CN111966973A - Copyright protection method and system based on picture pixel value steganography - Google Patents

Abstract

The invention provides a copyright protection method and system based on steganography of picture pixel values. The invention compares the parity of picture pixel points with the binary string, and fine-tunes the pixel points according to the agreed rules, so as to achieve no Under the premise of affecting the quality and size of the picture itself, the copyright traceability information of specific content is written into the file, and the copyright protection of the picture is realized through reverse forensics, so as to not affect the quality and display effect of the picture itself, but also Hide the effect of specific information in the picture, facilitate the traceability of the picture, help copyright protection, and ensure that it is not plagiarized by others.

Description

A copyright protection method and system based on image pixel value steganography

technical field

The invention relates to the technical field of copyright protection, in particular to a copyright protection method and system based on image pixel value steganography.

Background technique

During web development, the released products and user-uploaded resources contain a large number of image resources, some of which are copyrighted or contain sensitive information on the intranet. Traceability and protection of resources.

At present, the cost of obtaining valuable information on the Internet is getting higher and higher. Many website operators spend a lot of manpower and funds to collect the information, which is easily collected by some crawlers for their own use or profit. For example, the model photos of Taobao's online store, the photos taken by the model at a high cost are easily stolen by other online stores and used in their own online stores. For SNS user social networking sites, its users and user-generated content are the site's greatest wealth and a part of the site's competitiveness. A large number of beautiful photos uploaded by users of some social travel websites are scraped by competitors to enrich their own content with crawlers.

How to ensure that these contents are not easily plagiarized by others is an important problem that needs to be solved urgently in the field. The photos taken by the camera contain Exif (Exchangeable image file format, which records the attribute information and shooting data of digital photos) information, but these information are mainly equipment Information, shooting time, etc., the amount of stored information is small, can be easily erased and modified, and does not have the value of traceability.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a kind of copyright protection method and system based on picture pixel value steganography, aiming to solve the problem that the Exif information of the picture can be easily erased and modified in the prior art, and the copyright protection cannot be provided, so as to realize the protection of the picture. Tracing the source can help copyright protection and ensure that it is not plagiarized by others.

In order to achieve the above-mentioned technical purpose, the present invention provides a copyright protection method based on image pixel value steganography, and the method includes the following operations:

Sampling the pixels of the picture to determine the number of characters that can be written into the picture;

When the number of characters of the copyright information text to be encoded is less than the number of characters of the writable picture, convert the copyright information text to be encoded into a binary string;

One-to-one correspondence between the bits of the binary string and the RGB values of the sequentially arranged pixels, and fine-tune the pixels bit by bit through the binary string. According to the rule that 1 is odd and 0 is even, the RGB value of the pixel is odd and even. Add 1 or subtract 1 to the RGB values whose properties are different from the corresponding binary string bits, adjust them to correspond one-to-one with the numerical parity of the corresponding binary string bits, and write the total length of the required encoding into the last N pixels of the picture in reverse order. ;

When collecting evidence for copyright protection, extract the coding length A from the last N pixels of the picture, and obtain the RGB three values corresponding to the first A pixels of the picture according to the length value A. According to the rule that 1 is an odd number and 0 is an even number, the RGB value is The parity of is converted into a binary string containing 1, 0, and the binary string is decoded in reverse for the publishing rights information text.

Preferably, the copyright information is any one or a combination of file owner, file creation time, and unique serial number.

Preferably, the format of the picture needs to be converted into a unified format when the copyright information is written.

Preferably, the number of characters that can be written into the picture is:

When the characters are Chinese, the number of characters is the number of pixels/24; when the characters are English or numbers, the number of characters is the number of pixels/16.

The embodiment of the present invention also provides a copyright protection system based on image pixel value steganography, the system includes:

The picture pixel sampling module is used to sample the pixels of the picture to determine the number of characters that can be written into the picture;

A copyright information text conversion module, used for converting the required encoded copyright information text into a binary string when the number of characters of the required encoded copyright information text is less than the number of characters that can be written into the picture;

The pixel point RGB value fine-tuning module is used to correspond the bits of the binary string to the RGB three values of the sequentially arranged pixel points, and fine-tune the pixel points bit by bit through the binary string, according to 1 for odd and 0 for even. As a rule, add 1 or subtract 1 to the RGB value parity of the pixel point and the RGB value with different corresponding binary string bits, adjust it to correspond one-to-one with the numerical parity of the corresponding binary string bits, and set the total length of the required encoding according to Write the last N pixels of the picture in reverse order;

The copyright information extraction module is used to extract the coding length A for the last N pixels of the picture when collecting evidence of copyright protection, and obtain the RGB three values corresponding to the first A pixels of the picture according to the length value A, according to which 1 is an odd number and 0 is an odd number. The even number rule converts the parity of the RGB value into a binary string containing 1 and 0, and reversely decodes the binary string to publish the copyright information text.

Preferably, the copyright information is any one or a combination of file owner, file creation time, and unique serial number.

Preferably, the format of the picture needs to be converted into a unified format when the copyright information is written.

Preferably, the number of characters that can be written into the picture is:

When the characters are Chinese, the number of characters is the number of pixels/24; when the characters are English or numbers, the number of characters is the number of pixels/16.

The present invention also provides a copyright protection device based on picture pixel value steganography, including:

memory for storing computer programs;

The processor is configured to execute the computer program to implement the copyright protection method based on steganography of picture pixel values.

The present invention also provides a readable storage medium for storing a computer program, wherein when the computer program is executed by a processor, the copyright protection method based on picture pixel value steganography is implemented.

The effects provided in the summary of the invention are only the effects of the embodiments, rather than all the effects of the invention. One of the above technical solutions has the following advantages or beneficial effects:

Compared with the prior art, the present invention compares the parity of the pixel points of the picture with the binary string, and fine-tunes the pixels according to the agreed rules, so as to achieve the premise of not affecting the quality of the picture itself and the size of the volume. The copyright traceability information of specific content is written into the file, and the copyright protection of the picture is realized through reverse forensics, so as not to affect the quality and display effect of the picture itself, but also to hide the effect of specific information in the picture, which is convenient for the picture. Tracing the source can help copyright protection and ensure that it is not plagiarized by others.

Description of drawings

1 is a flowchart of a copyright protection method based on picture pixel value steganography provided in an embodiment of the present invention;

2 is an example of an encoding process provided in an embodiment of the present invention;

FIG. 3 is a block diagram of a copyright protection system based on image pixel value steganography provided in an embodiment of the present invention.

Detailed ways

In order to clearly illustrate the technical features of the solution, the present invention will be described in detail below through specific embodiments and in conjunction with the accompanying drawings. The following disclosure provides many different embodiments or examples for implementing different structures of the invention. In order to simplify the disclosure of the present invention, the components and arrangements of specific examples are described below. Furthermore, the present invention may repeat reference numerals and/or letters in different instances. This repetition is for the purpose of simplicity and clarity and does not in itself indicate a relationship between the various embodiments and/or arrangements discussed. It should be noted that the components illustrated in the figures are not necessarily drawn to scale. Descriptions of well-known components and processing techniques and processes are omitted from the present invention to avoid unnecessarily limiting the present invention.

A method and system for copyright protection based on image pixel value steganography provided by the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

As shown in FIG. 1, the present invention discloses a copyright protection method based on image pixel value steganography, and the method includes the following operations:

Sampling the pixels of the picture to determine the number of characters that can be written into the picture;

When the number of characters of the copyright information text to be encoded is less than the number of characters of the writable picture, convert the copyright information text to be encoded into a binary string;

One-to-one correspondence between the bits of the binary string and the RGB values of the sequentially arranged pixels, and fine-tune the pixels bit by bit through the binary string. According to the rule that 1 is odd and 0 is even, the RGB value of the pixel is odd and even. Add 1 or subtract 1 to the RGB values whose properties are different from the corresponding binary string bits, adjust them to correspond one-to-one with the numerical parity of the corresponding binary string bits, and write the total length of the required encoding into the last N pixels of the picture in reverse order. ;

When collecting evidence for copyright protection, extract the coding length A from the last N pixels of the picture, and obtain the RGB three values corresponding to the first A pixels of the picture according to the length value A. According to the rule that 1 is an odd number and 0 is an even number, the RGB value is The parity of is converted into a binary string containing 1, 0, and the binary string is decoded in reverse for the publishing rights information text.

In the embodiment of the present invention, by comparing the parity of the pixel points of the picture with the binary string, and fine-tuning the pixels according to the agreed rules, the copyright of specific content can be traced without affecting the quality and size of the picture itself. The information is written into the file, and the copyright protection of the picture is realized through reverse forensics.

Format conversion of pictures. Format conversion is to uniformly convert pictures of various formats into PNG format. The most common picture formats in the Web include GIF, PNG and JPG, etc. Other formats are not common, and it is feasible to convert between picture formats. Yes, the converted images are in PNG format, which reduces the complexity of subsequent processing.

Compress the format-converted pictures, and compress the pictures that are too large for transmission in the network. This operation depends on the network bandwidth and storage.

In order to determine the amount of information that can be written, image information sampling is performed. Get the pixel size of the picture, that is, the width and height of the picture. For example, the width and height of a mobile phone photo is 3120*4160, and the result is 12979200. A photo with a width and height of 3120*4160 has 12979200 pixels, and each pixel is a color block of a certain color. In the color system, it is considered that each color is composed of three colors of red, green and blue, namely red, green and blue. The harmonic value range of each color of RGB is 0-255. For example, red corresponds to The RGB is (255, 0, 0). Calculate the amount of information provided by the picture, 12979200 pixels correspond to 12979200*3 color value samples, that is, 12979200*3=38937600 bits of information can be stored.

Encode specific information such as file owner, file creation time, unique serial number and other information into the picture. Collect the information that needs to be hidden in the picture. The content is not limited. It can include the name of the creator of the picture, private information such as user account and email address, unique serial number, etc., as long as the information can be traced back and the source can be uniquely determined. As shown in Figure 2, the content is converted into a binary string according to UTF-8 encoding, each Chinese is 3 bytes, English and numbers are 2 bytes, each byte is 8 bits, such as "copyright" The binary corresponding to two Chinese characters is "111001111000100110001000

111001101001110110000011", calculate the number of characters occupied by the information to be encoded, and compare it with the value calculated in the image information sampling. For example, the information to be stored includes 100 characters including the uploader's name, user name, etc. A character occupies 2.5 bytes, and 38937600 bits are 38937600/8=4867200 bits, which can store 4867200/2.5=1946880 characters, far exceeding the 100 characters to be stored. Convert the input text to a binary string , and then correspond to the RGB pixel parity of the picture. For example, the RGB values of the first three pixels of the picture are (222, 111, 23), (222, 10, 23), (1, 58, 40), and the rule is that even values correspond to 0 , the odd value corresponds to 1, then the binary corresponding to the three pixels is 011001100. On the contrary, if the binary string of the input text is known, it is only necessary to fine-tune the pixels bit by bit, such as the binary string bit 110010, the first bit 1 is odd, then the first color value 222 of the first pixel needs to be changed to the nearest odd number 223 or 221, the second 1 is an odd number, and the second color value of the first pixel is 111 already. If it is an odd number, no adjustment is required, and so on, the coding information can be written. Due to the limited observation accuracy of the human eye, fine-tuning the pixels in this way cannot be recognized by the human eye and does not affect the picture itself. The display effect. The total length to be encoded is recorded in the last 5 pixels of the picture according to the encoding rules in the previous step, and recorded in reverse order.

Extract copyright information, extract coding information from pictures, determine whether the content of the pictures has been stolen, and extract hidden information for proof. Take the recorded coding length A from the last 5 pixels at the end of the picture, and obtain the RGB three values corresponding to the first A pixels of the picture according to the length value A, and obtain its parity. According to the even value corresponding to 0 and the odd value corresponding to 1 The rule is converted to a binary string represented by 0s and 1s, and the binary string is decoded in reverse to produce text.

The embodiment of the present invention writes the text information into the picture by encoding the pixels of the specific text and the picture according to the agreed rules, so as to achieve the effect of not affecting the quality and display effect of the picture itself, and hiding the specific information in the picture, The operation is reversible, and the text information stored in the picture can be extracted in reverse. For example, the food website A writes its own domain name www.aweb.com in the uploaded photo, and the website B uses the crawler program to steal the photo as its own. To use, website A only needs to save this photo from website B, perform the reverse operation, and extract the domain name information of A, which can prove that website B has infringement. Due to the limited observation accuracy of the human eye, such fine-tuning of pixels is carried out. , the human eye cannot recognize it, and it does not affect the display effect of the picture itself. Since the present invention only adjusts the parity value of the picture pixels, and does not add any other redundant information, this copyright protection and source traceability method will not have a significant impact on the picture volume.

As shown in FIG. 3 , an embodiment of the present invention further discloses a copyright protection system based on steganography of picture pixel values, the system includes:

The picture pixel sampling module is used to sample the pixels of the picture to determine the number of characters that can be written into the picture;

A copyright information text conversion module, used for converting the required encoded copyright information text into a binary string when the number of characters of the required encoded copyright information text is less than the number of characters that can be written into the picture;

The pixel point RGB value fine-tuning module is used to correspond the bits of the binary string to the RGB three values of the sequentially arranged pixel points, and fine-tune the pixel points bit by bit through the binary string, according to 1 for odd and 0 for even. As a rule, add 1 or subtract 1 to the RGB value parity of the pixel point and the RGB value with different corresponding binary string bits, adjust it to correspond one-to-one with the numerical parity of the corresponding binary string bits, and set the total length of the required encoding according to Write the last N pixels of the picture in reverse order;

The copyright information extraction module is used to extract the coding length A for the last N pixels of the picture when collecting evidence of copyright protection, and obtain the RGB three values corresponding to the first A pixels of the picture according to the length value A, according to which 1 is an odd number and 0 is an odd number. The even number rule converts the parity of the RGB value into a binary string containing 1 and 0, and reversely decodes the binary string to publish the copyright information text.

Format conversion of pictures. Format conversion is to uniformly convert pictures of various formats into PNG format. The most common picture formats in the Web include GIF, PNG, and JPG, etc. Other formats are not common, and it is feasible to convert between picture formats. Yes, the converted images are in PNG format, which reduces the complexity of subsequent processing.

Compress the format-converted pictures, and compress the pictures that are too large for transmission in the network. This operation depends on the network bandwidth and storage.

In order to determine the amount of information that can be written, image information sampling is performed. Get the pixel size of the picture, that is, the width and height of the picture. For example, the width and height of a mobile phone photo is 3120*4160, and the result is 12979200. A photo with a width and height of 3120*4160 has 12979200 pixels, and each pixel is a color block of a certain color. In the color system, it is considered that each color is composed of three colors of red, green and blue, namely red, green and blue. The harmonic value range of each color of RGB is 0-255. For example, red corresponds to The RGB is (255, 0, 0). Calculate the amount of information provided by the picture, 12979200 pixels correspond to 12979200*3 color value samples, that is, 12979200*3=38937600 bits of information can be stored.

Encode specific information such as file owner, file creation time, unique serial number and other information into the picture. Collect the information that needs to be hidden in the picture. The content is not limited. It can include the name of the creator of the picture, private information such as user account and email address, unique serial number, etc., as long as the information can be traced back and the source can be uniquely determined. Convert the content into a binary string according to UTF-8 encoding, each Chinese is 3 bytes, English and numbers are 2 bytes, each byte is 8 bits, such as the binary corresponding to the two Chinese characters "copyright" It is "111001111000100110001000111001101001110110000011", calculate the number of characters occupied by the information to be encoded, and compare it with the value calculated in the image information sampling. For example, the information to be stored includes the uploader's name, user name, etc. There are 100 characters. According to the calculation that each character occupies 2.5 bytes, 38937600 bits are 38937600/8=4867200 bits, which can store 4867200/2.5=1946880 characters, far exceeding the 100 characters to be stored. Convert the input text to a binary string, and then correspond to the RGB pixel parity of the image. For example, the RGB values of the first three pixels of the picture are (222, 111, 23), (222, 10, 23), (1, 58, 40). The rule is that even values correspond to 0, and odd values correspond to 1, then these three The binary corresponding to the pixel is 011001100. On the contrary, if the binary string of the input text is known, it is only necessary to fine-tune the pixels bit by bit. For example, the binary string bit is 110010, and the first 1 is an odd number, then the first color of the first pixel is 222. It needs to become the nearest odd number 223 or 221, the second bit 1 is an odd number, and the second color value of the first pixel is already an odd number 111, so no adjustment is required, and so on, the encoding information can be completed. of writing. Due to the limited observation accuracy of the human eye, the fine-tuning of the pixels in this way cannot be recognized by the human eye and does not affect the display effect of the picture itself. The total length to be encoded is recorded in the last 5 pixels of the picture according to the encoding rules in the previous step, and recorded in reverse order.

Extract copyright information, extract coding information from pictures, determine whether the content of the pictures has been stolen, and extract hidden information for proof. Take the recorded coding length A from the last 5 pixels at the end of the picture, and obtain the RGB three values corresponding to the first A pixels of the picture according to the length value A, and obtain its parity. According to the even value corresponding to 0 and the odd value corresponding to 1 The rule is converted to a binary string represented by 0s and 1s, and the binary string is decoded in reverse to produce text.

The embodiment of the present invention also discloses a copyright protection device based on picture pixel value steganography, including:

memory for storing computer programs;

The processor is configured to execute the computer program to implement the copyright protection method based on steganography of picture pixel values.

The embodiment of the present invention also discloses a readable storage medium for storing a computer program, wherein when the computer program is executed by a processor, the method for copyright protection based on steganography of picture pixel values is implemented.

The above descriptions are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention shall be included in the protection of the present invention. within the range.

Claims (10)

1. A copyright protection method based on picture pixel value steganography is characterized by comprising the following operations:

sampling pixel points of the picture, and determining the number of characters which can be written into the picture;

when the number of characters of the copyright information text required to be coded is less than that of the characters of the writable picture, converting the copyright information text required to be coded into a binary string;

the method comprises the steps of enabling bits of a binary string to correspond to RGB three values of pixels arranged in sequence one by one, finely adjusting the pixel points bit by bit through the binary string, adding 1 or subtracting 1 to the RGB value of the pixel points with different parity from the RGB value of the corresponding binary string according to the rule that 1 is an odd number and 0 is an even number, adjusting the value parity of the corresponding binary string to be in one-to-one correspondence, and writing the total length needing to be coded into the last N pixel points of a picture in a reverse order;

when the copyright protection is verified, the coding length A is extracted from the last N pixel points of the picture, the RGB three values corresponding to the A pixel points before the picture are obtained according to the length value A, the parity of the RGB values is converted into a binary string containing 1 and 0 according to the rule that 1 is an odd number and 0 is an even number, and the copyright information text is decoded reversely from the binary string.

2. The method according to claim 1, wherein the copyright information is any one or a combination of a file owner, a file creation time, and a unique serial number.

3. The method according to claim 1, wherein the format of the picture is converted into a uniform format when copyright information is written.

4. The method according to claim 1, wherein the number of characters that can be written into the picture is:

when the characters are Chinese, the number of the characters is the number of pixel points/24; when the characters are English or numbers, the number of the characters is the number of pixel points/16.

5. A picture pixel value steganography-based copyright protection system, the system comprising:

the picture pixel sampling module is used for sampling pixels of the picture and determining the number of characters which can be written into the picture;

the copyright information text conversion module is used for converting the copyright information text to be coded into a binary string when the number of the characters of the copyright information text to be coded is less than that of the characters of the writable picture;

the pixel RGB value fine tuning module is used for corresponding the bits of the binary string to RGB three values of pixels arranged in sequence one by one, finely tuning the pixel points bit by bit through the binary string, adding 1 or subtracting 1 to the RGB value of the pixel points with different RGB values corresponding to the binary string according to the rule that 1 is an odd number and 0 is an even number, adjusting the RGB value parity of the pixel points to be in one-to-one correspondence with the numerical value parity of the corresponding binary string, and writing the total length needing to be coded into the last N pixel points of the picture in a reverse sequence;

and the copyright information extraction module is used for extracting the coding length A from the last N pixel points of the picture when the copyright protection is proved, acquiring RGB three values corresponding to the A pixel points before the picture according to the length value A, converting the parity of the RGB values into binary strings containing 1 and 0 according to the rules that 1 is an odd number and 0 is an even number, and reversely decoding the binary strings to obtain the copyright information text.

6. The system of claim 5, wherein the copyright information is any one or a combination of a file owner, a file creation time, and a unique serial number.

7. The system according to claim 5, wherein the format of the picture is converted into a uniform format when the copyright information is written.

8. The system according to claim 5, wherein the number of characters of the writable picture is:

when the characters are Chinese, the number of the characters is the number of pixel points/24; when the characters are English or numbers, the number of the characters is the number of pixel points/16.

9. A copyright protection device based on steganography of picture pixel values, comprising:

a memory for storing a computer program;

a processor for executing the computer program to implement the method of copyright protection based on steganography of picture pixel values as claimed in any one of claims 1 to 4.

10. A readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the method of copy right protection based on steganography of picture pixel values according to any one of claims 1 to 4.

Images