and the original image are basically the same in appearance, so that the damage to the original image content is small.
S103, inputting the watermark image and the original image into an encoder model for training, and obtaining a trained encoder model;
specifically, a VIF value between the output hidden watermark pattern and the input original pattern is calculated, a loss value is calculated based on the VIF value, and the encoder model parameters are modified by back propagation until the loss value is close to 0.
Inputting a batch (at least 16) of original pictures and watermark pictures into an encoder model, calculating a VIF value between an output invisible watermark picture and the input original pictures, taking the square of the difference between the VIF value and 0.95 as a loss value, and modifying encoder model parameters by back propagation until the loss value is close to 0.
S104, obtaining watermark information to be encoded and original images to be encoded, and generating the watermark images to be encoded based on the watermark information.
Specifically, as shown in fig. 5, watermark information is converted into a binary form, a watermark pattern of black and white squares is generated based on the binary information, with white places representing 1 and black places representing 0.
S105, inputting the watermark pattern to be encoded and the original pattern to be encoded into a trained encoder model to obtain an invisible watermark pattern;
the method for encoding the invisible watermark of the screen acquires a certain number of watermark pictures and original pictures; constructing an encoder model based on the deep learning framework; inputting the watermark image and the original image into the encoder model for training to obtain a trained encoder model; obtaining watermark information to be encoded and an original image to be encoded, and generating a watermark image to be encoded based on the watermark information; and inputting the watermark image to be encoded and the original image to be encoded into the trained encoder model to obtain the invisible watermark image. The method solves the problems that in the prior art, the transparency of the screen invisible watermark is high, the transparency of the screen invisible watermark experienced by a user in screen use is low, and the success rate of watermark extraction is extremely low.
Fig. 2 is a flowchart of an embodiment of a method for decoding a screen invisible watermark according to the present invention, as shown in fig. 2, the method for decoding a screen invisible watermark according to the embodiment of the present invention includes the following steps:
s201, displaying the invisible watermark on a screen and shooting a screen;
specifically, the watermarked image is displayed on a screen, and a screen image is shot by a device such as a mobile phone.
S202, preprocessing a screen image to obtain a rectangular screen area image;
specifically, the screen area of the screen diagram is transformed into a rectangle through perspective transformation, and the area outside the screen is cut off, so that only a complete screen area of the rectangle is left.
S203, constructing a decoder model based on the deep learning framework;
specifically, the decoder model mainly includes an input layer, an output layer, and several hidden layers.
S204, training a decoder model based on the rectangular screen region diagram to obtain a trained decoder model;
specifically, a training set, a testing set and a verification set are constructed based on the rectangular screen area diagram;
the decoder model is trained based on the training set.
Performing performance evaluation on the trained decoder model based on the verification set to obtain a decoder model meeting performance conditions; and evaluating decoding results of the decoder model meeting the performance conditions based on the test set to obtain an evaluation index corresponding to the decoder model. Performing performance evaluation on the decoder model to obtain a percent score (namely, the highest score is 100 points and the lowest score is 0 points), and determining the decoder model with the score larger than a set value based on the percent score, wherein for example, the decoder model with the score larger than 90 points is the decoder model meeting the performance condition;
and performing evaluation index calculation on the decoder model meeting the performance condition to obtain evaluation indexes of the decoder model, and calculating to obtain an evaluation value corresponding to each evaluation index, wherein the evaluation value is used for representing the capability value of the decoder model on the evaluation indexes.
S205, inputting a rectangular screen area diagram to be decoded into a decoder model to obtain a watermark diagram with the same size as the input diagram;
the watermark pattern is shown in fig. 6.
S206, acquiring original watermark information based on a watermark pattern;
specifically, traversing the watermark image obtained in S206 in the form of rectangular blocks according to a preset step length K, wherein the white color is more than 1, the black color is more than 0, so as to obtain a binary information string, and converting the binary information string into decimal information to obtain the original watermark information.
The decoding method of the invisible watermark of the screen displays the invisible watermark image on the screen and shoots a screen image; preprocessing the screen image to obtain a rectangular screen area image; constructing a decoder model based on the deep learning framework; training the decoder model based on the rectangular screen region diagram to obtain a trained decoder model; inputting a rectangular screen area diagram to be decoded into a decoder model to obtain a watermark diagram with the same size as the input diagram; and acquiring original watermark information based on the watermark map. The problems of low transparency of the invisible watermark of the screen, low success rate of watermark extraction and low success rate of watermark extraction on pictures shot by the screen in the prior art are solved.
FIG. 3 is a flowchart of an embodiment of a coding system for invisible watermarking of a screen according to the present invention; as shown in fig. 3, the coding system for the invisible watermark of the screen provided by the embodiment of the invention comprises the following steps:
a first obtaining module 10, configured to obtain a certain number of watermark graphs and original graphs;
a first construction module 20 for constructing an encoder model 50 based on a deep learning framework;
the first training module 30 is configured to input the watermark pattern and the original pattern into the encoder model 50 for training, so as to obtain a trained encoder model 50;
a second obtaining module 40, configured to obtain watermark information to be encoded and an original image to be encoded, and generate a watermark image to be encoded based on the watermark information;
and the encoder model 50 is used for inputting the watermark image to be encoded and the original image to be encoded into the trained encoder model 50 to obtain the invisible watermark image.
The encoder model 50 includes a convolutional layer, a pooling layer, a BN layer, an upsampling layer, a downsampling layer, a skip connection, a stack splice, and an output layer connected in sequence, wherein the output layer is sized to be consistent with the artwork.
The training module is also configured to:
the VIF value between the output hidden watermark pattern and the input original pattern is calculated, a loss value is calculated based on the VIF value, and the encoder model 50 parameters are modified by back propagation until the loss value is close to 0.
According to the screen invisible watermark encoding system, a certain number of watermark images and original images are acquired through the first acquisition module 10; constructing, by the first construction module 20, an encoder model 50 based on the deep learning framework;
inputting the watermark pattern and the original pattern into the encoder model 50 for training through a first training module 30 to obtain a trained encoder model 50; obtaining watermark information to be encoded and an original image to be encoded through a second obtaining module 40, and generating a watermark image to be encoded based on the watermark information; and inputting the watermark pattern to be encoded and the original pattern to be encoded into the trained encoder model 50 through the encoder model 50 to obtain the invisible watermark pattern. The method for encoding the invisible watermark of the screen solves the problems that in the prior art, watermark information cannot be extracted due to low transparency of a watermark layer, and screen use experience of a user is seriously affected due to high transparency.
FIG. 4 is a flow chart of an embodiment of a decoding system for invisible watermarking of a screen according to the present invention; as shown in fig. 4, the decoding system for invisible watermarks on a screen provided by the embodiment of the invention includes the following steps:
a photographing module 60 for displaying the invisible watermark on a screen and photographing a screen;
a preprocessing module 70, configured to preprocess the screen map to obtain a rectangular screen area map;
a second construction module 80 for constructing a decoder model 100 based on the deep learning framework;
a second training module 90, configured to train the decoder model 100 based on the rectangular screen region map, to obtain a trained decoder model 100;
the decoder model 100 is used for inputting the rectangular screen area diagram to be decoded into the decoder model 100 to obtain a watermark diagram with the same size as the input diagram;
a third obtaining module 110, configured to obtain the original watermark information based on the watermark map.
The preprocessing module 70 is further configured to:
the screen area of the screen diagram is transformed into a rectangle through perspective transformation, and the area outside the screen is cut off, so that only a complete screen area of the rectangle is left.
The third obtaining module 110 is further configured to:
traversing the watermark image obtained in the step five in a rectangular block mode according to a preset step length K, wherein the white color is considered as 1, the black color is considered as 0, so that the information string in a binary form is obtained, and the information string is converted into decimal to obtain the original watermark information.
According to the decoding system of the invisible watermark of the screen, the invisible watermark image is displayed on the screen through the shooting module 60, and a screen image is shot; preprocessing the screen map through a preprocessing module 70 to obtain a rectangular screen area map; constructing, by the second construction module 80, the decoder model 100 based on the deep learning framework; training the decoder model 100 by a second training module 90 based on the rectangular screen region map to obtain a trained decoder model 100; inputting a rectangular screen area diagram to be decoded into the decoder model 100 through the decoder model 100 to obtain a watermark diagram with the same size as the input diagram; the original watermark information is acquired based on the watermark map by a third acquisition module 110. The method for encoding the invisible watermark of the screen solves the problem that tax returns of the month to which the same tax belongs cannot be classified based on the tax return mode and then summarized in the prior art.
Fig. 7 is a schematic diagram of an entity structure of an electronic device according to an embodiment of the present invention, as shown in fig. 7, an electronic device 120 includes: a processor 1201 (processor), a memory 1202 (memory), and a bus 1203;
the processor 1201 and the memory 1202 perform communication with each other through the bus 1203;
the processor 1201 is configured to invoke program instructions in the memory 1202 to perform the methods provided by the method embodiments described above, including, for example: obtaining a certain number of watermark pictures and original pictures; constructing an encoder model based on the deep learning framework; inputting the watermark image and the original image into the encoder model for training to obtain a trained encoder model; obtaining watermark information to be encoded and an original image to be encoded, and generating a watermark image to be encoded based on the watermark information; and inputting the watermark image to be encoded and the original image to be encoded into the trained encoder model to obtain the invisible watermark image.
Displaying the invisible watermark on a screen and shooting a screen; preprocessing the screen image to obtain a rectangular screen area image; constructing a decoder model based on the deep learning framework; training the decoder model based on the rectangular screen region diagram to obtain a trained decoder model; inputting a rectangular screen area diagram to be decoded into a decoder model to obtain a watermark diagram with the same size as the input diagram; and acquiring original watermark information based on the watermark map.
The present embodiment provides a non-transitory computer readable storage medium storing computer instructions that cause a computer to perform the methods provided by the above-described method embodiments, for example, including: obtaining a certain number of watermark pictures and original pictures; constructing an encoder model based on the deep learning framework; inputting the watermark image and the original image into the encoder model for training to obtain a trained encoder model; obtaining watermark information to be encoded and an original image to be encoded, and generating a watermark image to be encoded based on the watermark information; and inputting the watermark image to be encoded and the original image to be encoded into the trained encoder model to obtain the invisible watermark image.
Displaying the invisible watermark on a screen and shooting a screen; preprocessing the screen image to obtain a rectangular screen area image; constructing a decoder model based on the deep learning framework; training the decoder model based on the rectangular screen region diagram to obtain a trained decoder model; inputting a rectangular screen area diagram to be decoded into a decoder model to obtain a watermark diagram with the same size as the input diagram; and acquiring original watermark information based on the watermark map.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware associated with program instructions, where the foregoing program may be stored in a computer readable storage medium, and when executed, the program performs steps including the above method embodiments; and the aforementioned storage medium includes: various storage media such as ROM, RAM, magnetic or optical disks may store program code.
The apparatus embodiments described above are merely illustrative, wherein elements illustrated as separate elements may or may not be physically separate, and elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on such understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the embodiments or the methods of some parts of the embodiments.
While the invention has been described in detail in the foregoing general description and specific examples, it will be apparent to those skilled in the art that modifications and improvements can be made thereto. Accordingly, such modifications or improvements may be made without departing from the spirit of the invention and are intended to be within the scope of the invention as claimed.
Claims (10)
1. A method for encoding a invisible watermark on a screen, the method comprising:
obtaining a certain number of watermark pictures and original pictures;
constructing an encoder model based on the deep learning framework;
inputting the watermark image and the original image into the encoder model for training to obtain a trained encoder model;
obtaining watermark information to be encoded and an original image to be encoded, and generating a watermark image to be encoded based on the watermark information;
and inputting the watermark image to be encoded and the original image to be encoded into the trained encoder model to obtain the invisible watermark image.
2. The method for encoding a screen hidden watermark according to claim 1, wherein said constructing an encoder model based on a deep learning framework comprises:
the encoder model comprises a convolution layer, a pooling layer, a BN layer, an up-sampling layer, a down-sampling layer, a jump connection, a stacking splice and an output layer which are sequentially connected, wherein the size of the output layer is consistent with that of an original image.
3. The method for encoding a invisible screen watermark according to claim 1, wherein the step of inputting the watermark pattern and the original pattern into the encoder model for training to obtain a trained encoder model comprises;
and calculating a VIF value between the output invisible watermark and the input original image, calculating a loss value based on the VIF value, and back-propagating and modifying encoder model parameters until the loss value is close to 0.
4. A method for decoding a hidden watermark on a screen, the method comprising:
displaying the invisible watermark on a screen and shooting a screen;
preprocessing the screen image to obtain a rectangular screen area image;
constructing a decoder model based on the deep learning framework;
training the decoder model based on the rectangular screen region diagram to obtain a trained decoder model;
inputting a rectangular screen area diagram to be decoded into a decoder model to obtain a watermark diagram with the same size as the input diagram;
and acquiring original watermark information based on the watermark map.
5. The method for decoding an invisible screen watermark according to claim 4, wherein preprocessing the screen map to obtain a rectangular screen area map comprises:
the screen area of the screen diagram is transformed into a rectangle through perspective transformation, and the area outside the screen is cut off, so that only a complete screen area of the rectangle is left.
6. The method for decoding an invisible screen watermark according to claim 4, wherein said obtaining original watermark information based on said watermark pattern comprises:
traversing the watermark image obtained in the step five in a rectangular block mode according to a preset step length K, wherein the white color is considered as 1, the black color is considered as 0, so that the information string in a binary form is obtained, and the information string is converted into decimal to obtain the original watermark information.
7. A system for encoding a invisible screen watermark, comprising:
the first acquisition module is used for acquiring a certain number of watermark pictures and original pictures;
a first construction module for constructing an encoder model based on a deep learning framework;
the first training module is used for inputting the watermark image and the original image into the encoder model for training to obtain a trained encoder model;
the second acquisition module is used for acquiring watermark information to be encoded and original images to be encoded and generating a watermark image to be encoded based on the watermark information;
and the coder model is used for inputting the watermark image to be coded and the original image to be coded into the trained coder model to obtain the invisible watermark image.
8. A decoding system for a screen invisible watermark, comprising:
the shooting module is used for displaying the invisible watermark image on the screen and shooting a screen image;
the preprocessing module is used for preprocessing the screen image to obtain a rectangular screen area image;
a second construction module for constructing a decoder model based on the deep learning framework;
the second training module is used for training the decoder model based on the rectangular screen area diagram to obtain a trained decoder model;
the decoder model is used for inputting the rectangular screen area diagram to be decoded into the decoder model to obtain a watermark diagram with the same size as the input diagram;
and the third acquisition module is used for acquiring the original watermark information based on the watermark image.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and running on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1 to 3 or the steps of the method according to any one of claims 4 to 6 when the computer program is executed by the processor.
10. A non-transitory computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 3 or the steps of the method according to any one of claims 4 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310914931.1A CN116630131A (en) | 2023-07-25 | 2023-07-25 | Coding and decoding system and method for invisible screen watermark |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310914931.1A CN116630131A (en) | 2023-07-25 | 2023-07-25 | Coding and decoding system and method for invisible screen watermark |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116630131A true CN116630131A (en) | 2023-08-22 |
Family
ID=87592486
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310914931.1A Pending CN116630131A (en) | 2023-07-25 | 2023-07-25 | Coding and decoding system and method for invisible screen watermark |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116630131A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117495649A (en) * | 2024-01-02 | 2024-02-02 | 支付宝(杭州)信息技术有限公司 | Image processing method, device and equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150278980A1 (en) * | 2014-03-25 | 2015-10-01 | Digimarc Corporation | Screen watermarking methods and arrangements |
| CN108040190A (en) * | 2017-11-22 | 2018-05-15 | 明鉴方寸（北京）科技有限公司 | A kind of stealth watermark recognition methods, device and storage device |
| CN111079532A (en) * | 2019-11-13 | 2020-04-28 | 杭州电子科技大学 | Video content description method based on text self-encoder |
| CN111768327A (en) * | 2020-06-30 | 2020-10-13 | 苏州科达科技股份有限公司 | Watermark adding and extracting method and device based on deep learning and storage medium |
| CN112529757A (en) * | 2020-12-04 | 2021-03-19 | 平安科技（深圳）有限公司 | Screen information protection method and device, computer equipment and readable storage medium |
| CN113222800A (en) * | 2021-04-12 | 2021-08-06 | 国网江苏省电力有限公司营销服务中心 | Robust image watermark embedding and extracting method and system based on deep learning |
| CN113935883A (en) * | 2021-10-09 | 2022-01-14 | 深圳市联软科技股份有限公司 | Screen watermark identification method and system based on deep learning |
-
2023
- 2023-07-25 CN CN202310914931.1A patent/CN116630131A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150278980A1 (en) * | 2014-03-25 | 2015-10-01 | Digimarc Corporation | Screen watermarking methods and arrangements |
| CN108040190A (en) * | 2017-11-22 | 2018-05-15 | 明鉴方寸（北京）科技有限公司 | A kind of stealth watermark recognition methods, device and storage device |
| CN111079532A (en) * | 2019-11-13 | 2020-04-28 | 杭州电子科技大学 | Video content description method based on text self-encoder |
| CN111768327A (en) * | 2020-06-30 | 2020-10-13 | 苏州科达科技股份有限公司 | Watermark adding and extracting method and device based on deep learning and storage medium |
| CN112529757A (en) * | 2020-12-04 | 2021-03-19 | 平安科技（深圳）有限公司 | Screen information protection method and device, computer equipment and readable storage medium |
| WO2022116493A1 (en) * | 2020-12-04 | 2022-06-09 | 平安科技（深圳）有限公司 | Screen information protection method and apparatus, and computer device and readable storage medium |
| CN113222800A (en) * | 2021-04-12 | 2021-08-06 | 国网江苏省电力有限公司营销服务中心 | Robust image watermark embedding and extracting method and system based on deep learning |
| CN113935883A (en) * | 2021-10-09 | 2022-01-14 | 深圳市联软科技股份有限公司 | Screen watermark identification method and system based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 舒威: "基于信息安全的屏摄溯源取证***的设计与实现", 硕士学位论文, pages 20 - 22 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117495649A (en) * | 2024-01-02 | 2024-02-02 | 支付宝(杭州)信息技术有限公司 | Image processing method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021208247A1 (en) | Mimic compression method and apparatus for video image, and storage medium and terminal | |
| CN109087258A (en) | A kind of image rain removing method and device based on deep learning | |
| CN111353956B (en) | Image restoration method and device, computer equipment and storage medium | |
| Chen et al. | JSNet: a simulation network of JPEG lossy compression and restoration for robust image watermarking against JPEG attack | |
| CN112132158A (en) | Visual picture information embedding method based on self-coding network | |
| CN116630131A (en) | Coding and decoding system and method for invisible screen watermark | |
| CN116433914A (en) | Two-dimensional medical image segmentation method and system | |
| WO2016127271A1 (en) | An apparatus and a method for reducing compression artifacts of a lossy-compressed image | |
| CN116233445B (en) | Video encoding and decoding processing method and device, computer equipment and storage medium | |
| CN113724136A (en) | Video restoration method, device and medium | |
| CN111698508B (en) | Super-resolution-based image compression method, device and storage medium | |
| CN111355965B (en) | Image compression and restoration method and device based on deep learning | |
| CN114529441A (en) | Image frequency domain digital watermarking method, system, device and medium | |
| CN115908095A (en) | Hierarchical attention feature fusion-based robust image watermarking method and system | |
| CN115908205A (en) | Image restoration method and device, electronic equipment and storage medium | |
| CN114821100A (en) | Image compressed sensing reconstruction method based on structural group sparse network | |
| Liao et al. | GIFMarking: The robust watermarking for animated GIF based deep learning | |
| Liu et al. | Facial image inpainting using multi-level generative network | |
| CN110738666A (en) | discrete cosine transform-based image semantic segmentation method and device | |
| CN105488822A (en) | Reversible image hiding algorithm based on AMBTC algorithm | |
| CN113256744B (en) | Image coding and decoding method and system | |
| CN114173137A (en) | Video coding method and device and electronic equipment | |
| CN116912130A (en) | Image defogging method based on multi-receptive field feature fusion and mixed attention | |
| CN115861401A (en) | Binocular and point cloud fusion depth recovery method, device and medium | |
| CN114493971A (en) | Media data conversion model training and digital watermark embedding method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230822 |
|
| RJ01 | Rejection of invention patent application after publication |