Spatial domain thumbnail preserving encryption method for image privacy protection

Abstract

The invention discloses a spatial domain thumbnail preserving encryption method for image privacy protection. According to the method, the image privacy is kept, the availability of the image is taken into consideration, and the encryption and visualization balance of the image is realized through the thumbnail keeping encryption technology. According to the method, a mode of combining pixel prediction and label marking is adopted, an image owner preprocesses an original image, encrypts the image after embedding auxiliary information, and adjusts the encrypted image by using a residual space vacated by preprocessing, so that a ciphertext image thumbnail keeps an approximate visual effect compared with an original image thumbnail. According to the method, the calculation complexity is effectively reduced while the embedding capacity is improved, and excellent universality is shown in an actual image application scene. According to the scheme, the image encryption and information embedding process is completely reversible, and the reserved space of the auxiliary information embedding and image adjusting part is sufficient.

Description

Technical Field

[0001] The invention relates to a spatial domain thumbnail keeping encryption method for image privacy protection, belonging to the technical field of multimedia information security. Background Art

[0002] In recent years, with the rapid development of science and technology, individual users have become accustomed to capturing and recording moments in life through portable smartphones, forming a large amount of rich personal image data. In order to facilitate sharing, viewing and backing up these precious moments, many users choose to store their images in the cloud. However, many of these images contain private information that users do not want to disclose. Given the convenience of cloud storage, scholars have explored solutions to encrypt images to protect user privacy. However, traditional encryption methods usually sacrifice the availability of images, which is inconvenient for users to browse online and choose to download. To this end, thumbnail-preserving encryption technology has emerged, aiming to maintain the availability of images while protecting image privacy. The core idea of ​​thumbnail-preserving encryption technology is to maintain the visualization of images under the premise of encryption, so that image owners can easily select the required images through online browsing, and illegal users cannot obtain detailed information that leaks users' personal privacy from encrypted images. In the specific implementation, thumbnail-preserving encryption focuses on keeping the sum of encrypted pixel values ​​in image blocks close to or equal to the sum of original pixel values ​​to achieve visual availability. The uniqueness of this technology lies in that it provides a more flexible and convenient way of image management while protecting user privacy.

[0003] However, the thumbnail-preserving encryption scheme currently using reversible information hiding technology has a significant defect, that is, the reserved information storage space is insufficient, which results in the image being unable to normally display the integrity of the original image during the decryption and recovery process. This problem seriously affects the reliability and practicality of the encryption scheme. In addition, some existing thumbnail-preserving encryption schemes use a multi-round permutation encryption method, which leads to a high computational complexity. Such high complexity makes the encryption and decryption process time-consuming and resource-intensive, limiting the widespread application of such schemes in practical applications. Therefore, improving the efficiency and performance of thumbnail-preserving encryption schemes is an important challenge facing current technology. The present invention aims to address the above-mentioned problems and provide a more efficient and reliable thumbnail-preserving encryption method. Summary of the invention

[0004] Purpose of the invention: The present invention aims to overcome the limitations of existing methods and provide a thumbnail preservation encryption method based on the combination of pixel prediction and label marking. This method can not only significantly increase the embedding capacity and achieve perfect restoration of encrypted images, but also reduce the computational complexity and show excellent flexibility, so that it can be widely promoted in various practical images.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution provided by the present invention is as follows.

[0006] A spatial domain thumbnail preservation encryption method for image privacy protection comprises the following steps:

[0007] Step 1): First, preprocess the original image. Use the existing spatial domain information hiding algorithm to perform necessary reservation operations. During the image preprocessing process, the corresponding positions are reasonably divided to complete the operations of auxiliary information embedding and encrypted image adjustment required in the image restoration process. In particular, the pixel positions of these two types of operations must be guaranteed not to conflict;

[0008] Step 2): Embed the auxiliary information needed for image restoration into the labeled image, while ensuring that the capacity of the auxiliary information does not exceed the corresponding reserved space. Then, encrypt the processed image;

[0009] Step 3): Using the pre-divided adjustment space, the encrypted image embedded with the auxiliary information is adjusted to achieve a similar visual effect between the ciphertext image thumbnail and the original image thumbnail.

[0010] As a further optimization scheme of the spatial domain thumbnail encryption method for image privacy protection of the present invention, the specific steps of preprocessing the original image in step 1) are as follows:

[0011] Separate the R, G, and B channel components of the original image, and perform the following operations on each channel:

[0012] The first row and first column of the single-channel image are used as the reference pixel set. The predicted value of each pixel is calculated using the median edge predictor:

[0013]

[0014] In the formula, x pred (i, j) represents the predicted pixel value, a, b, c represent the pixel values ​​of the left, upper and upper left pixels of the current predicted pixel value respectively. Select the label parameter Ω (1≤Ω≤7), and calculate the error range S=2 allowed to be marked according to the binary code Ω . Calculate the prediction error e between the original image and the predicted image pred (i, j), and generate the prediction error position map α according to the prediction error and the label error range S determined by the label parameter:

[0015] e pred (i, j) = x(i, j) - x pred (i, j)

[0016] In the formula, x(i, j) represents the pixel value of a single channel of the original image with a horizontal coordinate of x and a vertical coordinate of y. The pixels are grouped according to the generated prediction error position map, and the pixels whose prediction error values ​​are located at The intervals are divided into embeddable pixels and those that do not meet the conditions are non-embeddable pixels, and a binary matrix is ​​generated to represent the embeddability of the pixels at the corresponding positions.

[0017] The pixel labeling scheme is used to mark pixels, and all pixels in the original image are divided into reference pixel set, embeddable pixel set, and non-embeddable pixel set. The reference pixel set remains unchanged during the pixel labeling process, and the Ω-LSB bits of the pixels in the embeddable pixel set are marked using a binary matrix.

[0018] As a further optimization scheme of the spatial domain thumbnail encryption method for image privacy protection of the present invention, the detailed steps of step 2) are as follows:

[0019] The auxiliary information is concatenated and embedded into the labeled image as a binary data stream. A checkerboard-style embedding strategy is used to embed the auxiliary information necessary for restoring the image using only the coordinates with the same parity in the reference pixel set and the embeddable pixel set. The pixels in the reference pixel set directly use bit replacement to embed the auxiliary information, and the embeddable pixels only replace the (8-Ω)-MSB bit. Generate the encryption matrix T:

[0020] T=Gen(A,F,K e )

[0021] In the formula, T is consistent with the matrix size of the original image, A is a random number, F is the information corresponding to each image, and K e is the encryption key. The encryption matrix and the marked image are subjected to XOR stream encryption to obtain the encrypted image.

[0022] As a further optimization scheme of the spatial domain thumbnail encryption method for image privacy protection of the present invention, the detailed steps of step 3) are as follows:

[0023] The encrypted image is divided into blocks, and the (8-Ω)-MSB position of the pixel is adjusted using the coordinates of different parities in the embeddable pixel set. The specific steps are as follows: first, the (8-Ω)-MSB position of the adjusted pixel is set to 0. If the value of the block pixel sum is greater than the value of the original block pixel sum at this time, the adjustment fails, otherwise proceed to the next step; the highest position of the adjustable pixel is set to 1. If the value of the block pixel sum is less than the value of the original block pixel sum at this time, the next MSB position is set to 1, otherwise the highest position is set to 0, and recursively from the next MSB position until the sum of the adjusted block pixel values ​​is close to the sum of the original block pixel values. The R, G, and B channels of the adjusted image are integrated to obtain the encrypted adjusted image.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] The embedding capacity of the encrypted image is improved, thus achieving perfect recovery of the encrypted image. It not only retains the usability of the image, but also effectively reduces the computational complexity of image decryption. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic diagram of the image auxiliary information embedding and pixel encryption adjustment process of the present invention;

[0027] Figure 2 It is a schematic diagram of the image restoration process of the present invention;

[0028] Figure 3 This is a schematic diagram of the labeling scheme used in the present invention; DETAILED DESCRIPTION

[0029] The following will be combined with the accompanying drawings of the embodiments of the present invention to clearly and completely describe the technical solution of the present invention. In order to make the purpose, features and advantages of the present invention more obvious, the following will be further described in detail by combining the accompanying drawings and specific implementation methods.

[0030] like Figure 1 As shown, the present invention discloses a spatial domain thumbnail encryption method for image privacy protection, comprising the following steps:

[0031] Step 1): First, preprocess the original image. Use the existing spatial domain information hiding algorithm to perform necessary reservation operations. During the image preprocessing process, the corresponding positions are reasonably divided to complete the operations of auxiliary information embedding and encrypted image adjustment required in the image restoration process. In particular, the pixel positions of these two types of operations must be guaranteed not to conflict;

[0032] Step 2): Embed the auxiliary information needed for image restoration into the labeled image, while ensuring that the capacity of the auxiliary information does not exceed the corresponding reserved space. Then, encrypt the processed image;

[0033] Step 3): Using the pre-divided adjustment space, the encrypted image embedded with the auxiliary information is adjusted to achieve a similar visual effect between the ciphertext image thumbnail and the original image thumbnail.

[0034] In step 1), the specific steps of preprocessing the original image are as follows:

[0035] Separate the R, G, and B channel components of the original image, and perform the following operations on each channel:

[0036] The first row and first column of the single-channel image are used as the reference pixel set. The predicted value of each pixel is calculated using the median edge predictor:

[0037]

[0038] In the formula, x pred (i, j) represents the predicted pixel value, a, b, c represent the pixel values ​​of the left, upper and upper left pixels of the current predicted pixel value respectively. Select the label parameter Ω (1≤Ω≤7), and calculate the error range S=2 allowed to be marked according to the binary code Ω . Calculate the prediction error e between the original image and the predicted image pred (i, j), and generate the prediction error position map α according to the prediction error and the label error range S determined by the label parameter:

[0039] e pred (i, j) = x(i, j) - x pred (i, j)

[0040] In the formula, x(i, j) represents the pixel value of a single channel of the original image with a horizontal coordinate of x and a vertical coordinate of y. The pixels are grouped according to the generated prediction error position map, and the pixels whose prediction error values ​​are located at The intervals are divided into embeddable pixels and those that do not meet the conditions are non-embeddable pixels, and a binary matrix is ​​generated to represent the embeddability of the pixels at the corresponding positions.

[0041] The pixel is labeled using a labeling scheme. The labeling scheme used in the present invention is as follows: Figure 3 As shown. All pixels in the original image are divided into a reference pixel set, an embeddable pixel set, and a non-embeddable pixel set. The reference pixel set remains unchanged during the pixel marking process, and the Ω-LSB bits of the pixels in the embeddable pixel set are marked using a binary matrix.

[0042] The detailed steps of step 2) are as follows:

[0043] The auxiliary information is concatenated and embedded into the labeled image as a binary data stream. The auxiliary information includes: parameter Ω, prediction error binary position map and reference pixel set. The checkerboard style embedding strategy is used to embed the auxiliary information necessary for restoring the image only with the coordinates of the reference pixel set and the embeddable pixel set that have the same parity. The pixels in the reference pixel set directly use bit replacement to embed the auxiliary information, and the embeddable pixels only replace the (8-Ω)-MSB bit. Generate the encryption matrix T:

[0044] T=Gen(A,F,K e )

[0045] In the formula, T is consistent with the matrix size of the original image, A is a random number, F is the information corresponding to each image, and Ke is the encryption key. The encryption matrix and the marked image are subjected to XOR stream encryption to obtain the encrypted image.

[0046] The detailed steps of step 3) are as follows:

[0047] The encrypted image is divided into blocks, and the (8-Ω)-MSB bits of the pixels with different parities in the embedded pixel set are adjusted. The specific steps are as follows:

[0048] First, the (8-Ω)-MSB position of the adjusted pixel is set to 0. If the value of the pixel sum of the block is greater than the value of the original pixel sum of the block, the adjustment fails. Otherwise, proceed to the next step. For the highest position of the adjustable pixel, set it to 1. If the value of the pixel sum of the block is less than the value of the original pixel sum of the block, continue to set the next MSB position to 1. Otherwise, set the highest position to 0 and recursively start from the next MSB position until the sum of the pixel values ​​of the adjusted block is close to the sum of the pixel values ​​of the original block. Finally, integrate the R, G, and B channels of the adjusted image to obtain the encrypted adjusted image.

[0049] The image decryption process diagram of the present invention is as follows Figure 2 As shown, the specific steps include:

[0050] First, the decryption matrix T is generated. The encrypted matrix T is the same as the encryption matrix. The encrypted image is decrypted by binary XOR. Then, auxiliary information is extracted from the reference pixel row: parameter Ω, prediction error binary position map, and reference pixel set. Secondly, the original pixel value of each pixel point is restored according to the extracted binary position map to reconstruct the original image.

[0051] By combining the research of visual psychology, the present invention introduces a design that is more in line with human visual habits in image processing, which improves the user experience of encrypted images. At the same time, the application of reversible information hiding technology based on the spatial domain provides more flexibility for encrypted images, so that encrypted images can maintain a high degree of usability while protecting privacy. This method not only focuses on privacy security, but also fully considers the user's intuitive perception of images, achieving innovation and progress in the field of image encryption.

[0052] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0053] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A spatial domain thumbnail-preserving encryption method for image privacy protection. It is characterized in that The following steps are involved: Step 1), first, preprocess the original image. By adopting the existing spatial domain information hiding algorithm, necessary reservation operations are performed. In the image preprocessing process, the corresponding positions are reasonably divided to complete the operations of auxiliary information embedding and encrypted image adjustment required in the image restoration process. In particular, it is required that the pixel bits of these two types of operations must ensure no conflict; Step 2), embed the auxiliary information needed for image restoration into the labeled image, while ensuring that the capacity of the auxiliary information does not exceed the corresponding reserved space. Then, encrypt the processed image; Step 3), using the pre-divided adjustment space, the encrypted image embedded with the auxiliary information is adjusted to achieve a similar visual effect between the ciphertext image thumbnail and the original image thumbnail.

2. The spatial domain thumbnail encryption method for image privacy protection according to claim 1, It is characterized in that In step 1), the original image is pre-processed as follows: Step 1.1), separate the R, G, and B channel components of the original image, and perform the following operations on each channel; Step 1.2), take the first row and first column of the single-channel image as the reference pixel set. Use the median edge predictor to calculate the predicted value of each pixel: In the formula, x pred (i, j) represents the predicted pixel value, a, b, c represent the pixel values ​​of the left, top and top-left pixels of the current predicted pixel value respectively; Step 1.3), select the label parameter Ω (1≤Ω≤7), and calculate the error range S=2 allowed to be labeled according to the binary code Ω ; Step 1.4), calculate the prediction error e between the original image and the predicted image pred (i, j), and generate the prediction error position map α according to the prediction error and the label error range S determined by the label parameter: e pred (i,j)=x(i,j)-x pred (i,j) Where x(i, j) represents the pixel value of a single channel of the original image with abscissa x and ordinate y; Step 1.5), group the pixels according to the generated prediction error position map, and group the pixels where the prediction error value is located The intervals are divided into embeddable pixels, and those that do not meet the conditions are non-embeddable pixels, and a binary matrix is ​​generated to represent the embeddability of the pixels at the corresponding positions; Step 1.6), use the labeling scheme to label pixels and divide all pixels in the original image into a reference pixel set, an embeddable pixel set, and a non-embeddable pixel set. The reference pixel set remains unchanged during the pixel labeling process, and the Ω-LSB bits of the pixels in the embeddable pixel set are labeled using a binary matrix.

3. The spatial domain thumbnail encryption method for image privacy protection according to claim 1, It is characterized in that The detailed steps of step 2) are as follows: Step 2.1), concatenate the auxiliary information and embed it into the labeled image as a binary data stream; Step 2.2), a checkerboard-style embedding strategy is selected, and only the coordinates with the same parity between the reference pixel set and the embeddable pixel set are used to embed the auxiliary information necessary for restoring the image. The pixels in the reference pixel set are directly embedded with the auxiliary information using bit replacement, and the embeddable pixels only replace the (8-Ω)-MSB bit; Step 2.3), generate the encryption matrix T: T=Gen(A,F,K e ) In the formula, T is consistent with the matrix size of the original image, A is a random number, F is the information corresponding to each image, and K e is the encryption key; Step 2.4) Perform XOR stream encryption on the encryption matrix and the marked image to obtain an encrypted image.

4. The spatial domain thumbnail preservation encryption method for image privacy protection according to claim 1, It is characterized in that The detailed steps of step 3) are as follows: Step 3.1), the encrypted image is divided into blocks, and the (8-Ω)-MSB bits of the pixels with different parities in the coordinate adjustment of the embedded pixel set are adjusted. The specific steps are as follows: Step 3.1.1), first set the (8-Ω)-MSB position of the adjusted pixel to 0. If the value of the pixel sum of each block is greater than the value of the pixel sum of the original block, the adjustment fails. Otherwise, proceed to step 3.1.

2. Step 3.1.2), set the highest position of the adjustable pixel to 1. If the value of the pixel sum of the block is less than the value of the pixel sum of the original block, continue to set the next MSB position to 1. Otherwise, set the highest position to 0 and recursively start from the next MSB position until the sum of the pixel values ​​of the adjusted block is close to the sum of the pixel values ​​of the original block; Step 3.2), integrate the R, G, and B channels of the adjusted image to obtain an encrypted adjusted image.

Images