A reversible information hiding and predictive recovery system and method for fully encrypted images

Abstract

The present invention belongs to the field of information security technology, and discloses a reversible information hiding and predictive recovery system and method for a fully encrypted image, including the following steps: S1: the image owner uses an encryption key to fully encrypt the original image, and uploads the encrypted image to a cloud server; S2: the information hider uses the information embedding key to embed additional information into the encrypted image, and obtains an encrypted image carrying additional information; S3: the image receiver extracts the embedded additional information or restores the original image according to different keys. The present invention can flexibly set the number of divided intervals according to the parameter value, and can increase the effective information embedding capacity to 1.5 bpp; it can realize lossless restoration of the original image when the embedding capacity of the smoothed image is small; information extraction and image restoration can be completely separated, and compared with other excellent algorithms, it has achieved both a higher embedding capacity and better restored image quality.

Description

Technical Field

[0001] The present invention belongs to the field of information security technology, and in particular relates to a reversible information hiding and predictive recovery system and method for a fully encrypted image. Background Art

[0002] With the rapid development of 5G communication and cloud storage technology, more and more users use cloud servers to save private images and use mobile networks to obtain the required images at any time. However, unauthorized attackers may intercept sensitive images during image transmission or even directly attack cloud servers to steal data, posing a network security threat. Therefore, protecting the data privacy of user images in cloud storage applications has received widespread attention. Among them, image encryption and reversible information hiding technology have become effective means to solve this problem.

[0003] Since the images stored by users in the cloud environment are highly private, traditional reversible information hiding techniques cannot be effectively applied to encrypted images. Therefore, in recent years, it has become crucial to explore the use of reversible information hiding techniques in encrypted images, namely, Reversible Data Hiding in Encrypted Image (RDHEI). Existing RDHEI schemes are mainly divided into two categories: one is to reserve redundant space before image encryption (RRBE), and the other is to vacate redundant space after image encryption (VRAE).

[0004] For the two existing RDHEI schemes, the RRBE scheme requires a lot of preprocessing before image encryption, which is not conducive to user-side use. The encrypted image security of the VRAE scheme based on block encryption or chaotic encryption is insufficient, which may cause information leakage. In the fully encrypted VRAE scheme, lossless extraction of embedded information can be achieved, but there are still problems such as low embedding capacity and poor image restoration quality. Summary of the invention

[0005] The purpose of the present invention is to provide a reversible information hiding and predictive recovery system and method for a fully encrypted image to solve the above-mentioned technical problems.

[0006] In order to solve the above technical problems, the specific technical solutions of a reversible information hiding and predictive recovery system and method of a fully encrypted image of the present invention are as follows:

[0007] A reversible information hiding and predictive recovery method for a fully encrypted image comprises the following steps:

[0008] S1: The image owner completely encrypts the original image using the image encryption key and uploads the encrypted image to the cloud server;

[0009] S2: The information hider uses the information hiding key to embed the additional information into the encrypted image to generate an encrypted image containing the additional information;

[0010] S3: The image receiver extracts the embedded additional information or restores the original image based on different keys.

[0011] Furthermore, the S1 comprises the following steps:

[0012] S11: The image owner generates a random number encryption matrix using his / her short key K e ;

[0013] S12: The image owner converts the original image and the encryption matrix into 8-bit binary;

[0014] S13: The image owner encrypts the original image;

[0015] S14: The image owner converts the binary encrypted image into decimal, implements image encryption and uploads the encrypted image to the cloud server. When decrypting the image, the pixels are also converted into binary, and then the decrypted binary pixel values ​​are obtained.

[0016] Furthermore, the S2 comprises the following steps:

[0017] S21: Information hider uses information to hide the key K h Encrypt additional information;

[0018] The additional binary information to be embedded is D , the length is The information hider generates a one-dimensional information hiding key with the same length as the additional binary information to be embedded by using the short key he holds. K h , the additional information is encrypted using the same bitwise XOR method as image encryption. The encrypted additional information can be securely embedded into the encrypted image to avoid information leakage;

[0019] S22: The information hider divides the encrypted image into hidden areas;

[0020] The hidden area division adopts an image pixel chessboard prediction method; using the image pixel chessboard prediction method, the encrypted image pixels are divided into two pixel groups of black and white grids. When information is hidden, the value of the white grid pixel group is kept unchanged, and information embedding is achieved by changing the black grid pixel value;

[0021] S23: The information hider divides the pixel interval to achieve information hiding;

[0022] S24: Process the black grid pixels in sequence until all black grid pixels are processed or all additional information is embedded, thereby obtaining an encrypted image containing secret information.

[0023] Further, the S23 comprises the following steps:

[0024] S231: Setting parameters , its value indicates the number of intervals formed after the overall pixel range is evenly divided;

[0025] S232: Calculate the starting pixel value after the interval is divided;

[0026] S233: Calculate the difference between the hidden pixel value and the starting pixel value of all intervals d ;

[0027] S234: In the difference d Find the minimum absolute value in ;

[0028] S235: Calculate the adjusted interval pixels according to the adjustment difference ;

[0029] S236: Obtain pixel value after embedding data e′ , for data embedding.

[0030] Furthermore, the S3 comprises the following steps:

[0031] S31: The receiver holds the information hiding key and can extract the secret information losslessly;

[0032] S32: The recipient holds the image encryption key and restores the original image with high quality.

[0033] Further, the S31 includes the following steps:

[0034] S311: The receiver divides the image into two parts according to the checkerboard pattern to obtain the positions of the pixels that can be hidden;

[0035] The chessboard division is a method for predicting the chessboard of image pixels;

[0036] S312: Calculate pixel division intervals according to parameter θ;

[0037] S313: Process the hidden pixels in the order of raster scanning, extract the decimal data according to the interval number of the pixel, and convert it into binary format, using the information to hide the key K hDecrypt the original extra information.

[0038] Further, the S32 includes the following steps:

[0039] S321: The receiver sends the image E′ Divide the pixels into two parts: hidden pixels and non-hidden pixels according to the chessboard;

[0040] S322: For white pixels that cannot be hidden, use the key K e Direct decryption;

[0041] S323: Obtain the prediction value corresponding to the black grid pixel according to the chessboard pixel prediction method p ;

[0042] S324: Divide the pixels that can be hidden in black grids into intervals, and obtain adjusted interval pixels;

[0043] S325: According to the key corresponding to the pixel K e , decrypt the interval pixels and obtain the decrypted interval pixels;

[0044] S326: Calculate the difference between all decrypted interval pixels and predicted pixels, and use the interval pixels with the smallest absolute value of the difference as the restored pixel value to achieve image restoration.

[0045] The present invention also discloses a reversible information hiding and predictive recovery system for a fully encrypted image, comprising:

[0046] The user encryption side is used to encrypt the original image based on the target encryption algorithm to obtain an encrypted image;

[0047] The information hiding side is used to embed additional information into the encrypted image based on the target information embedding algorithm to obtain the encrypted image carrying the additional information;

[0048] User recovery side: used to perform information extraction or image recovery operations based on the held key and the held encrypted image to obtain additional information or the original image.

[0049] Furthermore,

[0050] The user encryption side includes:

[0051] A first random key generating unit, used to generate a random key, the generated random key will be used for image encryption and decryption;

[0052] An encryption processing unit, used for encrypting the original image according to the random key to obtain an encrypted image;

[0053] The information hiding side includes:

[0054] A second random key generation unit, used to generate an information hiding key for encryption and decryption of hidden information;

[0055] an information embedding processing unit, used to embed information into the encrypted image according to the information hiding key to obtain an encrypted image embedded with additional information;

[0056] The user recovery side includes:

[0057] an information extraction unit, configured to determine the embedded encrypted additional information according to the embedding parameters and the encrypted image pixels embedded with the additional information and the information hiding key;

[0058] The decryption processing unit is used to perform decryption processing according to the encrypted image and the random key to obtain the original image.

[0059] Furthermore,

[0060] The encryption processing unit comprises:

[0061] An encryption parameter generation subunit, configured to generate a key matrix and an image parameter matrix based on the target encryption algorithm and the random key and the original image;

[0062] An encryption processing subunit, used for encrypting the key matrix and the image parameter matrix;

[0063] The information embedding processing unit comprises:

[0064] An information embedding parameter generating subunit, configured to generate an interval start pixel, an adjustment difference, and an adjustment interval pixel according to the encrypted image pixels and embedding parameters based on the target information embedding algorithm;

[0065] An information embedding processing subunit, which acquires an encrypted image pixel embedded with additional information according to an interval start pixel of the encrypted image pixel, an adjustment difference, an adjustment interval pixel and the encrypted additional information;

[0066] The decryption processing unit comprises:

[0067] A pixel prediction subunit, used for predicting pixel values ​​of encrypted pixels embedded with additional information according to the encrypted image to obtain predicted pixel values;

[0068] The decryption processing subunit is used to reversely process the encrypted pixels according to the encryption processing unit on the user encryption side and the information embedding unit on the information hiding side, and obtain the restored pixel values ​​according to the predicted pixel values.

[0069] The reversible information hiding and predictive recovery system and method of a fully encrypted image of the present invention have the following advantages: the reversible information hiding and predictive recovery system and method of a fully encrypted image of the present invention can flexibly set the number of divided intervals according to parameter values, and can increase the effective information embedding capacity to 1.5bpp; by retaining half of the pixel values ​​of the image and using pixel checkerboard prediction to obtain effective prediction values ​​of the hideable pixels, and using the prediction values ​​to assist in the effective recovery of the hideable pixels, the smooth image can be restored to the original image without loss when the embedding capacity is small; information extraction and image restoration can be completely separated, and compared with other excellent algorithms, it can achieve both higher embedding capacity and better restored image quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is a framework diagram of the algorithm of the present invention;

[0071] Figure 2 A schematic diagram of chessboard division and prediction in an embodiment;

[0072] Figure 3 This is an example diagram of image encryption, information hiding, information extraction and image restoration in the embodiment. DETAILED DESCRIPTION

[0073] In order to better understand the purpose, structure and function of the present invention, the following is a further detailed description of a reversible information hiding and predictive recovery system and method for a fully encrypted image of the present invention in conjunction with the accompanying drawings.

[0074] like Figure 1 As shown, a reversible information hiding and predictive recovery method of a fully encrypted image of the present invention comprises the following steps:

[0075] S1: The image owner completely encrypts the original image using the image encryption key and uploads the encrypted image to the cloud server;

[0076] S11: The image owner generates a value with a range of 0-255 using his / her short key, which is the same size as the image. Random number encryption matrix K e ;

[0077] S12: The image owner converts the original image and the encryption matrix into 8-bit binary using formula (1) and formula (2);

[0078] The formula (1) and formula (2) are:

[0079] , (1)

[0080] , (2)

[0081] The symbols in formula (1) and formula (2) are i , j is the coordinate, symbol Indicates rounding down. ε is a binary bit, o is the original image pixel, K e is the random number encryption matrix, The coordinates are i , j The original image pixels, The coordinates expressed in 8-bit binary are i , j The original image pixels, The coordinates are i , j The encryption matrix, The coordinates expressed in 8-bit binary are i , j The encryption matrix;

[0082] S13: The image owner encrypts the original image using formula (3);

[0083] The formula (3) is:

[0084] , (3)

[0085] The symbols in formula (3) are Represents the exclusive OR operation;

[0086] S14: The image owner uses formula (4) to convert the binary encrypted image into decimal, encrypts the image and uploads the encrypted image to the cloud server. When decrypting the image, the pixels are also converted into binary, and then the decrypted binary pixel value is obtained using formula (5);

[0087] The formula (4) and formula (5) are:

[0088] , (4)

[0089] , (5)

[0090] S2: The information hider uses the information hiding key to embed the additional information into the encrypted image to generate an encrypted image containing the additional information;

[0091] S21: Information hider uses information to hide the key K h Encrypt additional information;

[0092] Assume that the additional binary information to be embedded isD , the length is The information hider generates a one-dimensional information hiding key with the same length as the additional binary information to be embedded by using the short key he holds. K h , the additional information is encrypted using the same bitwise XOR method as image encryption, as shown in formula (3). The encrypted additional information can be securely embedded into the encrypted image to avoid information leakage;

[0093] S22: The information hider divides the encrypted image into hidden areas;

[0094] The hidden area division adopts the image pixel chessboard prediction method;

[0095] like Figure 2 As shown in the figure, the image pixel chessboard prediction method divides the original image into two parts, black grid pixel group and white grid pixel group, according to the international chessboard pattern. When predicting the black grid pixel, the average value of the surrounding white grid pixels adjacent to it is calculated by formula (6) to achieve prediction;

[0096] The formula (6) is:

[0097] , (6)

[0098] The symbols in formula (6) are p is the predicted pixel value;

[0099] Using the image pixel chessboard prediction method, the encrypted image pixels are divided into two pixel groups: black and white. When hiding information, the value of the white pixel group is kept unchanged, and information is embedded by changing the black pixel value;

[0100] S23: The information hider divides the pixel interval to achieve information hiding;

[0101] S231: Setting parameters θ , its value indicates the number of intervals formed by evenly dividing the overall pixel range of 0-255. θ The specific value of can be determined in advance so that the receiver can extract information and restore the image. The number of pixels in each interval after division is 256 / θ ;

[0102] S232: Calculate the starting pixel value after the interval division using formula (7) ;

[0103] The formula (7) is:

[0104] , (7)

[0105] In the formula (7),n is the pixel interval number, The pixel interval is n The starting pixel value of

[0106] like Figure 3 As shown, θ The value is 4, and the starting pixel values ​​are 0, 64, 128, and 192 respectively;

[0107] S233: Calculate the difference between the hidden pixel value and the starting pixel value of all intervals using formula (8) d ;

[0108] The formula (8) is:

[0109] , (8)

[0110] like Figure 3 As shown, the original pixel 187 is generated into an encrypted pixel 212 after being XOR-encrypted by the encryption key 111, and the difference between the encrypted pixel 212 and the starting pixels of these intervals is calculated to obtain 212, 148, 84, and 20;

[0111] S234: Using formula (9) in the difference d Find the minimum absolute value in ;

[0112] The formula (9) is:

[0113] , (9)

[0114] The min in the formula (9) represents a minimum value operation;

[0115] S235: Calculate the adjusted interval pixel value using formula (10) according to the adjustment difference

[0116] The formula (10) is:

[0117] , (10)

[0118] like Figure 3 As shown, the adjusted interval pixels are 20, 84, 148, and 212 respectively;

[0119] S236: The number of binary bits that can be embedded in a pixel is log2 θ , extract log2 from the encrypted extra information in order θ bits of data, denoted as α , use formula (11) to get the pixel value after embedding data e′ ;

[0120] The formula (11) is:

[0121] , (11)

[0122] The decimal in the formula (11) represents converting the bit to be embedded into the corresponding decimal number;

[0123] like Figure 3 As shown, assuming that the value of the additional information to be embedded is 01, 84 is selected as the pixel value after embedding;

[0124] S24: processing the black grid hidden pixels in sequence until all black grid pixels are processed or all the additional information has been embedded, thereby obtaining an encrypted image containing the secret information;

[0125] S3: The image receiver extracts the embedded additional information or restores the original image based on different keys;

[0126] S31: The receiver holds the information hiding key and can extract the secret information losslessly;

[0127] S311: The receiver sends the image E′ Divide the chessboard into two parts to obtain the location of the hidden pixels;

[0128] The chessboard division is a method for predicting the chessboard of image pixels;

[0129] S312: According to the parameters θ The value of and formula (7) calculate the pixel division interval;

[0130] S313: Process the hidden pixels in the order of raster scanning, extract the decimal data according to the interval number of the pixel, and convert it into binary format, using the information to hide the key K h Decrypt the original extra information;

[0131] like Figure 3 As shown, according to the hidden pixel 84 and the pixel division interval, it can be determined that the embedded information is 01;

[0132] S32: The receiver holds the image encryption key and restores the original image with high quality;

[0133] S321: The receiver sends the image E′ Divide the pixels into two parts: hidden pixels and non-hidden pixels according to the chessboard;

[0134] S322: For white pixels that cannot be hidden, use formula (5) and the key K e Direct decryption to achieve lossless recovery;

[0135] S323: According to the chessboard pixel prediction method, use formula (6) to obtain the prediction value corresponding to the black grid pixel p ;

[0136] S324: for pixels that can be hidden in black grids, use formula (7) to divide the interval, and use formulas (8)-(10) to obtain adjusted interval pixels;

[0137] like Figure 3 As shown, according to the hidden pixel 84 and the calculated divided interval pixels 0, 64, 128, 192, the adjustment interval pixels 20, 84, 148, 212 are obtained;

[0138] S325: According to the key corresponding to the pixel K e , use formula (5) to decrypt the interval pixels and obtain the decrypted interval pixels;

[0139] like Figure 3 As shown, the adjustment interval pixels are XORed with the encryption key 111 to obtain the decrypted interval pixels 123, 59, 251, 187;

[0140] S326: Calculate the difference between all decrypted interval pixels and predicted pixels using formula (12), and use the interval pixel with the smallest absolute value of the difference as the restored pixel value to achieve image restoration;

[0141] The formula (12) is:

[0142] , (12)

[0143] The symbols in formula (12) are r is the pixel value after restoration, symbol To decrypt the interval pixels, the symbol is the pixel interval number where the pixel with the smallest absolute difference value is located, To decrypt the pixel value of the interval, the symbol p is the predicted pixel value;

[0144] like Figure 3 As shown, the calculated absolute values ​​of the differences are 61, 125, 67, and 3, respectively. The minimum difference is 3, and the corresponding pixel is 187. 187 is used as the restored pixel.

[0145] A reversible information hiding and predictive recovery system for a fully encrypted image of the present invention comprises:

[0146] The user encryption side is used to encrypt the original image based on the target encryption algorithm to obtain an encrypted image;

[0147] The information hiding side is used to embed additional information into the encrypted image based on the target information embedding algorithm to obtain the encrypted image carrying the additional information;

[0148] User recovery side: used to perform information extraction or image recovery operations based on the held key and the held encrypted image to obtain additional information or the original image.

[0149] The user encryption side includes:

[0150] A first random key generating unit, used to generate a random key, the generated random key will be used for image encryption and decryption;

[0151] An encryption processing unit, used for encrypting the original image according to a random key to obtain an encrypted image;

[0152] The encryption processing unit comprises:

[0153] An encryption parameter generation subunit, used for generating a key matrix and an image parameter matrix according to a random key and an original image based on a target encryption algorithm;

[0154] The encryption processing subunit is used for encrypting the key matrix and the image parameter matrix.

[0155] The information hiding side includes:

[0156] A second random key generation unit, used to generate an information hiding key for encryption and decryption of hidden information;

[0157] An information embedding processing unit, used for embedding information into the encrypted image according to the information hiding key to obtain an encrypted image embedded with additional information;

[0158] The information embedding processing unit comprises:

[0159] An information embedding parameter generating subunit, for generating interval starting pixels, adjusting difference values, and adjusting interval pixels according to encrypted image pixels and embedding parameters based on a target information embedding algorithm;

[0160] The information embedding processing subunit obtains the encrypted image pixel embedded with the additional information according to the interval start pixel of the encrypted image pixel, the adjustment difference, the adjustment interval pixel and the encrypted additional information.

[0161] The user recovery side includes:

[0162] an information extraction unit, configured to determine the embedded encrypted additional information according to the embedding parameters and the encrypted image pixels embedded with the additional information and the information hiding key;

[0163] A decryption processing unit, used for performing decryption processing according to the encrypted image and the random key to obtain the original image;

[0164] The decryption processing unit comprises:

[0165] A pixel prediction subunit, used for predicting pixel values ​​of encrypted pixels embedded with additional information according to the encrypted image to obtain predicted pixel values;

[0166] The decryption processing subunit is used to reversely process the encrypted pixels according to the encryption processing unit on the user encryption side and the information embedding unit on the information hiding side, and obtain the restored pixel values ​​according to the predicted pixel values.

[0167] It is to be understood that the present invention is described by some embodiments, and it is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.

Claims

1. A reversible information hiding and predictive recovery method for a fully encrypted image, characterized in that: The steps include: S1: The image owner completely encrypts the original image using the image encryption key and uploads the encrypted image to the cloud server; S2: The information hider uses the information hiding key to embed the additional information into the encrypted image to generate an encrypted image containing the additional information. This step specifically includes: S21: Information hider uses information to hide the key K h Encrypt additional information; The additional binary information to be embedded is D , the length is l The information hider generates a one-dimensional information hiding key with the same length as the additional binary information to be embedded by using the short key he holds. K h , the additional information is encrypted using the same bitwise XOR method as image encryption. The encrypted additional information can be securely embedded into the encrypted image to avoid information leakage; S22: The information hider divides the encrypted image into hidden areas; The hidden area division adopts an image pixel chessboard prediction method; using the image pixel chessboard prediction method, the encrypted image pixels are divided into two pixel groups of black and white grids. When information is hidden, the value of the white grid pixel group is kept unchanged, and information embedding is achieved by changing the black grid pixel value; S23: The information hider divides the pixel interval to achieve information hiding. This step specifically includes: S231: Setting parameters θ , its value indicates the number of intervals formed after the overall pixel range is evenly divided; S232: Calculate the starting pixel value after the interval is divided; S233: Calculate the difference between the hidden pixel value and the starting pixel value of all intervals d ; S234: In the difference d Find the minimum absolute value in ; S235: Calculate the adjusted interval pixels according to the adjustment difference ; S236: Obtain pixel value after embedding data e′ , perform data embedding; S24: processing the black grid hidden pixels in sequence until all black grid pixels are processed or all the additional information has been embedded, thereby obtaining an encrypted image containing the secret information; S3: The image receiver extracts the embedded additional information or restores the original image based on different keys.

2. The reversible information hiding and predictive recovery method of a fully encrypted image according to claim 1, characterized in that: The S1 comprises the following steps: S11: The image owner generates a random number encryption matrix using his / her short key K e ; S12: The image owner converts the original image and the encryption matrix into 8-bit binary; S13: The image owner encrypts the original image; S14: The image owner converts the binary encrypted image into decimal, implements image encryption and uploads the encrypted image to the cloud server. When decrypting the image, the pixels are also converted into binary, and then the decrypted binary pixel values ​​are obtained.

3. The reversible information hiding and predictive recovery method of a fully encrypted image according to claim 1, characterized in that: The S3 comprises the following steps: S31: The receiver holds the information hiding key and can extract the secret information losslessly; S32: The recipient holds the image encryption key and restores the original image with high quality.

4. The reversible information hiding and predictive recovery method of a fully encrypted image according to claim 3 is characterized in that: The S31 comprises the following steps: S311: The receiver divides the image into two parts according to the checkerboard pattern to obtain the positions of the pixels that can be hidden; The chessboard division is a method for predicting the chessboard of image pixels; S312: Calculate pixel division intervals according to parameter θ; S313: Process the hidden pixels in the order of raster scanning, extract the decimal data according to the interval number of the pixel, and convert it into binary format, using the information to hide the key K h Decrypt the original extra information.

5. The reversible information hiding and predictive recovery method of a fully encrypted image according to claim 3, characterized in that: The S32 comprises the following steps: S321: The receiver sends the image E′ Divide the pixels into two parts: hidden pixels and non-hidden pixels according to the chessboard; S322: For white pixels that cannot be hidden, use the key K e Direct decryption; S323: Obtain the prediction value corresponding to the black grid pixel according to the chessboard pixel prediction method p ; S324: Divide the pixels that can be hidden in black grids into intervals, and obtain adjusted interval pixels; S325: According to the key corresponding to the pixel K e , decrypt the interval pixels and obtain the decrypted interval pixels; S326: Calculate the difference between all decrypted interval pixels and predicted pixels, and use the interval pixels with the smallest absolute value of the difference as the restored pixel value to achieve image restoration.

Images