CN102270335A - Method for embedding additional information into image - Google Patents

Abstract

The present invention provides a method and device for embedding additional information in an image, including: selecting at least one region from the image according to the time domain characteristics of the image itself, the region is composed of the region boundary and the region interior; Set the corresponding least significant bit LSB level for the time-domain features respectively; for each area, according to the corresponding LSB level and adopt the LSB concealment method to embed the pre-set boundary mark into the area boundary, according to the corresponding LSB level and adopt the LSB concealment method to embed Additional information is embedded inside the region. The application of the solution of the present invention not only realizes the hiding of additional information, but also embeds the additional information more flexibly under the condition that the image is not perceived.

Description

A method and device for embedding additional information in an image

technical field

The invention relates to digital image processing technology, in particular to a method and device for embedding additional information in an image.

Background technique

In digital image processing technology, certain algorithm can be used to hide some additional information in image data, such as hiding copyright protection information, image annotation or another digital image, etc. This digital image technology is usually divided into two types of algorithms based on the frequency domain and based on the time domain. Among them, the algorithms based on the time domain are mostly implemented using the least significant bit (LSB) principle, that is, the low-order bits of image pixels are used to hide additional information, and the high-order bits are still used to represent the pixel value of the point. Since the contribution of the low-order bits to the pixel value is relatively small, the human eye is not sensitive to changes in the low-order bits of the pixel. If additional information is embedded in the low-order bits, it can reduce the damage to the image without affecting the human eye. Visual experience, so as to achieve the effect of hiding additional information.

However, the existing technology is usually aimed at the entire image, and uses a unified LSB number (or LSB level) to hide additional information, which is not well correlated with time-domain features such as image content, brightness, texture, and position, and cannot be flexible. Hide additional information.

Contents of the invention

In view of the above-mentioned defects in the prior art, the object of the present invention is to provide a method and device for embedding additional information in an image, which can embed additional information more flexibly while ensuring that the image is not perceived.

For above-mentioned first invention object, the technical scheme that the present invention proposes is:

A method of embedding additional information in an image, the method comprising:

Selecting at least one region from the image according to the temporal characteristics of the image itself, where the region consists of a region boundary and a region interior;

Set the corresponding least significant bit LSB level according to the time domain characteristics of each area;

For each region, according to the corresponding LSB level and using the LSB hiding method, the pre-set boundary identifier is embedded into the region boundary, and according to the corresponding LSB level and using the LSB hiding method, the additional information is embedded into the region.

In the above solution, the time-domain feature includes image content, brightness, color, or position in the image.

In the above scheme, the method for selecting an area includes:

A selected area is formed by the user-specified area boundary, the specified area boundary, and the enclosed area interior; or,

A region boundary is recognized from the image by using a region boundary recognition algorithm, and the recognized region boundary and the surrounding region constitute a selected region.

In the above solution, the method for setting the corresponding LSB level according to the time domain characteristics of each area includes:

First set the default LSB level for each area, and then adjust the LSB level one by one according to the time domain characteristics of each area to obtain the corresponding LSB level for each area; or,

For each area, the feature value of the area is calculated according to the time-domain feature correlation algorithm, and then the LSB level of the area is determined according to the corresponding relationship between the feature value and the LSB level set in advance, so as to obtain the LSB level corresponding to each area.

In the above solution, for each region, the method of embedding the preset boundary identifier into the region boundary includes: for each region, determining the pixel points of the region boundary according to the coordinates of the region boundary, and determining the region according to the LSB level corresponding to the region The bits used for hiding in the border pixels, and embedding the border identifier into the bits used for hiding in the border pixels of the region;

For each area, the method for embedding additional information into the area includes: for each area, determining the pixel points inside the area according to the coordinates of the area boundary, and determining the pixel points in the area for hiding according to the corresponding LSB level of the area. bits, and embed additional information into the hidden bits in the pixels inside the region.

For above-mentioned second invention object, the technical scheme that the present invention proposes is:

A device for embedding additional information in an image, the device comprising:

The area selection module selects at least one area from the image according to the time domain characteristics of the image itself, and notifies the selected area to the least significant bit LSB level setting module and the information embedding module, the area is composed of the area boundary and the interior of the area;

The least significant bit LSB level setting module sets the corresponding least significant bit LSB level respectively according to the time domain characteristics of each area, and notifies the information embedding module;

The information embedding module, for each region, according to the corresponding LSB level of the region and using the LSB concealment method, embeds the pre-set boundary mark into the region boundary of the region, according to the corresponding LSB level of the region and adopts the LSB concealment method to embed the additional information Embedded inside the region's region.

In the above solution, the time-domain feature includes brightness, color or position in the image.

In the above scheme, the information embedding module includes:

The first storage module is used to save the border identifier;

The second storage module is used to store additional information;

The third storage module is used to save the LSB level corresponding to each area;

The first execution module, for each region, determines the pixel points of the region border according to the coordinates of the region border, determines the bits used for hiding in the region border pixel points according to the corresponding LSB level of the region saved by the third storage module, and sends the second A boundary identifier saved by a storage module is embedded in the bit for hiding in the boundary pixels of the region;

The second execution module, for each region, determines the pixels inside the region according to the coordinates of the region boundary, determines the bits used for hiding in the pixels inside the region according to the corresponding LSB level of the region saved by the third storage module, and sends the The additional information saved by the second storage module is embedded in the bits used for hiding in the pixels inside the area.

The present invention provides a method and device for embedding additional information in an image. At least one region can be selected according to the temporal characteristics of the image itself, and then corresponding LSB levels can be set respectively according to the temporal characteristics of each region. Each region embeds the boundary identifier into the region boundary according to the corresponding LSB level, and embeds additional information into the region according to the corresponding LSB level, which not only realizes the hiding of additional information, but also ensures that the image is not perceived, more flexible embedded additional information.

Description of drawings

Fig. 1 is a flow chart of the method for embedding additional information in an image according to the present invention.

Fig. 2 is a schematic diagram of an image to hide additional information in an embodiment of the present invention.

Fig. 3 is a flowchart of an embodiment of the present invention.

FIG. 4 is a schematic diagram of region selection from FIG. 2 .

Fig. 5 is a schematic diagram of bits of information that can be embedded in an embodiment of the present invention.

Fig. 6 is a schematic diagram of an image before embedding additional information using the solution of an embodiment of the present invention.

Fig. 7 is a schematic diagram of an image after embedding additional information using the solution of the embodiment of the present invention.

FIG. 8 is a schematic structural diagram of a device for embedding additional information in an image according to the present invention.

FIG. 9 is a schematic diagram of the internal structure of the information embedding module 803 in FIG. 8 .

Detailed ways

The core idea of the present invention is to set different least significant bit (LSB) levels for different regions according to the time-domain characteristics of the image itself, and embed boundary identification and additional information into the region boundary and region interior respectively. In this way, not only the purpose of embedding additional information can be achieved, but also, since different LSB levels are set for different regions, the additional information can be embedded more flexibly while ensuring that the image is not perceived.

Fig. 1 is a flow chart of the method for embedding additional information in an image according to the present invention. As shown in Figure 1, the method includes:

Step 101: Select at least one region from the image according to the time-domain characteristics of the image itself, and the region is composed of a boundary of the region and an interior of the region.

Here, the temporal features of an image refer to image content, brightness, color, or position in the image, etc. The reason why the present invention selects the region according to the above-mentioned features of the image is because: some time-domain features have intuitive meanings to people (such as image content), and some time-domain features are not sensitive to human eyes (such as highlighted areas, rich image textures, etc.) area, the border area of the image). Based on this principle, the present invention can use these selected regions to hide information later.

In addition, there are many ways to select an area. For example, the user can manually specify the area boundary, and the specified area boundary and the surrounding area constitute a selected area. As for how to implement it, it should be easy for those skilled in the art to implement. I won't repeat them here. For another example: a region boundary recognition algorithm may be used to recognize a region boundary from an image, and the recognized region boundary and the surrounding region constitute a selected region. The area boundary identification algorithm described here may include existing technologies such as differential boundary detection and gradient boundary detection, which will not be described in detail here.

Step 102: Set the corresponding least significant bit LSB level respectively according to the time domain characteristics of each area.

In order to make full use of the time domain characteristics of different regions, the present invention can set different LSB levels for different regions. The LSB level mentioned here is the lowest number of bits allowed to hide information in a pixel point. During specific implementation, it can be determined according to the time domain characteristics of a certain area that the area is suitable for hiding information. For example: select 3 areas that need to hide information. Among them, the brightness value of the first area is high, which is convenient to hide more information, so a higher LSB level can be set for it, so that the lowest 3 bits of pixels in this area are used to hide information, and recorded as LSB=3. The luminance value of the second area is relatively low, which is inconvenient to hide more information, so a lower LSB level can be set for it, so that the lowest 1 bit of the pixel in this area is used to hide information, and recorded as LSB=1. Although the brightness value of the third area is relatively low, it belongs to the area image boundary, and it is also convenient to hide more information. A slightly higher LSB level can be set for it, so that the lowest 2 bits of pixels in this area are used to hide information and record LSB=2.

Of course, there are many ways to specifically determine the LSB level, for example: first set the default LSB level for each area, and then adjust the LSB level one by one according to the time domain characteristics of each area to obtain the corresponding LSB level for each area. Another example: for each area, calculate the feature value of the area according to the time-domain feature correlation algorithm, and then determine the LSB level of the area according to the corresponding relationship between the feature value and the LSB level set in advance, so as to obtain the corresponding LSB level of each area . The time-domain feature correlation algorithm described here may be an image texture information acquisition algorithm, an image brightness information acquisition algorithm, etc., and the specific algorithms may be implemented using existing technologies, and will not be repeated here.

Step 103: For each region, according to the corresponding LSB level and using the LSB hiding method, embed the pre-set boundary identifier into the region boundary, and according to the corresponding LSB level and using the LSB hiding method, embed additional information into the region.

This step not only embeds the additional information into the selected area, but also embeds the boundary identifier into the area boundary of each area. The reason why the region identification is embedded is that the present invention does not embed additional information in the whole image as in the prior art, but embeds additional information in one or some regions of the image. Those skilled in the art know that if the additional information is embedded in the entire image, the area where the additional information is read is known, and the additional information can be easily read. However, if the additional information is embedded in one or some regions of the image, the additional information cannot be successfully read out without knowing the location of the region. Based on this, in order to facilitate the subsequent reading of additional information, the present invention also carries the information identifying the boundary of the region in the image, that is, the boundary identifier is embedded in the boundary of the region.

Specifically,

According to the corresponding LSB level and using the LSB concealment method, the method of embedding the boundary identifier into the region boundary includes: for each region, determining the pixel points of the region boundary according to the coordinates of the region boundary, and determining the pixel points of the region boundary according to the corresponding LSB level of the region Bits for hiding, and embedding the boundary identifier into the bit for hiding in the boundary pixels of the region.

The method of embedding additional information into the region according to the corresponding LSB level and using the LSB hiding method includes: for each region, determine the pixel points inside the region according to the coordinates of the region boundary, and determine the pixels in the region according to the corresponding LSB level of the region bits for hiding, and embed additional information into the bits for hiding in the pixels inside the region.

In order to better illustrate the solution of the present invention, a preferred embodiment is used for a detailed description below.

FIG. 2 is a schematic diagram of an image in which additional information is assumed to be hidden in this embodiment. The image is a bitmap (BMP), where each small square represents a pixel.

The method flow of this embodiment is shown in Figure 3, and may include the following flow:

Step 301: Read a BMP image.

This embodiment is described by taking a BMP image as an example, and in actual application, an image in other formats such as Joint Photographic Experts Group (JPG or JPEG), JPG2000, etc. may also be used.

Step 302: Select three regions (region 1, region 2, and region 3) from the image according to the temporal characteristics of the image itself, and each region consists of a region boundary and a region interior.

As mentioned above, in this embodiment, regions can be manually selected, or regions can be automatically selected using algorithms such as differential boundary detection and gradient boundary detection. As for how to implement it, it is up to the user who applies the solution of the present invention to determine. In this embodiment, it is assumed that three regions are selected from FIG. 2 , and the selection situation is shown in FIG. 4 .

Step 303: Record the coordinates of the boundaries of each area.

In practical applications, the area boundary coordinates of the selected area can be recorded first, so as to facilitate subsequent operations on the area. For example: the boundary coordinates of a certain square area are recorded as (0,0)...(0,10)...(10,10)...(10,0), the area in this embodiment can also be according to this method to record the boundary coordinates. In this way, these coordinates can be used to determine the pixel point where the area is located, so as to accurately embed information.

Step 304: Set the LSB level corresponding to area 1 to 3, the LSB level corresponding to area 2 to 1, and the LSB level corresponding to area 3 to 2 according to the respective time domain characteristics of the three selected areas.

In this embodiment, assuming that area 1 is a boundary area, the human eye is not sensitive, and the LSB can be set higher; area 2 is in the middle of the image, and the human eye is more sensitive than the boundary, and the LSB can be set lower; 3 belongs to the highlight area, and the human eye is not too sensitive, so it can be set a little higher. No matter the setting basis is brightness, color, image content, or position in the image, these features can be collectively referred to as temporal features of the image in this implementation, and the correspondence between these temporal features and LSB levels can be based on experience or The actual situation of the image is determined.

That is to say, the method of setting the LSB level in this embodiment can be manually adjusted by the user as mentioned above: first set the LSB level of area 1 to area 3 to the default level, that is, LSB=1; for area 1 , this area belongs to the boundary area, which can be readjusted to LSB=3; for area 2, this area is in the middle of the image, and its default level can be kept unchanged; for area 3, this area belongs to the highlighted area, Then it can be properly adjusted to LSB=2.

In practical applications, the LSB level can also be adjusted automatically. For example: first determine from the boundary coordinates of each area that area 1 belongs to the boundary area, and areas 2 and 3 do not belong to the boundary area, so set the LSB level corresponding to area 1 to 3; Calculate the average luminance values of areas 2 and 3 as X and Y respectively. After searching the relationship table corresponding to the luminance value and LSB level set in advance, it is determined that the LSB level corresponding to X is 1, and the LSB level corresponding to Y is 2. Therefore, the The LSB level corresponding to area 2 is set to 1, and the LSB level corresponding to area 3 is set to 2. What is calculated here is the location of the area and the average brightness of the area. In practical applications, other temporal features such as image texture information of the area can also be calculated, and the corresponding relationship between these calculated feature values and the LSB level can be found to determine the corresponding LSB. level.

In short, regardless of the manual or automatic method, no matter which time-domain feature is used, an appropriate LSB level can be set for each region in the end, and will not be listed here.

Step 305: Record the LSB levels set for each area.

Step 306: Set the boundary flag K.

As mentioned above, the boundary of the region used to hide the additional information also needs to embed the boundary identifier K, so as to determine the location of the region when the additional information is read. Assuming that the LSB level of a certain area has been set to 2 in step 304, the K value mentioned in this step is the value of the lowest two bits of the border pixel that needs to be embedded in the area. Of course, the value of the boundary identifier K can be set by the user who applies the solution of the present invention. For example: still taking the LSB level as 2 as an example, in order to reduce the impact on the boundary of the region, the boundary identifier K can be set to "00", that is, "00" is embedded in the lowest two bits of the pixel where the boundary of the region is located . If you don't consider the influence of the area boundary, you can also embed other values, such as "10" and so on. In addition, in practical applications, in order to reduce the probability of conflict between the border mark and the pixel value of the image pixel itself, the border mark can also be embedded in the four pixels of the "Tian"-shaped position in the two rows and two columns of the border, that is, in The boundary logo is embedded in the outermost two weeks of pixels of the region boundary, as shown in Figure 4. Of course, in practical applications, whether to embed the boundary mark in the outermost pixel points of the area or to embed the boundary mark in the outermost two pixel points can be determined by the user applying the solution of the present invention.

Step 307: Set additional information 1 for area 1, set additional information 2 for area 2, and set additional information 3 for area 3.

Step 308: For area 1, determine the pixel points of the area boundary according to the coordinates of the area boundary, determine the hidden bits in the area boundary pixel points according to the LSB level corresponding to area 1, and embed the boundary identifier K into the area boundary pixel Bits used for hiding in the point; for area 1, determine the pixel points inside the area according to the coordinates of the area boundary, determine the bits used for hiding in the pixels inside the area according to the corresponding LSB level of the area, and embed additional information 1 Bits used for hiding in pixels inside the region.

Step 309: For area 2, determine the pixel points of the area boundary according to the coordinates of the area boundary, determine the hidden bits in the area boundary pixel points according to the LSB level corresponding to area 2, and embed the boundary identifier K into the area boundary pixel Bits used for hiding in the point; for area 2, determine the pixel points inside the area according to the coordinates of the area boundary, determine the bits used for hiding in the pixels inside the area according to the corresponding LSB level of the area, and embed additional information 2 Bits used for hiding in pixels inside the region.

Step 310: For area 3, determine the pixel points of the area boundary according to the coordinates of the area boundary, determine the hidden bits in the area boundary pixel points according to the LSB level corresponding to area 3, and embed the boundary identifier K into the area boundary pixel Bits used for hiding in the point; for area 3, determine the pixel points inside the area according to the coordinates of the area boundary, determine the bits used for hiding in the pixels inside the area according to the corresponding LSB level of the area, and embed additional information 3 Bits used for hiding in pixels inside the region.

In this embodiment, steps 308 to 310 are key steps for embedding a border marker at the border of the selected region and embedding additional information inside the region. Whether it is embedding boundary markers or embedding additional information, the embedding methods all use the LSB hidden method. Taking the LSB level as 2 as an example, as shown in Figure 5, assuming that a pixel includes three colors of red (Red), green (Green), and blue (Blue), each color occupies 8 bits, and the lowest two bits are enough. As the bits of embedded information, how to specifically embed them belongs to the prior art and will not be repeated here.

Next, the technical solution of the present invention is used to verify a specific embodiment. Assume that the read BMP original image, the selected area, and the additional information to be embedded are shown in Figure 6. That is to say, 2 regions are selected in the image to embed additional information. Region 1 is located in the upper half of the image, and the set LSB level is assumed to be 2. Region 2 is located in the lower left corner of the image, and the set LSB level is assumed to be 1.

In addition, the additional information that needs to be embedded in area 1 is:

Model: SAMSUNG LA52A650

Display Size: 52”

Resolution: 1920×1080

Aspect Ratio: 16:9

Dimensions: 1276*326*866

Launch Date: April, 2008

The additional information that needs to be embedded in area 2 is:

Copyright1995-2009

SAMSUNG

All rights reserved

After embedding the additional information, the image carries the additional information and region boundary identification. Afterwards, it can be set to display the additional information embedded in this area in real time when the mouse etc. enters a certain area. As shown in Figure 7: when the mouse is positioned on the upper half of the image, the additional information embedded in area 1 is immediately displayed.

For the above method, the present invention also proposes a device for embedding additional information in an image. FIG. 8 shows a schematic diagram of the basic structure of the device. As shown in FIG. 8 , the device may include: a region selection module 801 , an LSB level setting module 802 , and an information embedding module 803 . in,

The area selection module 801 is used to select at least one area from the image according to the time domain characteristics of the image itself, and notify the selected area to the least significant bit level setting module 802 and the information embedding module 803, and the area is composed of the area boundary and the area internal composition.

Here, the temporal features of an image refer to image content, brightness, color, or position in the image, etc. As described in the previous method, there are many ways to select an area, and the user can manually specify the area boundary, and the area selection module 801 finally determines the area according to the specified area boundary. Certainly, a region boundary recognition algorithm may also be used to recognize the region boundary from the image, and the recognized region boundary and the surrounding region constitute a selected region. The area boundary recognition algorithm described here may include existing technologies such as differential boundary detection and gradient boundary detection, which will not be described in detail here.

The least significant bit level setting module 802 sets the corresponding least significant bit LSB level according to the time domain characteristics of each area, and notifies the information embedding module 803 .

Here, there are many ways to specifically determine the LSB level, for example: firstly set a default LSB level for each area, and then adjust the LSB level one by one according to the time domain characteristics of each area to obtain the LSB level corresponding to each area. Another example: for each area, calculate the feature value of the area according to the time-domain feature correlation algorithm, and then determine the LSB level of the area according to the corresponding relationship between the feature value and the LSB level set in advance, so as to obtain the corresponding LSB level of each area . The time-domain feature correlation algorithm described here may be an image texture information acquisition algorithm, an image brightness information acquisition algorithm, etc., and specific algorithms may be implemented using existing technologies.

The information embedding module 803, for each area, according to the LSB level corresponding to the area and using the LSB hiding method, embeds the pre-set boundary identifier into the area boundary of the area, according to the LSB level corresponding to the area and using the LSB hiding method, the additional Information is embedded inside the region's region.

In practical applications, a schematic diagram of the internal structure of the information embedding module 803 may be shown in Figure 9, including:

The first storage module 901 is configured to save the boundary identifier.

The second storage module 902 is configured to store additional information.

The third storage module 903 is configured to store the LSB level corresponding to each area.

The first execution module 904, for each region, determines the pixel points of the region border according to the coordinates of the region border, and determines the hidden bits in the region border pixels according to the corresponding LSB level of the region saved by the third storage module 903, and Embedding the boundary identifier saved by the first storage module 901 into the hidden bits in the region boundary pixel.

The second execution module 905, for each region, determines the pixels inside the region according to the coordinates of the region boundary, and determines the hidden bits in the pixels inside the region according to the corresponding LSB level of the region saved by the third storage module 903, and Embedding the additional information stored in the second storage module 902 into the hidden bits in the pixels inside the region.

That is to say, for an image, the region selection module 801 selects at least one region from the image, and notifies the selected region to the least significant bit level setting module 802 and the information embedding module 803 . The least significant bit level setting module 802 respectively sets the corresponding least significant bit LSB level according to the time domain characteristics of each area, and saves it in the third storage module 903 . In addition, the apparatus also stores the set boundary identification information in the first storage module 901 , and stores the set additional information in the second storage module 902 . In this way, when embedding the boundary identifier, for each region, the first execution module 904 determines the pixel points of the region boundary according to the coordinates of the region boundary, and determines the pixel points of the region boundary according to the corresponding LSB level of the region saved by the third storage module 903. Bits used for hiding, embedding the boundary identifier saved by the first storage module 901 into the bits used for hiding in the boundary pixels of the region. Similarly, for each region, the second execution module 905 determines the pixel points inside the region according to the coordinates of the region boundary, and determines the pixel points used for hiding in the region internal pixels according to the LSB level corresponding to the region saved by the third storage module 903 bits, and embed the additional information stored in the second storage module 902 into the hidden bits in the pixels inside the region.

By applying the solution of the present invention, since different LSB levels can be set according to the time-domain characteristics of each region in the image, the boundary identifier and additional information are hidden in the region boundaries and the region interior of each region according to different LSB levels, which not only realizes The hiding of additional information also embeds additional information more flexibly while ensuring that the image is not perceived.

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

Claims (8)

1. A method of embedding additional information in an image, the method comprising:

selecting at least one region from the image according to the time domain characteristics of the image, wherein the region consists of a region boundary and a region interior;

respectively setting the LSB level of the corresponding least significant bit according to the time domain characteristics of each region;

and embedding a preset boundary identifier into the boundary of each region according to the corresponding LSB level by adopting an LSB hiding method, and embedding additional information into the region according to the corresponding LSB level by adopting the LSB hiding method.

2. The method of claim 1, wherein the temporal feature comprises image content, brightness, color, or location in an image.

3. The method according to claim 1 or 2, wherein the method of selecting a region comprises:

the user designates the area boundary, and the designated area boundary and the enclosed area form a selected area; or,

and identifying the region boundary from the image by using a region boundary identification algorithm, wherein the identified region boundary and the enclosed region form a selected region.

4. The method according to claim 1 or 2, wherein the method for respectively setting the corresponding LSB level according to the time domain characteristics of each region comprises:

firstly, setting a default LSB level for each region, and then adjusting the LSB levels one by one according to the time domain characteristics of each region to obtain the LSB levels corresponding to each region; or,

and aiming at each region, calculating the characteristic value of the region according to a time domain characteristic correlation algorithm, and determining the LSB level of the region according to the preset corresponding relation between the characteristic value and the LSB level so as to obtain the LSB level corresponding to each region.

5. The method according to claim 1 or 2, wherein the method for embedding the boundary identifier set in advance into the boundary of the region comprises the following steps of: for each region, determining pixel points of the region boundary according to the coordinates of the region boundary, determining a bit for hiding in the region boundary pixel points according to the LSB level corresponding to the region, and embedding a boundary identifier into the bit for hiding in the region boundary pixel points;

for each region, the method of embedding additional information inside the region includes: and aiming at each region, determining pixel points inside the region according to the coordinates of the region boundary, determining the bit for hiding in the pixel points inside the region according to the LSB level corresponding to the region, and embedding additional information into the bit for hiding in the pixel points inside the region.

6. An apparatus for embedding additional information in an image, the apparatus comprising:

the image processing device comprises a region selection module, a least significant bit LSB level setting module and an information embedding module, wherein the region selection module selects at least one region from an image according to the time domain characteristics of the image and informs the selected region to the LSB level setting module and the information embedding module, and the region consists of a region boundary and a region interior;

the LSB level setting module of the least significant bit respectively sets the corresponding LSB level of the least significant bit according to the time domain characteristics of each region and informs the information embedding module;

and the information embedding module is used for embedding a preset boundary identifier into the region boundary of each region by adopting an LSB hiding method according to the LSB level corresponding to the region and embedding additional information into the region of the region by adopting the LSB hiding method according to the LSB level corresponding to the region.

7. The apparatus of claim 6, wherein the temporal feature comprises a brightness, a color, or a location in an image.

8. The apparatus of claim 6 or 7, wherein the information embedding module comprises:

the first storage module is used for storing the boundary identifier;

the second storage module is used for storing additional information;

the third storage module is used for storing the LSB level corresponding to each area;

the first execution module is used for determining pixel points of the region boundary according to the coordinates of the region boundary, determining bit positions for hiding in the region boundary pixel points according to LSB levels corresponding to the region and stored in the third storage module, and embedding the boundary identification stored in the first storage module into the bit positions for hiding in the region boundary pixel points;

and the second execution module is used for determining pixel points in the region according to the coordinates of the region boundary, determining bit positions for hiding in the pixel points in the region according to the LSB level corresponding to the region stored by the third storage module, and embedding the additional information stored by the second storage module into the bit positions for hiding in the pixel points in the region.

Images