WO2021093648A1

WO2021093648A1 - Watermark information embedding method and apparatus

Info

Publication number: WO2021093648A1
Application number: PCT/CN2020/126360
Authority: WO
Inventors: 刘永亮; 曾晶华
Original assignee: 阿里巴巴集团控股有限公司
Priority date: 2019-11-11
Filing date: 2020-11-04
Publication date: 2021-05-20
Also published as: CN112788342A; CN112788342B

Abstract

Disclosed in the present application are a watermark information embedding method and apparatus. The method comprises: obtaining a carrier object; obtaining watermark information to be embedded; obtaining a target pilot sequence, wherein the target pilot sequence satisfies the following condition: after the target pilot sequence is shifted, an obtained degree of discrimination between a shifted sequence and the target pilot sequence is greater than a preset degree of discrimination; and embedding the watermark information to be embedded and the target pilot sequence into the carrier object. By means of the method, the watermark information completely consistent with the embedded watermark information can be extracted, thereby avoiding the existing problem of failing to ensure the uniqueness of the watermark information because the embedded watermark information is not completely matched with the extracted watermark information.

Description

Method and device for embedding watermark information

This application claims the priority of the Chinese patent application filed on November 11, 2019 with the application number 201911093370.3 and the invention title "A method and device for embedding watermark information", the entire content of which is incorporated into this application by reference.

Technical field

This application relates to the field of computer technology, in particular to a method and device for embedding watermark information. This application also relates to an electronic device. This application also relates to a method for extracting watermark information, a device for extracting watermark information, and an electronic device.

Background technique

With the rapid development of communication technology and multimedia technology, the use of digital watermarking technology to solve the problems of digital media security, copyright protection and authentication has become a research hotspot. Convenient network communication conditions have made illegal distribution increasingly rampant and severely infringed. The basic interests of copyright owners. In order to prevent digital film and television works from being illegally distributed and tampered with, digital watermarking technology, as an effective method to realize copyright protection and anti-counterfeiting tracking, has been widely used in multimedia information, documents, software and other carrier objects.

Digital watermarking technology embeds the watermark information into the carrier object to be protected. Without affecting the normal use of the carrier object, encryption and other methods are used to prevent the content from being copied or changed at will, for example, when there is a piracy or copyright dispute , Can extract watermark information from disputed works as evidence of copyright, thereby safeguarding the rights and interests of the owner; in addition, video watermarking technology can also determine whether the carrier object has been arbitrarily changed through the embedded identification information, thereby tracking the change To achieve the purpose of protecting the basic rights and interests of copyright owners, video watermarking technology is widely used in the copyright protection of digital film and television, information security and other fields.

The signal processing process related to video images includes channel noise, filtering, digital/analog and analog/digital conversion, resampling, image cropping, image shifting, image scale change, and image compression coding. Robustness means that the watermark information can still maintain integrity and be accurately detected after a variety of unintentional or intentional signal processing procedures. The existing watermark embedding and extraction algorithms for video sequences can effectively resist the signal processing processes of image compression, noise addition and filtering, but they lack robustness against geometric attacks. That is, after the attacker performs geometric transformations that do not affect the visual effect of the target image, most watermark detectors cannot detect the embedded watermark information again. For example, the watermark embedding method based on image block, When the above geometric transformation process destroys the synchronization between image blocks, the same block method as when embedding watermark information cannot be obtained, resulting in watermark information extraction errors. Another example is the use of image feature points-based watermark embedding methods to resist geometry Attacks require a lot of searches, and the realization of this method depends on whether the image feature points are accurately located, which increases the complexity of the watermark embedding process.

Since the shape of the image's gray histogram is invariant after geometric transformations such as image rotation and scaling, the watermark embedding method based on the image's gray histogram has achieved good results in resisting geometric attacks. The watermarking algorithm based on gray histogram searches for the average gray value of the image to determine the embedding position of the watermark information. In this process, multiple watermark information needs to be detected and matched with the original watermark information. This determines whether the original watermark information is embedded in the detected image.

However, the above-mentioned watermark embedding method based on image gray histogram will bring a great false positive probability in the process of extracting watermark information, that is, in the process of determining the embedding position of watermark information, due to a large number of watermark sequences It has a high degree of similarity before and after the shift. Therefore, even if the detected watermark information is the watermark information after the shift, when it is matched with the original watermark information, a higher degree of credibility can be obtained Or, use a completely wrong watermark information (non-original watermark information) to match multiple watermark information detected, or extract watermark information with high reliability from the carrier image. Therefore, the watermark information cannot be guaranteed. Uniqueness. For a video frame sequence, since the embedding position of the watermark information also needs to be determined, and multiple detected watermark information needs to be matched with the original watermark information, the uniqueness of the watermark information cannot also be achieved.

Summary of the invention

The embodiment of the present application provides a watermark information embedding method to solve the problem that the uniqueness of the watermark information cannot be guaranteed in the existing watermark information embedding method. Another embodiment of the present application provides a watermark information embedding device and an electronic device. This application also provides a method for extracting watermark information, a device for extracting watermark information, and an electronic device.

The embodiment of the present application provides a method for embedding watermark information, including:

Obtain the carrier object; obtain the watermark information to be embedded; obtain the target preamble sequence; embed the watermark information to be embedded and the target preamble sequence into the carrier object; wherein the target preamble sequence meets the following conditions: After the target preamble sequence is shifted, the degree of discrimination between the obtained shifted sequence and the target preamble sequence is greater than the preset degree of discrimination.

Optionally, the target amble sequence is a binary bit sequence, and the embedding the watermark information to be embedded and the target amble sequence into the carrier object includes: adding the target amble sequence to Obtaining the target bit sequence at the front end or the back end of the watermark information to be embedded; embedding the target bit sequence in the carrier object.

Optionally, the carrier object includes a carrier image, and embedding the target bit sequence into the carrier object includes: obtaining a grayscale histogram of the carrier image; The shape of the grayscale histogram is adjusted to obtain an adjusted grayscale histogram; according to the adjusted grayscale histogram, the grayscale value of the pixel in the carrier image is adjusted to obtain the The target image of the target bit sequence.

Optionally, the adjusting the shape of the grayscale histogram according to the target bit sequence to obtain the adjusted grayscale histogram includes: obtaining the target grayscale interval of the grayscale histogram; For the number of bits of the target bit sequence, the target gray-scale interval is divided to obtain gray-scale sub-intervals corresponding to the number of bits, wherein each gray-scale sub-interval includes at least two adjacent gray-scale sub-intervals. Degree level, the gray level is used to indicate the number of pixels with the same gray value; obtain the number relationship information of the pixels in the gray sub-interval corresponding to the bit value; according to the target bit sequence and the bit The number relationship information of the pixel points in the gray level sub-interval corresponding to the value, the number of pixels contained in at least two adjacent gray levels of the gray-level sub-interval is adjusted to obtain an adjusted gray histogram .

Optionally, the obtaining the target grayscale interval of the grayscale histogram includes: calculating the grayscale mean value of the carrier image; based on the representation range of the grayscale histogram and the number of bits of the target bit sequence Calculating the target grayscale interval of the grayscale histogram according to the grayscale average value.

Optionally, the obtaining the number relationship information of the pixel points in the gray sub-interval corresponding to the bit value includes: when the bit value is 1, obtaining two adjacent gray levels in the gray sub-interval corresponding to the bit value The first comparison relationship between the ratio of the number of pixel points contained in the level and the predetermined embedding intensity; and, when the bit value is 0, obtain two adjacent gray levels in the gray level subinterval corresponding to the bit value The second comparison relationship between the ratio of the number of pixels included in the level and the predetermined embedding intensity;

According to the target bit sequence and the number relationship information of the pixel points in the gray-scale sub-interval corresponding to the bit value, the determination of the pixel points contained in at least two adjacent gray-scale levels in the gray-scale sub-interval is Adjust the quantity to obtain the adjusted gray histogram, including:

The value of the bit to be embedded in the target bit sequence is 1, if the ratio of the number of pixels contained in two adjacent gray levels in the gray subinterval corresponding to the bit value to be embedded is less than the predetermined embedding intensity The relationship between the two is consistent with the first comparison relationship, then the pixels are not adjusted; if the ratio of the number of pixels contained in two adjacent gray levels in the gray sub-interval corresponding to the bit value to be embedded, and If the relationship between the predetermined embedding intensities does not conform to the first comparison relationship, the first number of pixels are selected from the pixels contained in the gray level with a larger gray value and moved to the gray with a smaller gray level. Degree

The value of the bit to be embedded in the target bit sequence is 0, if the ratio of the number of pixels contained in two adjacent gray levels in the gray sub-interval corresponding to the bit value to be embedded is less than the predetermined embedding intensity The relationship between the two is consistent with the second comparison relationship, then the pixels are not adjusted; if the ratio of the number of pixels contained in two adjacent gray levels in the gray sub-interval corresponding to the bit value to be embedded, and If the relationship between the predetermined embedding intensities does not conform to the second comparison relationship, a second number of pixels are selected from the pixels contained in the gray level with a smaller gray value and moved to the gray with a larger gray value. Degree

Wherein, the first number and the second number are calculated based on the predetermined embedding intensity and the number of pixels included in two adjacent gray levels in the gray sub-interval.

Optionally, the selecting a first number of pixel points from the pixel points included in the gray level with a larger gray value to move to the gray level with a smaller gray level includes: adopting a random method from the gray level Select the first number of pixels from the pixels contained in the gray level with the larger value to move to the gray level with the smaller gray level;

The selecting the second number of pixel points from the pixel points included in the gray level with the smaller gray value to move to the gray level with the larger gray value includes: adopting a random method from the smaller gray value The pixel points of the second number selected from the pixel points included in the gray level are moved to a gray level with a larger gray value.

Optionally, before obtaining the grayscale histogram of the carrier image, the method further includes: performing Gaussian filtering processing on the carrier image to obtain the low-frequency signal part of the carrier image; and obtaining the grayscale of the carrier image The histogram includes: obtaining a grayscale histogram of the low-frequency signal part of the carrier image.

Optionally, the obtaining a grayscale histogram of the low-frequency signal part of the carrier image includes: performing block processing on the low-frequency signal part of the carrier image to obtain a block image; calculating the grayscale of the block image Degree average value; according to the average gray value of the block image, the gray level histogram of the block image is statistically obtained.

Optionally, the method further includes: calculating and obtaining the mean square error of the block image; the adjusting the shape of the grayscale histogram according to the target bit sequence to obtain the adjusted grayscale histogram includes: According to the order of the mean square error of the segmented images, the gray histograms of a predetermined number of segmented images are selected; according to the target bit sequence, the grayscale histograms of the predetermined number of segmented images The shape of the image is adjusted to obtain the grayscale histogram of the adjusted block image.

Optionally, the carrier object includes a video frame sequence, and the embedding of the target bit sequence into the carrier object includes: obtaining the first bit value “0” in the target bit sequence. Sub-bit sequence, and obtaining a second sub-bit sequence used to represent the bit value "1" in the target bit sequence, the first sub-bit sequence being different from the second sub-bit sequence; obtaining the video frame The target video image in the sequence, the target video image refers to the video image to be embedded in the target bit sequence; the target bit value of the target video image to be embedded is obtained; the first bit value that will be used to represent the target bit value A sub-bit sequence or the second sub-bit sequence is embedded in the target video image.

Optionally, the first sub-bit sequence satisfies the following condition: after the first sub-bit sequence is shifted, the difference between the obtained shifted bit sequence and the first sub-bit sequence Degree is greater than the preset discrimination degree; and, the second sub-bit sequence satisfies the following conditions: after the second sub-bit sequence is shifted, the obtained shifted bit sequence is the same as the second sub-bit sequence. The degree of discrimination between sequences is greater than the preset degree of discrimination.

Optionally, the first sub-bit sequence includes a first sub-preamble sequence and first sub-watermark information, and the first sub-preamble sequence satisfies the following condition: After processing, the degree of discrimination between the obtained shifted bit sequence and the first sub-amble sequence is greater than the preset degree of discrimination; and, the second sub-bit sequence includes a second sub-amble sequence and a second sub-amble sequence. Sub-watermark information, the second sub-preamble sequence satisfies the following condition: after the second sub-preamble sequence is shifted, the obtained shifted bit sequence and the second sub-preamble sequence The degree of discrimination between the two is greater than the preset degree of discrimination; wherein, the first sub-watermark information is different from the second sub-watermark information.

Optionally, the first sub-preamble sequence is the same as the second sub-preamble sequence.

Optionally, the first sub-preamble sequence and the target preamble sequence are the same bit sequence, and the second sub-preamble sequence and the target preamble sequence are the same bit sequence.

Optionally, the obtaining the target video image in the video frame sequence includes: according to a manner of embedding a bit value in a frame of the video image of the video frame sequence, according to the bit value contained in the target bit sequence The target video image to be embedded in the target bit sequence is obtained from the video frame sequence.

Optionally, the embedding the first sub-bit sequence or the second sub-bit sequence used to represent the target bit value into the target video image includes: obtaining a gray scale of the target video image Degree histogram; according to the first sub-bit sequence or the second sub-bit sequence used to represent the target bit value, the shape of the gray-scale histogram of the target video image is adjusted to obtain an adjustment After the grayscale histogram; according to the adjusted grayscale histogram, adjust the grayscale value of the pixel in the target video image to obtain the first sub-bit sequence or the second sub-bit sequence embedded The target video image of the bit sequence.

Optionally, the carrier object includes a video frame sequence, and embedding the target bit sequence into the carrier object includes: repeatedly embedding the target bit sequence into the video frame sequence according to a predetermined number of embedding times in.

The embodiment of the application also provides a method for extracting watermark information, including:

Obtain the object to be detected; Obtain the reference bit sequence containing the pilot code sequence and the original watermark information. The pilot code sequence satisfies the following conditions: after the pilot code sequence is shifted, the obtained shifted sequence is the same as the The degree of discrimination between the preamble sequences is greater than the preset degree of discrimination. Extracting a target bit sequence from the object to be detected; matching the target bit sequence with the reference bit sequence to determine whether the target bit sequence is a reference bit sequence embedded in the object to be detected.

Optionally, the preamble sequence is set at the front end of the original watermark information, and the extracting the target bit sequence from the object to be detected includes: taking the start position of the embedding range of the target bit sequence as The initial detection point of the target bit sequence is extracted, and the target bit sequence is extracted from the object to be detected.

Optionally, the amble sequence is set at the back end of the original watermark information, and the extracting the target bit sequence from the object to be detected includes: taking the end position of the embedding range of the target bit sequence as The initial detection point of the target bit sequence is extracted, and the target bit sequence is extracted from the object to be detected.

Optionally, the object to be detected includes an image to be detected, and extracting a target watermark sequence from the object to be detected includes: obtaining a grayscale histogram of the image to be detected; based on the grayscale histogram , Extract the target watermark sequence from the object to be detected.

Optionally, the object to be detected includes a video frame sequence, and the extracting a target watermark sequence from the object to be detected includes: obtaining a first bit value "0" in the target bit sequence. A reference sub-bit sequence, and obtaining a second reference sub-bit sequence used to represent a bit value "1" in the target bit sequence, where the first reference sub-bit sequence is different from the second reference sub-bit sequence; Extract multiple target sub-bit sequences from the video image of the video frame sequence; compare the target sub-bit sequence with the first reference sub-bit sequence and the second reference sub-bit sequence to determine The value of the bit embedded in the video image of the video frame sequence is "0" or "1".

Optionally, the extracting the target watermark sequence from the object to be detected based on the gray-level histogram includes: calculating the gray-level average value of the image to be detected; and calculating the gray-level average value according to the gray-level average value. The target grayscale interval of the grayscale histogram; according to the predetermined number of bit values of the watermark sequence embedded in the image to be detected, the target grayscale interval is divided to obtain the bit value of the embedded watermark sequence The number of gray-scale sub-intervals corresponding to the number of gray-scale sub-intervals, each of the gray-scale sub-intervals includes at least two adjacent gray levels, and the gray levels are used to represent the number of pixels with the same gray value; Bit value extraction data; according to the number of pixels in at least two adjacent gray levels of the gray sub-interval and the bit value extraction data, the bit values embedded in the image to be detected are respectively extracted, Obtain the target bit sequence.

Another embodiment of the present application also provides a watermark information embedding device, including:

The carrier object obtaining unit is used to obtain the carrier object;

A unit for obtaining watermark information to be embedded, configured to obtain watermark information to be embedded;

The target amble sequence obtaining unit is configured to obtain a target amble sequence, and the target amble sequence satisfies the following condition: after the target amble sequence is shifted, the obtained shifted sequence is the same as the target amble sequence. The degree of discrimination between code sequences is greater than the preset degree of discrimination;

The information embedding unit is used to embed the watermark information to be embedded and the target preamble sequence into the carrier object.

Another embodiment of the present application further provides an electronic device, including: a processor and a memory, the memory is used to store a watermark information embedding program, and when the program is read and executed by the processor, the following operations are performed: Obtain a carrier object ; Obtain the watermark information to be embedded; Obtain the target amble sequence, the target amble sequence meets the following conditions: after the target amble sequence is shifted, the obtained shifted sequence and the target amble sequence The degree of discrimination between is greater than the preset degree of discrimination; the watermark information to be embedded and the target preamble sequence are embedded in the carrier object.

Another embodiment of the present application also provides a watermark information extraction device, including:

The object to be detected obtaining unit is used to obtain the object to be detected;

The reference bit sequence obtaining unit is configured to obtain a preset reference bit sequence containing a amble sequence and original watermark information, where the amble sequence satisfies the following condition: after shifting the amble sequence, the obtained shift The degree of discrimination between the post-bit sequence and the preamble sequence is greater than the preset degree of discrimination;

A target bit sequence extraction unit, configured to extract a target bit sequence from the object to be detected;

The information matching unit is configured to match the target bit sequence with the reference bit sequence, and determine whether the target bit sequence is a reference bit sequence embedded in the object to be detected.

Another embodiment of the present application further provides an electronic device, including: a processor and a memory, the memory is used to store a watermark information extraction program, and when the program is read and executed by the processor, the following operations are performed: Object; Obtain a preset reference bit sequence containing the amble sequence and the original watermark information, the amble sequence meets the following conditions: after the amble sequence is shifted, the obtained shifted sequence and the The degree of discrimination between the preamble sequences is greater than the preset degree of discrimination; the target bit sequence is extracted from the object to be detected; the target bit sequence is matched with the reference bit sequence to determine whether the target bit sequence is A reference bit sequence embedded in the object to be detected.

Another embodiment of the present application further provides a method for embedding watermark information, including: obtaining a carrier object; obtaining at least two pieces of watermark information to be embedded; adding different target code sequences to the at least two pieces of watermark information to be embedded to obtain at least Two target embedding sequences; embedding the at least two target embedding sequences into the carrier object; wherein, the target preamble sequence satisfies the following condition: after shifting the target preamble sequence, obtain The degree of discrimination between the shifted sequence of and the target preamble sequence is greater than the preset degree of discrimination.

Optionally, the adding different target code sequences to the at least two watermark information to be embedded includes: adding and the watermark information to be embedded to each of the at least two watermark information to be embedded. The target preamble sequence corresponding to the watermark information.

Compared with the prior art, this application has the following advantages:

In the watermark information embedding method provided by this application, when the watermark information is embedded in the carrier object, the target amble sequence is embedded at the same time, and the target amble sequence meets the following conditions: after the target amble sequence is shifted, the obtained shift The degree of discrimination between the post-bit sequence and the target preamble sequence is greater than the preset degree of discrimination. Since the target preamble sequence is more distinguishable after the shift and before the shift, when the watermark information is extracted at the watermark information extraction end, accurate matching of the preamble sequence can be achieved, and the embedding interval of the watermark information can be accurately located. So as to realize the accurate positioning of the watermark information. By using this method, the watermark information that is completely consistent with the embedded watermark information can be extracted, and the existing problem that the uniqueness of the watermark information cannot be guaranteed because the embedded watermark information and the extracted watermark information are not completely matched.

Description of the drawings

Fig. 1 is a flowchart of a watermark information embedding method provided by the first embodiment of the present application;

Fig. 1-A is a schematic diagram of a gray-scale histogram of a carrier image after embedding watermark information provided by the first embodiment of the present application;

Figure 1-B is a schematic diagram of embedding the watermark information to be embedded and the target preamble sequence into a carrier object provided by the first embodiment of the present application;

Figure 2 is a flow chart of the watermark information extraction method provided by the second embodiment of the present application;

Fig. 3 is a block diagram of the watermark information embedding device provided by the third embodiment of the present application;

4 is a schematic diagram of a logical structure of an electronic device provided by a fourth embodiment of the present application;

Fig. 5 is a block diagram of the unit of the watermark information extraction device provided by the fifth embodiment of the present application;

FIG. 6 is a schematic diagram of the logical structure of an electronic device provided by a sixth embodiment of the present application.

Detailed ways

In the following description, many specific details are set forth in order to fully understand this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar promotion without violating the connotation of this application. Therefore, this application is not limited by the specific implementation disclosed below.

For watermark information embedding and extraction scenarios, in order to accurately locate the watermark information during the watermark information extraction process to extract watermark information that is completely consistent with the embedded watermark information, this application provides a watermark information embedding method, For the watermark information embedding device and electronic equipment corresponding to this method, this application also provides a watermark information extraction method, a watermark information extraction device and electronic equipment corresponding to the method. Examples are provided below to describe the method, device, and electronic equipment in detail.

The first embodiment of the present application provides a method for embedding watermark information, which is described below with reference to FIG. 1. The embodiments involved in the following description are used to explain the principle of the method, and are not a limitation of actual use.

As shown in Fig. 1, in step S101, a carrier object is obtained.

Digital watermarking technology is a content-based, non-cryptographic computer information hiding technology, which means that it will not be easy to detect and modify, and can be identified and identified by a designated subject without affecting the use value of the carrier object. The watermark information is directly embedded in the carrier object, or the structure of a specific area of the carrier object is modified. Through these watermark information hidden in the carrier object, the purpose of confirming the content creator, purchaser, transmitting secret information or judging whether the carrier object has been tampered with can be achieved. It is an effective way to protect information security, realize anti-counterfeiting traceability, and copyright protection. Carrier objects include multimedia information, documents, software, etc. In this embodiment, the carrier object is mainly a carrier image or a sequence of video frames.

As shown in Fig. 1, in step S102, the watermark information to be embedded is obtained.

Watermark information is a kind of protection information embedded in a carrier object using a computer algorithm, which is mainly a digital sequence or a bit sequence, for example, watermark information 1010110101 containing ten bit values. The watermark information to be embedded refers to the watermark information preset to be embedded in the above-mentioned carrier image or video frame sequence.

As shown in Fig. 1, in step S103, a target preamble sequence is obtained.

The target amble sequence satisfies the following conditions: after the target amble sequence is shifted, the degree of discrimination between the obtained shifted sequence and the target amble sequence is greater than the preset degree of discrimination, that is, according to the target amble sequence After shifting it forward or backward in the sorting direction, the degree of discrimination between the obtained shifted sequence and the target preamble sequence is greater than the preset degree of discrimination. Here the discrimination degree is greater than the preset discrimination degree, which is used to indicate that the target code sequence has strong discrimination before and after the shift, for example, for two bit sequences with the same number of bits and strong discrimination After aligning the two bit sequences according to the number of bits, they are compared, and the number of the same bit values corresponding to each bit is less than a predetermined threshold. In this embodiment, the target code sequence is preferably a binary bit sequence. For example, the bit sequence 110110010 is shifted forward (the bit sequence moves to the left as a whole) by one bit, and the shifted bit sequence is 100110010, and the last bit sequence is 100110010. The bit corresponds to a random bit value of 0 or 1. After aligning the bit sequence after the shift with the bit sequence before the shift, it can be seen that the difference between the bit sequence after the shift and the bit sequence before the shift The bit sequence is relatively large; the bit sequence is shifted forward by two bits (the bit sequence is shifted to the left as a whole). The shifted bit sequence is 00110010, and the last two bits are random. The shifted bit sequence is the same as the shifted bit sequence. The distinction between the previous bit sequence is relatively large; the bit sequence is reversed (shifted to the right as a whole) by one bit. The last nine bits of the resulting bit sequence have a value of 11011001, and the first bit is random , Assuming that the bit value corresponding to the first bit is 0, the bit sequence after the shift is 0110011001, which is equally distinguishable from the bit sequence before the shift. Therefore, the bit sequence 110110010 can be used as the target preamble sequence.

In the existing watermark information embedding method, in the extraction stage of the watermark information, it is necessary to perform a detailed search on the embedding position of the watermark information. For example, in the process of embedding and extracting the watermark information of a single image, the watermark algorithm based on gray histogram is adopted. It is assumed that the carrier image after the watermark information is embedded is I _W. When the carrier image is subjected to various possible attacks, it will be The carrier image after the attack is I′ _W. In the process of determining the embedding range of the watermark information, _{the gray average value calculated from I′ W and} the gray average value of I _W

Different, so according to the formula

Set the search interval, and traverse the search interval, extract multiple watermark information, and compare it with the input original watermark information. However, in the process of determining the embedding position of the watermark information, due to the large number of watermark sequences in the process of determining the embedded position of the watermark information. Before and after the shift, there is a high degree of similarity. Therefore, even if the detected watermark information is the watermark information after the shift, when the original watermark information is aligned with the number of bits and then matched, it can be obtained High credibility, or, using a completely wrong watermark information (non-original watermark information) to match multiple watermark information detected, it is also possible to extract watermark information with high credibility from the carrier image. There is a false positive problem, that is, inputting unembedded watermark information may also extract watermark information that is very similar to the original watermark information. Therefore, the above method cannot guarantee the uniqueness of the watermark information.

For example, taking two gray levels as a gray sub-interval, a watermark sequence containing 10 bit values is embedded in the embedding range of the gray histogram: 1010110101, as shown in Figure 1-A (the image of the carrier image after the watermark is embedded) The schematic diagram of the grayscale histogram) shows: in Figure 1-A, the grayscale with a grayscale value of 12 is used as the left end point (the initial embedding position of the watermark information) of the original watermark information. When extracting the watermark information , It is necessary to search the above-mentioned left endpoint according to the mean value of the carrier image, the length of the watermark information, and the size of the group, and extract the watermark information in order based on the left endpoint. However, when the gray level with the gray value of 15 is searched, the bit sequence opposite to the original watermark information can be extracted, and when the gray level with the gray value of 16 is searched, the original watermark information can be extracted to the left After shifting the bit sequence, the bit value corresponding to the first nine bits of the bit sequence is: 01011010, and the last bit bit corresponds to the random bit value 0 or 1. Through comparison, it can be seen that the above-mentioned bit sequence opposite to the original watermark information and the bit sequence obtained after the original watermark information is shifted to the left have a higher degree of matching with the original watermark information, so they all have a higher degree of credibility, for example, If the original watermark information is very similar to the above-mentioned shifted watermark information, in the process of searching the left endpoint, the bit sequence obtained after the original watermark information is shifted is matched with the original watermark information, and a high credibility can be obtained. Degree of watermark information cannot guarantee the uniqueness of watermark information.

Based on the foregoing reasons, in the embodiments of the present application, before embedding the above-mentioned watermark information to be embedded into the carrier object, the above-mentioned target amble sequence is obtained first, and the amble sequence is used to locate the above-mentioned watermark information to be embedded. For example, the watermark information to be embedded is 1010110101, and the target preamble sequence is 1100110010.

As shown in FIG. 1, in step S104, the watermark information to be embedded and the target preamble sequence are embedded into the carrier object.

After obtaining the watermark information to be embedded and the target code sequence in the above steps, as shown in Figure 1-B, this step is used to embed the watermark information to be embedded and the target code sequence into the carrier object to obtain Embed the target object to be embedded with the watermark information and the target preamble sequence. The process specifically includes: adding the target preamble sequence to the front or back end of the watermark information to be embedded to obtain a target bit sequence; embedding the target bit sequence into the carrier object, and in this process, The target preamble sequence can be used as the unique identification code of the watermark information to be embedded. For example, adding the target code sequence 110110010 to the front end of the watermark information 1010110101 to be embedded, the obtained target bit sequence is 110110010 1010110101, and the target bit sequence is embedded into the carrier object according to a predetermined watermark information embedding method.

In this embodiment, the aforementioned carrier object may refer to a single frame of carrier image or a sequence of video frames.

When the carrier object is a carrier image, the foregoing process of embedding the target bit sequence into the carrier object may include the following content:

A. Obtain the grayscale histogram of the carrier image.

In this embodiment, the process of obtaining the grayscale histogram of the carrier image is specifically as follows:

First, the following formula is used to perform Gaussian filtering processing on the carrier image to obtain the low-frequency signal part of the carrier image: I _Low (x, y) = G (x, y, σ) * I (x, y), where , I represents the carrier image, I _Low represents the low-frequency signal part of the carrier image obtained after the carrier image is Gaussian filtering, and G represents the Gaussian low-pass filter; after Gaussian filtering is performed on the carrier image, the watermark information can filter, Attacks such as adding noise have good robustness, that is, embedding watermark information in the low-frequency signal part of the carrier image can make the watermark information more robust. For the two-dimensional carrier image, the Gaussian low-pass filter is defined as follows:

Among them, σ is the standard deviation of the Gaussian distribution.

Secondly, the gray histogram of the low-frequency signal part of the carrier image is obtained, that is, the gray histogram of the statistics I _Low . The gray histogram is the statistical feature of the image, which depends on the gray level contained in the image. The number of pixels has nothing to do with the specific location of the pixels. The shape of the grayscale histogram is used to indicate the ratio of the number of pixels contained in each grayscale to the total number of pixels in the image. When an image is geometrically attacked, the size and spatial position of the image will change accordingly, but the ratio of the number of pixels contained in each gray level to the total number of pixels in the image does not change, that is, the gray value The shape of the square graph can remain stable under geometric attacks, so it can be used in watermarking algorithms against geometric attacks.

B. Adjust the shape of the grayscale histogram according to the target bit sequence to obtain an adjusted grayscale histogram.

In this embodiment, the process specifically includes the following content:

B-1. Obtain a target grayscale interval of the grayscale histogram, where the target grayscale interval is used to indicate the embedding range of the target bit sequence. The process of obtaining the target gray-scale interval of the gray-scale histogram is specifically: calculating and obtaining the gray-scale mean value of the carrier image

Then use the following formula to calculate and obtain the target gray-scale interval B of the gray-scale histogram:

Among them, the positive decimal λ satisfies the following conditions: B cannot exceed the representation range of the gray histogram, and B satisfies the number of bits of the target bit sequence.

B-2. Divide the target gray-scale interval according to the number of bits of the target bit sequence to obtain gray-scale sub-intervals corresponding to the number of bits, wherein each gray-scale sub-interval includes at least Two adjacent gray levels, where the gray levels are used to indicate the number of pixels with the same gray value;

B-3. Obtain the quantity relationship information of the pixel points in the gray sub-interval corresponding to the bit value. In this embodiment, the process of obtaining the quantity relationship information of pixels in the gray sub-interval corresponding to the bit value is specifically: when the bit value is 1, obtain two adjacent gray sub-intervals corresponding to the bit value. The ratio of the number of pixels contained in the gray level and the first comparison relationship with the predetermined embedding intensity; and, when the bit value is 0, obtain two adjacent gray sub-intervals corresponding to the bit value The second comparison relationship between the ratio of the number of pixels contained in the gray level and the predetermined embedding intensity. For example, the number relationship information of pixels in the gray-scale sub-interval corresponding to the bit value is:

Among them, w(i) represents the bit value, a and b respectively represent the number of pixels contained in two adjacent gray levels in the gray sub-interval, and a represents the gray level bin1 with a smaller gray value. The number of pixels included, b represents the number of pixels included in the gray level bin2 with a larger gray value, and T represents the predetermined embedding intensity.

B-4. According to the target bit sequence and the number relationship information of the pixel points in the gray sub-interval corresponding to the bit value, the pixels corresponding to at least two adjacent gray levels in the gray sub-interval Adjust the number of points to obtain the adjusted gray histogram.

In this embodiment, the process of adjusting the pixel points is specifically: when the value of the bit to be embedded in the target bit sequence is 1, if the value of the bit to be embedded corresponds to two adjacent gray levels in the gray sub-interval The relationship between the ratio of the number of pixel points included in the level and the predetermined embedding intensity meets the above-mentioned first comparison relationship, the pixel points are not adjusted; if the two phases in the gray sub-interval corresponding to the bit value to be embedded The relationship between the ratio of the number of pixels contained in adjacent gray levels and the predetermined embedding intensity does not conform to the above-mentioned first comparison relationship, and the first comparison is selected from the pixels contained in the gray level with the larger gray value. A number of pixels are moved to a gray level with a smaller gray level;

When the value of the bit to be embedded in the target bit sequence is 0, if the ratio of the number of pixels contained in two adjacent gray levels in the gray sub-interval corresponding to the bit value to be embedded is equal to the predetermined embedding value The relationship between the intensities conforms to the second comparison relationship, then the pixels are not adjusted; if the bit value to be embedded corresponds to the ratio of the number of pixels contained in two adjacent gray levels in the gray sub-interval , The relationship between the predetermined embedding intensity and the predetermined embedding intensity does not meet the second comparison relationship, and the second number of pixels are selected from the pixels contained in the gray level with the smaller gray value and moved to the larger gray value In the gray scale.

For example, if the value of the bit to be embedded in the target bit sequence is 1, if a/b≥T, there is no need to adjust the pixel points. If a/b<T, I1 pixel points are selected from the pixels contained in the gray level bin2 Move to the gray level bin1; the value of the bit to be embedded in the target bit sequence is 0, if b/a≥T, there is no need to adjust the pixel point, if b/a<T, from the gray level bin1 contained I0 pixel points are selected from the pixel points and moved to the gray level bin2.

Wherein, the first number and the second number are calculated according to the embedding intensity and the number of pixels contained in two adjacent gray levels in the gray sub-interval, for example:

In this embodiment, the above-mentioned selection of the first number of pixels from the pixels contained in the gray level with a larger gray value to move to the gray level with a smaller gray level may refer to: adopting a random method Select the first number of pixels from the pixels contained in the gray level with the larger gray value to move to the gray level with the smaller gray level; Selecting the second number of pixels from the pixels to move to the gray level with a larger gray value can mean: randomly selecting the second number from the pixels contained in the gray level with a smaller gray value The pixel points of move to the gray level with larger gray value. For example, using a random method to select I1 pixels from the pixels contained in the gray level bin2 to move to the gray level bin1, and using a random method to select I0 pixels from the pixels contained in the gray level bin1 to move To the gray level bin2.

The following figure is a schematic diagram of the specific adjustment of embedding a bit corresponding to the four relative sizes of a and b. Among them, I ₁ and I ₀ are the minimum number of pixels to be adjusted, (a)1<a/b<T;(b)a/b>T>1;(c)1<b/a<T;(d)b/a>T>1

Said selecting I1 pixel points from the pixel points included in gray level bin2 and moving to gray level bin1, or said selecting I0 pixel points from the pixel points included in gray level bin1 and moving to gray level In level bin2, it is implemented in the following way:

Among them, M is the width of the gray level, f ₁ (i) is the i-th pixel selected from bin1, and f ₂ (j) is the j-th pixel selected from bin2.

It should be noted that the above process of obtaining the gray-scale histogram of the low-frequency signal part of the carrier image may also be the following content: block the low-frequency signal part of the carrier image to obtain the block image, for example, the carrier image The low-frequency signal part of the image is divided into blocks according to 8*8; calculate the average gray value of each block image; according to the gray average value of each block image, statistically obtain the gray histogram of each block image, and calculate The mean square error of each block image, and the mean square error of the block image can be used to indicate the smoothness of the block image. The adjusting the shape of the gray-scale histogram according to the target code sequence to obtain the adjusted gray-scale histogram may refer to: for the gray-scale sub-intervals for which the number of pixels needs to be adjusted, according to the block image Select a predetermined number of block images with a larger mean square error, and adjust the shape of the grayscale histogram of the block image with a larger mean square error according to the above target code sequence to obtain The adjusted grayscale histogram of the segmented image. In this way, non-smooth areas of the carrier image can be selected in advance for modification, so as to improve the quality of the carrier image.

C. Adjust the gray values of pixels in the carrier image according to the adjusted gray histogram to obtain a target image embedded with the target bit sequence.

When the aforementioned carrier object is a video frame sequence, the process of embedding the target bit sequence into the carrier object may include the following:

A1. Obtain a first sub-bit sequence representing the bit value "0" in the target bit sequence, and obtain a second sub-bit sequence representing the bit value "1" in the target bit sequence, so The first sub-bit sequence is different from the second sub-bit sequence.

In this embodiment, the above-mentioned first sub-bit sequence and the second sub-bit sequence can be set in the manner of the above-mentioned target amble sequence, and the purpose is to embed the first sub-bit sequence or the second sub-bit sequence in the video frame sequence. In the video image of the second sub-bit sequence, in the detection link of the video image, the embedded first sub-bit sequence or the second sub-bit sequence can be accurately detected through the correlation characteristics of the target code sequence. For example, the first sub-bit sequence may satisfy the following condition: after the first sub-bit sequence is shifted, the degree of discrimination between the obtained shifted bit sequence and the first sub-bit sequence Greater than the preset degree of discrimination, that is, the difference between the obtained shifted sequence and the first sub-bit sequence after shifting it forward or backward according to the ordering direction of the first sub-bit sequence Degree is greater than the preset discrimination degree; the second sub-bit sequence may satisfy the following condition: after the second sub-bit sequence is shifted, the obtained shifted bit sequence is the same as the second sub-bit sequence The degree of discrimination between is greater than the preset degree of discrimination, that is, after the second sub-bit sequence is shifted forward or backward according to the ordering direction of the second sub-bit sequence, the obtained shifted sequence is compared with the second sub-bit sequence. The degree of discrimination between sequences is greater than the preset degree of discrimination.

In this embodiment, the above-mentioned first sub-bit sequence and second sub-bit sequence may also be set as follows:

The first sub-bit sequence includes a first sub-preamble sequence and first sub-watermark information, and the first sub-preamble sequence may satisfy the following conditions: after shifting the first sub-preamble sequence, the first sub-preamble sequence is obtained The degree of discrimination between the shifted bit sequence and the first sub-amble sequence is greater than the preset degree of discrimination, that is, the first sub-amble sequence is shifted forward or backward according to the ordering direction of the first sub-amble sequence Then, the degree of discrimination between the obtained shifted sequence and the first sub-amble sequence is greater than the preset degree of discrimination; the second sub-bit sequence includes a second sub-amble sequence and second sub-watermark information, The second sub-amble sequence may satisfy the following condition: after the second sub-amble sequence is shifted, the degree of discrimination between the obtained shifted bit sequence and the second sub-amble sequence is greater than The preset degree of discrimination, that is, after the second sub-amble sequence is shifted forward or backward according to the ordering direction of the second sub-amble sequence, the obtained shifted sequence is between the second sub-amble sequence The degree of discrimination is greater than the preset degree of discrimination; wherein, the above-mentioned first sub-watermark information is different from the second sub-watermark information. The first sub-preamble sequence and the second sub-preamble sequence may be the same bit sequence, and the first sub-preamble sequence and the second sub-preamble sequence may be the same bit as the target sequence. Sequence, for example, 0: 1100110010 1100110010; 1: 1100110010 0011001101. In this way, the first sub-bit sequence or the second sub-bit sequence is set. When the first sub-bit sequence or the second sub-bit sequence is embedded in the video image in the video frame sequence, it is embedded in the carrier image to be embedded. The watermark information is in the same manner as the target preamble sequence, and its purpose is that for a single frame of video image used to embed the first sub-bit sequence or the second sub-bit sequence in the video frame sequence, each can be adjusted according to a unified preamble sequence. The watermark information embedded in the carrier object (video image and video frame sequence) is located.

B1. Obtain a target video image in the video frame sequence, where the target video image refers to the video image to be embedded in the target bit sequence. In this embodiment, according to the manner of embedding a bit value in a frame of the video image of the video frame sequence, according to the number of bit values contained in the target bit sequence, the video frame sequence to be embedded is obtained from the video frame sequence. The target video image of the target bit sequence. For example, if the target bit sequence is 1100110010 1010110101, the 20 bit values contained in the target bit sequence need to be embedded in 20 frames of video images of the video frame sequence. The 20 frames of video images are used to embed the target bit sequence. Target video image.

C1: Obtain the target bit value of the target video image to be embedded. For example, if the current target video image to be embedded is the 9th frame of the above 20 target video images, the value of the target bit to be embedded is 1.

D1, embedding the first sub-bit sequence or the second sub-bit sequence used to represent the target bit value into the target video image. For example, the bit value "1" corresponds to the second sub-bit sequence, so the second sub-bit sequence is embedded in the 9th frame image. In this embodiment, the embedding process includes the following: obtaining a grayscale histogram of the target video image; according to the first sub-bit sequence or the second sub-bit sequence used to represent the target bit value Bit sequence, adjust the shape of the grayscale histogram of the target video image to obtain the adjusted grayscale histogram; according to the adjusted grayscale histogram, calculate the grayscale of the pixels in the target video image The degree value is adjusted to obtain a target video image embedded with the first sub-bit sequence or the second sub-bit sequence. For the implementation details of the embedding process, please refer to the above process of embedding the target bit sequence in the carrier image, which will not be repeated here.

In this embodiment, in order to resist frame rate conversion attacks (for example, the video frame sequence is changed from 50 frames to 25 frames, and the watermark information embedded in the video frame sequence is lost due to the loss of the video image), the target bit sequence is embedded In the process of the video frame sequence, embedding may be repeated multiple times in the time sequence, that is, the target bit sequence is repeatedly embedded into the video frame sequence according to a predetermined number of embedding times. For example, after embedding the target bit sequence 110110010-1010110101 in the video frame sequence, the above embedding process is repeated for each frame of the video image of the video frame sequence multiple times subsequently.

As shown in the following table, the values in the following table represent the watermark information detection result obtained after matching the input reverse watermark sequence of the original watermark information with the extracted watermark information in the watermark information detection process. This embodiment Compared with the provided watermark information embedding method using the target amble sequence, the watermark information embedding method obtained by the former is significantly lower than the latter.

Therefore, the watermark information embedding method using the target pilot code sequence has a low false detection rate in the watermark information detection process, and it can be judged whether the input watermark information is the original watermark information embedded in the carrier object.

In this embodiment, the above-mentioned target preamble sequence can correspond to multiple watermark information, that is, when multiple watermark information is embedded in the same carrier object or embedded in multiple carrier objects, the same target preamble sequence can be combined It is added to the front end or the back end of the multiple watermark information to respectively locate the multiple watermark information. For example, multiple watermark information of the same category can correspond to a target code sequence, or, according to different use requirements and applicable scenarios of the watermark information (for example, the different encryption levels provided by the watermark information), match them with the use requirements The target preamble sequence corresponding to the applicable scenario.

In this embodiment, the above method can also be used to simultaneously embed multiple watermark information for a carrier object. The process specifically includes the following content: obtaining a carrier object, which can be a carrier image or a sequence of video frames; obtaining at least two pending objects. Embedding watermark information, the at least two watermark information to be embedded may be multiple watermark information of different categories, or multiple watermark information of the same category; adding different target code sequences to the at least two watermark information to be embedded , To obtain at least two target embedding sequences, for example, for different types of watermark information, a predetermined target amble sequence corresponding to the type of watermark information can be added to them, wherein the target amble sequence is shifted After bit processing, the degree of discrimination between the obtained shifted sequence and the target preamble sequence is greater than the preset degree of discrimination; the at least two target embedding sequences are embedded in the carrier object. The foregoing manner of adding different target code sequences to the at least two watermark information to be embedded may specifically be: adding a predetermined and the watermark information to be embedded to each of the at least two watermark information to be embedded. The corresponding target preamble sequence. Through this method, multiple watermark information containing positioning information (target code sequence) can be embedded in a carrier object. Correspondingly, multiple watermark information can be accurately extracted in the process of extracting the watermark, so as to achieve multiple encryption Effect.

In the watermark information embedding method provided in this embodiment, when the watermark information is embedded in the carrier object, the target amble sequence is embedded at the same time, and the target amble sequence meets the following conditions: The degree of discrimination between the shifted sequence and the target preamble sequence is greater than the preset degree of discrimination. Since the target preamble sequence is more distinguishable after the shift and before the shift, when the watermark information is extracted at the watermark information extraction end, accurate matching of the preamble sequence can be achieved, and the embedding interval of the watermark information can be accurately located. So as to realize the accurate positioning of the watermark information. By using this method, the watermark information that is completely consistent with the embedded watermark information can be extracted, avoiding the existing problem that the uniqueness of the watermark information cannot be guaranteed because the embedded watermark information does not completely match the extracted watermark information.

Corresponding to the watermark embedding method provided in the foregoing first embodiment, the second embodiment of the present application provides a watermark information extraction method, which is described below with reference to FIG. 2.

As shown in Fig. 2, in step S201, the object to be detected is obtained. The object to be detected may be an image to be detected or a sequence of video frames to be detected.

As shown in FIG. 2, in step S202, a reference bit sequence containing a amble sequence and original watermark information is obtained, and the amble sequence satisfies the following condition: after shifting the amble sequence, the obtained shift is The degree of discrimination between the post-bit sequence and the preamble sequence is greater than the preset degree of discrimination. The amble sequence may be set at the front end or the back end of the original watermark information.

This step is used to extract the watermark sequence embedded in the object to be detected, and may specifically include the following content: obtaining the grayscale histogram of the image to be detected; calculating the average gray value of the image to be detected; according to the gray average value Calculate and obtain the target grayscale interval of the grayscale histogram; divide the target grayscale interval according to the predetermined number of bits embedded in the watermark sequence to obtain the grayscale subinterval corresponding to the number of bits, each Each of the gray-scale sub-intervals includes at least two adjacent gray-levels; the extraction strategy corresponding to the bit value is obtained; and the number of pixels in the at least two adjacent gray-levels of the gray-level sub-intervals and the bit The extraction strategy corresponding to the value extracts the watermark sequence of the image to be detected, and obtains the watermark sequence embedded in the image to be detected.

As shown in FIG. 2, in step S203, a target bit sequence is extracted from the object to be detected.

In this embodiment, the method of extracting the target bit sequence is related to the predetermined original watermark information and the setting method of the amble sequence. When the amble sequence is set at the front end of the original watermark information, the target bit sequence is extracted according to the forward direction. Extraction is performed in the manner of extraction, that is, the start position of the embedding range of the target bit sequence is used as the initial detection point for extracting the target bit sequence, and the target bit sequence is sequentially extracted from the object to be detected; When the sequence is set at the back end of the original watermark information, the target bit sequence is extracted according to the reverse extraction method, that is, the end position of the embedding range of the target bit sequence is used as the initial point for extracting the target bit sequence. For detection points, the target bit sequence is sequentially extracted from the object to be detected.

In this embodiment, when the object to be detected is an image to be detected, the foregoing extraction of the target bit sequence from the object to be detected specifically includes the following content:

First, obtain the grayscale histogram of the image to be detected.

Secondly, based on the above gray histogram, the target bit sequence is extracted from the object to be detected. For example, calculating the average gray value of the image to be detected; calculating the target grayscale interval of the grayscale histogram according to the average gray value; according to a predetermined bit value of the watermark sequence embedded in the image to be detected The target gray-scale interval is divided to obtain the gray-scale sub-intervals corresponding to the number of bit values of the embedded watermark sequence, each of the gray-scale sub-intervals includes at least two adjacent gray levels The gray level is used to represent the number of pixels with the same gray value; to obtain a predetermined bit value to extract data; according to the number of pixels in at least two adjacent gray levels in the gray sub-interval, and The bit value extracting data, respectively extracting the bit value embedded in the image to be detected, to obtain the target bit sequence. This process is the reverse process of embedding watermark information in the carrier image in the first embodiment of this application. Its implementation is similar to the existing watermark information extraction process. For the specific implementation process, please refer to the existing watermark information extraction process, here No longer.

When the object to be detected is a video frame sequence, extracting the target bit sequence from the object to be detected above specifically includes the following content:

First, obtain a first reference sub-bit sequence used to represent the bit value "0" in the target bit sequence, and obtain a second reference sub-bit sequence used to represent the bit value "1" in the target bit sequence , The first reference sub-bit sequence is different from the second reference sub-bit sequence. For the relevant content of the first reference sub-bit sequence and the second reference sub-bit sequence here, please refer to the first sub-bit sequence and the second sub-bit sequence in the above-mentioned first embodiment, which will not be repeated here.

Secondly, multiple target sub-bit sequences are extracted from the video images of the video frame sequence. Each video image embedded with the original watermark information or amble sequence corresponds to a target sub-bit sequence. For example, for each video frame of the video frame sequence, the watermark information is extracted separately, and the maximum election method is used to determine the sub-bit sequence. The information extracted from each video frame of the video frame sequence to be detected is counted, and finally multiple target sub-bit sequences consistent with the number of bit values in the target bit sequence are obtained.

Finally, the target sub-bit sequence is compared with the first reference sub-bit sequence and the second reference sub-bit sequence to determine the value of the bit embedded in the video image of the video frame sequence. 0" or "1", and so on, until all the bit values embedded in the video frame sequence are extracted, and the extracted bit values constitute the target bit sequence.

It should be noted that the execution order of the above step S202 and step S203 is not limited, that is, after the target bit sequence is extracted from the object to be detected, the reference bit sequence including the amble sequence and the original watermark information can be obtained.

As shown in FIG. 2, in step S204, the target bit sequence is matched with the reference bit sequence to determine whether the target bit sequence is a reference bit sequence embedded in the object to be detected. For example, according to the amble sequence in the reference bit sequence set at the front or back end of the original watermark information, the target bit sequence is combined with the target bit sequence in a forward matching or reverse matching method (consistent with the above-mentioned forward extraction or reverse extraction). The reference bit sequence is matched. Since the information obtained after the amble sequence is shifted forward or reversely according to its sorting direction, the degree of discrimination from the amble sequence is greater than the preset degree of discrimination, therefore, the reference bit sequence and the extraction The exact match of the target bit sequence.

The watermark information extraction method provided in this embodiment can achieve accurate matching of the preamble sequence when extracting the watermark information, and can accurately locate the embedding interval of the watermark information, thereby achieving accurate positioning of the watermark information. By using this method, the watermark information that is completely consistent with the embedded watermark information can be extracted, and the existing problem that the uniqueness of the watermark information cannot be guaranteed because the embedded watermark information and the extracted watermark information are not completely matched.

The above first embodiment provides a watermark information embedding method. Correspondingly, the third embodiment of this application also provides a watermark information embedding device. Since the device embodiment is basically similar to the method embodiment, the description is It is relatively simple. For details of related technical features, please refer to the corresponding description of the method embodiment provided above. The following description of the device embodiment is only illustrative.

Please refer to FIG. 3 to understand this embodiment. FIG. 3 is a unit block diagram of the apparatus provided in this embodiment. As shown in FIG. 3, the apparatus provided in this embodiment includes:

The carrier object obtaining unit 301 is used to obtain the carrier object;

The to-be-embedded watermark information obtaining unit 302 is configured to obtain the to-be-embedded watermark information;

The target amble sequence obtaining unit 303 is configured to obtain a target amble sequence, which satisfies the following condition: after performing shift processing on the amble sequence, the obtained shifted sequence and the amble are The degree of discrimination between sequences is greater than the preset degree of discrimination;

The information embedding unit 304 is configured to embed the watermark information to be embedded and the target preamble sequence into the carrier object.

The target preamble sequence is a binary bit sequence, and the embedding the watermark information to be embedded and the target preamble sequence into the carrier object includes: adding the target preamble sequence to the to be embedded The front end or the back end of the watermark information obtains a target bit sequence; and the target bit sequence is embedded in the carrier object.

The carrier object includes a carrier image, and the embedding of the target bit sequence into the carrier object includes: obtaining a grayscale histogram of the carrier image; and calculating the grayscale histogram according to the target bit sequence The shape of the graph is adjusted to obtain an adjusted gray histogram; according to the adjusted gray histogram, the gray values of pixels in the carrier image are adjusted to obtain the target bit sequence embedded Target image.

The adjusting the shape of the grayscale histogram according to the target bit sequence to obtain the adjusted grayscale histogram includes: obtaining the target grayscale interval of the grayscale histogram; and according to the target bit sequence. The number of bits in the sequence is divided into the target gray-scale interval to obtain gray-scale sub-intervals corresponding to the number of bits, wherein each of the gray-scale sub-intervals includes at least two adjacent gray levels, so The gray level is used to indicate the number of pixels with the same gray value; obtain the number relationship information of the pixels in the gray sub-interval corresponding to the bit value; according to the target bit sequence and the gray corresponding to the bit value The number relationship information of the pixel points in the degree sub-interval is adjusted to the number of pixels contained in at least two adjacent gray levels of the gray-level sub-interval to obtain an adjusted gray-level histogram.

The obtaining the target gray scale interval of the gray scale histogram includes: calculating the gray scale mean value of the carrier image; based on the representation range of the gray scale histogram and the number of bits of the target bit sequence, according to the The gray-level mean value calculates the target gray-level interval of the gray-level histogram.

The obtaining the quantity relationship information of the pixels in the gray sub-interval corresponding to the bit value includes: when the bit value is 1, obtaining the information contained in two adjacent gray levels in the gray sub-interval corresponding to the bit value The first comparison relationship between the ratio of the number of pixel points and the predetermined embedding intensity; and, when the bit value is 0, obtain the two adjacent gray levels contained in the gray sub-interval corresponding to the bit value The second comparison relationship between the ratio of the number of pixels and the predetermined embedding intensity;

The selecting the first number of pixel points from the pixel points included in the gray level with the larger gray value to move to the gray level with the smaller gray level includes: randomly selecting from the gray level with the larger gray value Select the first number of pixels from the pixels included in the gray level to move to a gray level with a smaller gray level;

Before obtaining the grayscale histogram of the carrier image, the method further includes: performing Gaussian filtering processing on the carrier image to obtain the low-frequency signal part of the carrier image; and obtaining the grayscale histogram of the carrier image includes : Obtain a grayscale histogram of the low-frequency signal part of the carrier image.

The obtaining the grayscale histogram of the low-frequency signal portion of the carrier image includes: performing block processing on the low-frequency signal portion of the carrier image to obtain a block image; calculating the gray average value of the block image; The gray-level average value of the block image is statistically obtained to obtain the gray-level histogram of the block image.

It also includes: calculating the mean square error of the block image; said adjusting the shape of the grayscale histogram according to the target bit sequence to obtain the adjusted grayscale histogram, including: according to the division Select the grayscale histograms of a predetermined number of block images in descending order of the mean square error of the block images; adjust the shape of the grayscale histograms of the predetermined number of block images according to the target bit sequence , To obtain the gray histogram of the adjusted block image.

The carrier object includes a video frame sequence, and embedding the target bit sequence into the carrier object includes: obtaining a first sub-bit sequence that is used to represent a bit value "0" in the target bit sequence, And obtaining a second sub-bit sequence for representing the bit value "1" in the target bit sequence, where the first sub-bit sequence is different from the second sub-bit sequence; and obtaining the target in the video frame sequence Video image, the target video image refers to the video image to be embedded in the target bit sequence; the target bit value of the target video image to be embedded is obtained; the first sub-bit sequence to be used to represent the target bit value Or the second sub-bit sequence is embedded in the target video image.

The first sub-bit sequence satisfies the following condition: after the first sub-bit sequence is shifted, the degree of discrimination between the obtained shifted bit sequence and the first sub-bit sequence is greater than a preset Degree of discrimination; and, the second sub-bit sequence satisfies the following conditions: after the second sub-bit sequence is shifted, the obtained shifted bit sequence is between the second sub-bit sequence The degree of discrimination is greater than the preset degree of discrimination.

The first sub-bit sequence includes a first sub-preamble sequence and first sub-watermark information, and the first sub-preamble sequence satisfies the following condition: after performing shift processing on the first sub-preamble sequence, obtain The degree of discrimination between the shifted bit sequence and the first sub-amble sequence is greater than the preset degree of discrimination; and, the second sub-bit sequence includes a second sub-amble sequence and second sub-watermark information, The second sub-amble sequence satisfies the following condition: after the second sub-amble sequence is shifted, the degree of discrimination between the obtained shifted bit sequence and the second sub-amble sequence Greater than the preset discrimination degree; wherein, the first sub-watermark information is different from the second sub-watermark information.

The first sub-preamble sequence is the same as the second sub-preamble sequence.

The first sub-preamble sequence and the target preamble sequence are the same bit sequence, and the second sub-preamble sequence and the target preamble sequence are the same bit sequence.

The obtaining the target video image in the video frame sequence includes: embedding a bit value in a frame of the video frame sequence of the video frame sequence, and according to the number of bit values contained in the target bit sequence, from Obtain a target video image to be embedded in the target bit sequence from the video frame sequence.

The embedding the first sub-bit sequence or the second sub-bit sequence used to represent the target bit value into the target video image includes: obtaining a grayscale histogram of the target video image; According to the first sub-bit sequence or the second sub-bit sequence used to represent the target bit value, the shape of the grayscale histogram of the target video image is adjusted to obtain the adjusted grayscale Histogram; according to the adjusted gray histogram, adjust the gray value of the pixel in the target video image to obtain the target embedded with the first sub-bit sequence or the second sub-bit sequence Video image.

The carrier object includes a video frame sequence, and the embedding of the target bit sequence into the carrier object includes: repeating the embedding of the target bit sequence into the video frame sequence according to a predetermined number of embedding times.

In the above-mentioned embodiments, a watermark information embedding method and a watermark information embedding device are provided. In addition, the fourth embodiment of the present application also provides an electronic device. Since the electronic device embodiment is basically similar to the method embodiment, Therefore, the description is relatively simple. For details of related technical features, please refer to the corresponding description of the method embodiment provided above. The following description of the electronic device embodiment is only illustrative. An example of the electronic device is as follows:

Please refer to FIG. 4 to understand this embodiment. FIG. 4 is a schematic diagram of the electronic device provided in this embodiment.

As shown in FIG. 4, the electronic device includes: a processor 401; a memory 402;

The memory 402 is configured to store a watermark information embedding program, and when the program is read and executed by the processor, the following operations are performed:

Obtain the carrier object;

Obtain the watermark information to be embedded;

Obtain a target preamble sequence, the target preamble sequence satisfies the following condition: after the shift processing is performed on the preamble sequence, the degree of discrimination between the obtained shifted sequence and the preamble sequence is greater than the preset distinction degree;

Embedding the watermark information to be embedded and the target preamble sequence into the carrier object.

The above-mentioned second embodiment provides a watermark information extraction method. Correspondingly, the fifth embodiment of the present application also provides a watermark information extraction device. Since the device embodiment is basically similar to the method embodiment, the description is as follows: It is relatively simple. For details of related technical features, please refer to the corresponding description of the method embodiment provided above. The following description of the device embodiment is only illustrative.

Please refer to FIG. 5 to understand this embodiment. FIG. 5 is a unit block diagram of the apparatus provided in this embodiment. As shown in FIG. 5, the apparatus provided in this embodiment includes:

The object to be detected obtaining unit 501 is configured to obtain the object to be detected;

The reference bit sequence obtaining unit 502 is configured to obtain a preset reference bit sequence containing a amble sequence and original watermark information, where the amble sequence satisfies the following conditions: the amble sequence is obtained after shifting the amble sequence The degree of discrimination between the shifted sequence and the preamble sequence is greater than the preset degree of discrimination;

The target bit sequence extraction unit 503 is configured to extract a target bit sequence from the object to be detected;

The information matching unit 504 is configured to match the target bit sequence with the reference bit sequence, and determine whether the target bit sequence is a reference bit sequence embedded in the object to be detected.

The preamble sequence is set at the front end of the original watermark information, and the extraction of the target bit sequence from the object to be detected includes: taking the start position of the embedding range of the target bit sequence as the extraction target The initial detection point of the bit sequence extracts the target bit sequence from the object to be detected.

The preamble sequence is set at the back end of the original watermark information, and the extracting the target bit sequence from the object to be detected includes: taking the end position of the embedding range of the target bit sequence as the extracting target The initial detection point of the bit sequence extracts the target bit sequence from the object to be detected.

The object to be detected includes an image to be detected, and extracting a target watermark sequence from the object to be detected includes: obtaining a grayscale histogram of the image to be detected; The target watermark sequence is extracted from the object to be detected.

The object to be detected includes a sequence of video frames, and extracting a target watermark sequence from the object to be detected includes: obtaining a first reference sub-bit sequence for representing a bit value "0" in the target bit sequence , And obtaining a second reference sub-bit sequence used to represent the bit value "1" in the target bit sequence, where the first reference sub-bit sequence is different from the second reference sub-bit sequence; from the video frame A plurality of target sub-bit sequences are extracted from the sequence of video images; the target sub-bit sequence is compared with the first reference sub-bit sequence and the second reference sub-bit sequence, respectively, to determine the video frame The value of the bit embedded in the video image of the sequence is "0" or "1".

The extracting the target watermark sequence from the object to be detected based on the gray level histogram includes: calculating the gray level average value of the image to be detected; calculating the gray level histogram according to the gray level average value The target gray-scale interval; according to the predetermined number of bit values of the watermark sequence embedded in the image to be detected, the target gray-scale interval is divided to obtain a number corresponding to the number of bit values of the embedded watermark sequence Each of the gray-scale sub-intervals includes at least two adjacent gray-scale levels, and the gray-scale levels are used to indicate the number of pixels with the same gray-scale value; obtaining predetermined bit values to extract data According to the number of pixels in at least two adjacent gray levels of the gray sub-interval and the bit value extraction data, the bit values embedded in the image to be detected are extracted respectively to obtain the target Bit sequence.

In the above-mentioned embodiments, a method for extracting watermark information and a device for extracting watermark information are provided. In addition, the sixth embodiment of the present application also provides an electronic device. Since the electronic device embodiment is basically similar to the method embodiment, Therefore, the description is relatively simple. For details of related technical features, please refer to the corresponding description of the method embodiment provided above. The following description of the electronic device embodiment is only illustrative. An example of the electronic device is as follows:

Please refer to FIG. 6 to understand this embodiment. FIG. 6 is a schematic diagram of the electronic device provided in this embodiment.

As shown in FIG. 6, the electronic device includes: a processor 601; a memory 602;

The memory 602 is configured to store a watermark information extraction program, and when the program is read and executed by the processor, the following operations are performed:

Obtain the object to be detected;

Obtain a reference bit sequence containing a amble sequence and original watermark information, where the amble sequence satisfies the following conditions: after the amble sequence is shifted, the obtained shifted sequence is between the sequence and the amble sequence The degree of discrimination is greater than the preset degree of discrimination;

Extracting a target bit sequence from the object to be detected;

The target bit sequence is matched with the reference bit sequence, and it is determined whether the target bit sequence is a reference bit sequence embedded in the object to be detected.

In a typical configuration, the computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include non-permanent memory in a computer readable medium, random access memory (RAM) and/or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM). Memory is an example of computer readable media.

1. Computer-readable media include permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disc (DVD) or other optical storage, Magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media can be used to store information that can be accessed by computing devices. According to the definition in this article, computer-readable media does not include non-transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.

2. Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems or computer program products. Therefore, this application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

Although this application is disclosed as above in preferred embodiments, it is not intended to limit this application. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of this application. Therefore, this application The scope of protection shall be subject to the scope defined by the claims of this application.

Claims

A method for embedding watermark information, which is characterized in that it comprises:

Obtain the carrier object;

Obtain the watermark information to be embedded;

Obtain the target preamble sequence;

Embedding the watermark information to be embedded and the target preamble sequence into the carrier object;

Wherein, the target preamble sequence satisfies the following condition: after performing shift processing on the target preamble sequence, the degree of discrimination between the obtained shifted sequence and the target preamble sequence is greater than the preset degree of discrimination.
The method according to claim 1, wherein the target amble sequence is a binary bit sequence, and the embedding the watermark information to be embedded and the target amble sequence into the carrier object comprises:

Adding the target preamble sequence to the front end or the back end of the watermark information to be embedded to obtain a target bit sequence;

The target bit sequence is embedded in the carrier object.
The method according to claim 2, wherein the carrier object comprises a carrier image, and the embedding of the target bit sequence into the carrier object comprises:

Obtaining a grayscale histogram of the carrier image;

Adjusting the shape of the grayscale histogram according to the target bit sequence to obtain an adjusted grayscale histogram;

According to the adjusted gray histogram, the gray values of the pixels in the carrier image are adjusted to obtain the target image embedded with the target bit sequence.
The method according to claim 3, wherein the adjusting the shape of the grayscale histogram according to the target bit sequence to obtain the adjusted grayscale histogram comprises:

Obtaining the target gray-scale interval of the gray-scale histogram;

According to the number of bits in the target bit sequence, the target gray-scale interval is divided to obtain gray-scale sub-intervals corresponding to the number of bits, wherein each gray-scale sub-interval includes at least two adjacent gray-scale sub-intervals. Gray level, the gray level is used to indicate the number of pixels with the same gray value;

Obtain the number relationship information of the pixel points in the gray sub-interval corresponding to the bit value;

According to the target bit sequence and the number of pixels in the gray sub-interval corresponding to the bit value, the number of pixels contained in at least two adjacent gray levels in the gray sub-interval is performed Adjust to obtain the adjusted gray histogram.
The method according to claim 4, wherein the obtaining the target gray scale interval of the gray scale histogram comprises:

Calculating the average gray value of the carrier image;

Based on the representation range of the grayscale histogram and the number of bits of the target bit sequence, the target grayscale interval of the grayscale histogram is calculated according to the grayscale mean value.
The method according to claim 4, wherein the obtaining the quantity relationship information of pixels in the gray-scale sub-interval corresponding to the bit value comprises:

When the bit value is 1, obtaining the ratio of the number of pixels contained in two adjacent gray levels in the gray sub-interval corresponding to the bit value and the first comparison relationship with the predetermined embedding intensity; and

When the bit value is 0, obtain the ratio of the number of pixels contained in two adjacent gray levels in the gray sub-interval corresponding to the bit value and the second comparison relationship between the predetermined embedding intensity;

According to the target bit sequence and the number relationship information of the pixel points in the gray-scale sub-interval corresponding to the bit value, the determination of the pixel points contained in at least two adjacent gray-scale levels in the gray-scale sub-interval is Adjust the quantity to obtain the adjusted gray histogram, including:

The value of the bit to be embedded in the target bit sequence is 1, if the ratio of the number of pixels contained in two adjacent gray levels in the gray subinterval corresponding to the bit value to be embedded is less than the predetermined embedding intensity The relationship between the two is consistent with the first comparison relationship, then the pixels are not adjusted; if the ratio of the number of pixels contained in two adjacent gray levels in the gray sub-interval corresponding to the bit value to be embedded, and If the relationship between the predetermined embedding intensities does not conform to the first comparison relationship, the first number of pixels are selected from the pixels contained in the gray level with a larger gray value and moved to the gray with a smaller gray level. Degree

The value of the bit to be embedded in the target bit sequence is 0, if the ratio of the number of pixels contained in two adjacent gray levels in the gray sub-interval corresponding to the bit value to be embedded is less than the predetermined embedding intensity The relationship between the two is consistent with the second comparison relationship, then the pixels are not adjusted; if the ratio of the number of pixels contained in two adjacent gray levels in the gray sub-interval corresponding to the bit value to be embedded, and If the relationship between the predetermined embedding intensities does not conform to the second comparison relationship, a second number of pixels are selected from the pixels contained in the gray level with a smaller gray value and moved to the gray with a larger gray value. Degree

Wherein, the first number and the second number are calculated based on the predetermined embedding intensity and the number of pixels included in two adjacent gray levels in the gray sub-interval.
The method according to claim 6, wherein said selecting a first number of pixel points from the pixel points contained in a gray level with a larger gray value and moving to a gray level with a smaller gray level , Including: using a random method to select a first number of pixel points from the pixel points contained in a gray level with a larger gray value to move to a gray level with a smaller gray level;

The selecting the second number of pixel points from the pixel points included in the gray level with the smaller gray value to move to the gray level with the larger gray value includes: adopting a random method from the smaller gray value The pixel points of the second number selected from the pixel points included in the gray level are moved to a gray level with a larger gray value.
The method according to claim 2, wherein before obtaining the grayscale histogram of the carrier image, the method further comprises: performing Gaussian filtering processing on the carrier image to obtain the low-frequency signal part of the carrier image;

The obtaining the grayscale histogram of the carrier image includes:

A gray histogram of the low-frequency signal part of the carrier image is obtained.
8. The method according to claim 8, wherein the obtaining a grayscale histogram of the low-frequency signal part of the carrier image comprises:

Performing block processing on the low-frequency signal part of the carrier image to obtain a block image;

Calculating the average gray value of the segmented image;

According to the average gray value of the block image, the gray scale histogram of the block image is statistically obtained.
The method according to claim 9, further comprising: calculating the mean square error of the block image;

The adjusting the shape of the grayscale histogram according to the target bit sequence to obtain the adjusted grayscale histogram includes:

Selecting a predetermined number of grayscale histograms of the block images according to the order of the mean square error of the block images from large to small;

According to the target bit sequence, the shape of the grayscale histogram of the predetermined number of block images is adjusted to obtain the adjusted grayscale histogram of the block image.
The method according to claim 2, wherein the carrier object comprises a video frame sequence, and the embedding of the target bit sequence into the carrier object comprises:

Obtain a first sub-bit sequence representing the bit value "0" in the target bit sequence, and obtain a second sub-bit sequence representing the bit value "1" in the target bit sequence, the first A sub-bit sequence is different from the second sub-bit sequence;

Obtaining a target video image in the video frame sequence, where the target video image refers to a video image to be embedded in the target bit sequence;

Obtaining the target bit value of the target video image to be embedded;

Embedding the first sub-bit sequence or the second sub-bit sequence used to represent the target bit value into the target video image.
The method according to claim 11, wherein the first sub-bit sequence satisfies the following condition: after the first sub-bit sequence is shifted, the obtained shifted bit sequence is the same as the The degree of discrimination between the first sub-bit sequence is greater than the preset degree of discrimination; and, the second sub-bit sequence satisfies the following condition: after shifting the second sub-bit sequence, the obtained shifted The degree of discrimination between the bit sequence and the second sub-bit sequence is greater than the preset degree of discrimination.
The method according to claim 11, wherein the first sub-bit sequence includes a first sub-amble sequence and first sub-watermark information, and the first sub-amble sequence satisfies the following conditions: After the first sub-amble sequence is shifted, the degree of discrimination between the obtained shifted bit sequence and the first sub-amble sequence is greater than the preset degree of discrimination; and, the second sub-bit sequence includes The second sub-preamble sequence and the second sub-watermark information, the second sub-preamble sequence satisfies the following condition: after the second sub-preamble sequence is shifted, the obtained shifted bit sequence and The degree of discrimination between the second sub-amble sequences is greater than the preset degree of discrimination; wherein, the first sub-watermark information is different from the second sub-watermark information.
The method according to claim 13, wherein the first sub-preamble sequence is the same as the second sub-preamble sequence.
The method according to claim 13, wherein the first sub-preamble sequence and the target preamble sequence are the same bit sequence, and the second sub-preamble sequence and the target preamble sequence are The same bit sequence.
The method according to claim 11, wherein said obtaining the target video image in the sequence of video frames comprises:

According to the manner of embedding a bit value in one frame of video image of the video frame sequence, according to the number of bit values contained in the target bit sequence, obtain the target to be embedded in the target bit sequence from the video frame sequence Video image.
The method according to any one of claims 11-15, wherein the first sub-bit sequence or the second sub-bit sequence used to represent the target bit value is embedded in the The target video image includes:

Obtaining a grayscale histogram of the target video image;

According to the first sub-bit sequence or the second sub-bit sequence used to represent the target bit value, the shape of the grayscale histogram of the target video image is adjusted to obtain the adjusted grayscale Histogram

According to the adjusted gray histogram, the gray value of the pixel in the target video image is adjusted to obtain the target video image embedded with the first sub-bit sequence or the second sub-bit sequence.
The method according to claim 2, wherein the carrier object comprises a video frame sequence, and the embedding of the target bit sequence into the carrier object comprises:

The target bit sequence is repeatedly embedded into the video frame sequence according to the predetermined number of embedding times.
A method for extracting watermark information, which is characterized in that it comprises:

Obtain the object to be detected;

Obtain a reference bit sequence containing a amble sequence and original watermark information, where the amble sequence satisfies the following conditions: after the amble sequence is shifted, the obtained shifted sequence is between the sequence and the amble sequence The degree of discrimination is greater than the preset degree of discrimination;

Extracting a target bit sequence from the object to be detected;

The target bit sequence is matched with the reference bit sequence, and it is determined whether the target bit sequence is a reference bit sequence embedded in the object to be detected.
The method according to claim 19, wherein the amble sequence is set at the front end of the original watermark information, and the extracting the target bit sequence from the object to be detected comprises:

The start position of the embedding range of the target bit sequence is used as the initial detection point for extracting the target bit sequence, and the target bit sequence is extracted from the object to be detected.
The method according to claim 19, wherein the amble sequence is set at the back end of the original watermark information, and the extracting the target bit sequence from the object to be detected comprises:

The end position of the embedding range of the target bit sequence is used as the initial detection point for extracting the target bit sequence, and the target bit sequence is extracted from the object to be detected.
The method according to any one of claims 19-21, wherein the object to be detected includes an image to be detected, and extracting a target watermark sequence from the object to be detected comprises:

Obtaining a grayscale histogram of the image to be detected;

Based on the grayscale histogram, a target watermark sequence is extracted from the object to be detected.
The method according to any one of claims 19-21, wherein the object to be detected comprises a sequence of video frames, and said extracting a target watermark sequence from the object to be detected comprises:

Obtain a first reference sub-bit sequence used to represent the bit value "0" in the target bit sequence, and obtain a second reference sub-bit sequence used to represent the bit value "1" in the target bit sequence, so The first reference sub-bit sequence is different from the second reference sub-bit sequence;

Extracting multiple target sub-bit sequences from the video images of the video frame sequence;

The target sub-bit sequence is respectively compared with the first reference sub-bit sequence and the second reference sub-bit sequence to determine that the value of the bit embedded in the video image of the video frame sequence is "0" Or "1".
The method according to claim 22, wherein said extracting a target watermark sequence from said object to be detected based on said gray histogram comprises:

Calculating the average gray value of the image to be detected;

Calculating the target grayscale interval of the grayscale histogram according to the grayscale average;

According to the predetermined number of bit values of the watermark sequence embedded in the image to be detected, the target gray-scale interval is divided to obtain the gray-level sub-intervals corresponding to the number of bit values of the embedded watermark sequence, Each of the gray sub-intervals includes at least two adjacent gray levels, and the gray levels are used to indicate the number of pixels with the same gray value;

Obtain a predetermined bit value to extract data;

According to the number of pixels in at least two adjacent gray levels in the gray sub-interval and the bit value extraction data, the bit values embedded in the image to be detected are respectively extracted to obtain the target bit sequence.
A watermark information embedding device is characterized in that it comprises:

The carrier object obtaining unit is used to obtain the carrier object;

A unit for obtaining watermark information to be embedded, configured to obtain watermark information to be embedded;

The target amble sequence obtaining unit is configured to obtain a target amble sequence, and the target amble sequence satisfies the following condition: after the target amble sequence is shifted, the obtained shifted sequence is the same as the target amble sequence. The degree of discrimination between code sequences is greater than the preset degree of discrimination;

The information embedding unit is used to embed the watermark information to be embedded and the target preamble sequence into the carrier object.
An electronic device, characterized in that it comprises:

processor;

The memory is used to store a watermark information embedding program, and when the program is read and executed by the processor, the following operations are performed:

Obtain the carrier object;

Obtain the watermark information to be embedded;

Obtain a target preamble sequence, and the target preamble sequence meets the following condition: after the target preamble sequence is shifted, the degree of discrimination between the obtained shifted sequence and the target preamble sequence is greater than that of the target preamble sequence. Set the degree of distinction;

Embedding the watermark information to be embedded and the target preamble sequence into the carrier object.
A watermark information extraction device, which is characterized in that it comprises:

The object to be detected obtaining unit is used to obtain the object to be detected;

The reference bit sequence obtaining unit is configured to obtain a preset reference bit sequence containing a amble sequence and original watermark information, where the amble sequence satisfies the following condition: after shifting the amble sequence, the obtained shift The degree of discrimination between the post-bit sequence and the preamble sequence is greater than the preset degree of discrimination;

A target bit sequence extraction unit, configured to extract a target bit sequence from the object to be detected;

The information matching unit is configured to match the target bit sequence with the reference bit sequence, and determine whether the target bit sequence is a reference bit sequence embedded in the object to be detected.
An electronic device, characterized in that it comprises:

processor;

The memory is used to store a watermark information extraction program, and when the program is read and executed by the processor, the following operations are performed:

Obtain the object to be detected;

Obtain a preset reference bit sequence containing a amble sequence and original watermark information, where the amble sequence satisfies the following conditions: after shifting the amble sequence, the obtained shifted sequence and the amble are The degree of discrimination between sequences is greater than the preset degree of discrimination;

Extracting a target bit sequence from the object to be detected;

The target bit sequence is matched with the reference bit sequence, and it is determined whether the target bit sequence is a reference bit sequence embedded in the object to be detected.
A method for embedding watermark information, which is characterized in that it comprises:

Obtain the carrier object;

Obtain at least two watermark information to be embedded;

Adding different target preamble sequences to the at least two watermark information to be embedded to obtain at least two target embedding sequences;

Embedding the at least two target embedding sequences into the carrier object;

Wherein, the target preamble sequence satisfies the following condition: after performing shift processing on the target preamble sequence, the degree of discrimination between the obtained shifted sequence and the target preamble sequence is greater than the preset degree of discrimination.
The method according to claim 29, wherein the adding different target preamble sequences to the at least two watermark information to be embedded comprises:

A target preamble sequence corresponding to the watermark information to be embedded is added to each watermark information to be embedded in the at least two watermark information to be embedded.