CN115379066A - Encrypted image reversible data encryption and decryption method based on self-adaptive compression coding - Google Patents

Encrypted image reversible data encryption and decryption method based on self-adaptive compression coding Download PDF

Info

Publication number
CN115379066A
CN115379066A CN202211034832.6A CN202211034832A CN115379066A CN 115379066 A CN115379066 A CN 115379066A CN 202211034832 A CN202211034832 A CN 202211034832A CN 115379066 A CN115379066 A CN 115379066A
Authority
CN
China
Prior art keywords
data
block
image
secret
pixel
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202211034832.6A
Other languages
Chinese (zh)
Other versions
CN115379066B (en
Inventor
隋连升
韩凯峰
肖照林
王战敏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xian University of Technology
Original Assignee
Xian University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xian University of Technology filed Critical Xian University of Technology
Priority to CN202211034832.6A priority Critical patent/CN115379066B/en
Publication of CN115379066A publication Critical patent/CN115379066A/en
Application granted granted Critical
Publication of CN115379066B publication Critical patent/CN115379066B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/32—Circuits or arrangements for control or supervision between transmitter and receiver or between image input and image output device, e.g. between a still-image camera and its memory or between a still-image camera and a printer device
    • H04N1/32101—Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title
    • H04N1/32144—Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title embedded in the image data, i.e. enclosed or integrated in the image, e.g. watermark, super-imposed logo or stamp
    • H04N1/32149—Methods relating to embedding, encoding, decoding, detection or retrieval operations
    • H04N1/32267—Methods relating to embedding, encoding, decoding, detection or retrieval operations combined with processing of the image
    • H04N1/32272—Encryption or ciphering
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/32—Circuits or arrangements for control or supervision between transmitter and receiver or between image input and image output device, e.g. between a still-image camera and its memory or between a still-image camera and a printer device
    • H04N1/32101—Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title
    • H04N1/32144—Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title embedded in the image data, i.e. enclosed or integrated in the image, e.g. watermark, super-imposed logo or stamp
    • H04N1/32149—Methods relating to embedding, encoding, decoding, detection or retrieval operations
    • H04N1/32267—Methods relating to embedding, encoding, decoding, detection or retrieval operations combined with processing of the image
    • H04N1/32277—Compression
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/32—Circuits or arrangements for control or supervision between transmitter and receiver or between image input and image output device, e.g. between a still-image camera and its memory or between a still-image camera and a printer device
    • H04N1/32101—Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title
    • H04N1/32144—Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title embedded in the image data, i.e. enclosed or integrated in the image, e.g. watermark, super-imposed logo or stamp
    • H04N1/32149—Methods relating to embedding, encoding, decoding, detection or retrieval operations
    • H04N1/32347—Reversible embedding, i.e. lossless, invertible, erasable, removable or distorsion-free embedding
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses an encrypted image reversible data encryption method based on self-adaptive compression coding, which comprises the following steps: 1. obtaining a corrected image from the original image; 2. dividing the corrected image into blocks; 3. marking the blocks to generate a position map; 4. compressing the block, generating auxiliary data, generating a matrix, and carrying out XOR on the matrix and the image bit of the recovered data to generate an encrypted image; 5. and calculating the length of the auxiliary data, storing and encrypting to obtain secret data, and 6, obtaining a secret image. The invention also discloses a decryption method of the encryption method, which comprises the following steps: 1. extracting and recovering auxiliary data; 2. extracting a segment of the secret data; 3. obtaining complete secret data and decrypting the secret data to obtain embedded data; 4. recovering the data; 5. generating a matrix, and carrying out bit XOR on the matrix and the image of the recovered data; 6. the original image is obtained. The encryption and decryption method solves the problems of low data embedding rate and low algorithm robustness.

Description

Encryption and decryption method for reversible data of encrypted image based on adaptive compression coding
Technical Field
The invention belongs to the technical field of image processing, relates to an encrypted image reversible data encryption method based on self-adaptive compression coding, and further relates to a decryption method of an encrypted image obtained by encryption by using the encryption method.
Background
With the continuous development of social productivity and internet technology, digital image information is used as an important carrier of internet information and plays a very important role in the interaction of the internet information. In many fields related to personal privacy, business secrets, defense secrets, national security, etc., the security requirements for digital image information are also increasing. The reversible data hiding can completely extract secret data, and a host image can be recovered in a lossless manner, so that the ciphertext and a carrier image can be protected at the same time, and the reversibility is particularly important in various application fields such as medicine, military, legal evidence obtaining and the like. Many studies have been extensively focused on reversible data hiding methods, and embedding mechanisms thereof can be mainly classified into three major categories, histogram shifting, difference expansion and lossless compression. At present, reversible data hiding technology is quite mature, but most of the embedding rate is still not high.
Disclosure of Invention
The invention aims to provide an encrypted image reversible data encryption method based on self-adaptive compression coding, which solves the problems of low data embedding rate and low algorithm robustness in the prior art.
The invention also provides a decryption method for the encryption method of the reversible data of the encrypted image based on the adaptive compression coding.
The technical scheme adopted by the invention is that the reversible data encryption method of the encrypted image based on the self-adaptive compression coding comprises the following steps:
step 1, firstly, an original image I with the size of M multiplied by N o Binary sequence L for original data of least significant bit plane of all pixels in the image sb Storing, and then setting all of them to zero to obtain corrected image I m ；
Step 2, correcting the image I obtained in the step 1 m Dividing into M non-overlapping blocks of size N × N, where M = M × N/N 2 ；
Step 3, setting a threshold value T for the m non-overlapped blocks obtained in the step 2, judging the first H most significant bits of each block independently from the threshold value T, marking the m non-overlapped blocks with two types of smooth blocks and rough blocks, and generating a position map L according to the judgment result;
and 4, respectively compressing the pixels in the smooth block and the rough block in the step 3 by adopting self-adaptive coding to generate necessary auxiliary data: huffman coding rule, reference pixel and recovery sequence R q A label graph; simultaneously, a binary matrix S is generated by utilizing an encryption key in a pseudo-random manner, and the binary matrix S and the corrected image I generated in the step 1 are combined m Performing bit XOR to generate an encrypted image I e ；
Step 5, calculating the length of the auxiliary data generated in the step 4 and recording the length as Len aux Then let Len aux With Huffman coding rule, reference pixel, position diagram L and recovery sequence R q Binary sequence L of label graph and original graph sb Sequentially storing to the encrypted image I generated in step 4 e In the least significant bit plane of; the embedded data d is encrypted by using the data hidden secret key to obtain secret data d e ；
Step 6, secret data d obtained in step 5 e Storing the encrypted image I generated in step 4 e The middle flat sliding block and the rough block are compressed to provide an embeddable space to obtain a secret image I carrying secret data ew 。
In step 3, the specific method for marking the non-overlapped blocks as the smooth blocks and the rough blocks comprises the following steps: for I m For each block in the block, retrieving the first H e {4,5,6,7} most significant bits, i.e. HMSB, of each pixel in the block, and classifying the same HMSB in the block into one, if there are λ kinds of HMSB in the block and not greater than a preset threshold T e {3,4,5,6,7}, then the current block is marked as a flat block, otherwise, the current block is marked as a rough block;
the method for generating the position map L comprises the following steps: each block is marked with a binary bit, with a "1" for smooth blocks and a "0" for rough blocks, resulting in the formula:
Figure BDA0003818706880000031
the compression method of the flat sliding block in the step 4 comprises the following steps: set up (alpha) 1 ,α 2 ,α 3 ) For expressing the lambda of the slider, i.e. (alpha) 1 ,α 2 ,α 3 ) Representing how many HMSB types are in common in the smooth block, and connecting the original data of the HMSB types in series according to the sequence of the HMSB types from high to low in the block to form a binary sequence, and connecting three bits of (alpha) 1 ,α 2 ,α 3 ) And λ × H bits of the original HMSB data to generate the auxiliary data required for the current slider block, representing the number of slider blocks as u (≦ m), scanning these slider blocks and calculating their auxiliary data, and concatenating them sequentially to obtain a recovery sequence Rq whose Rq has a length η:
Figure BDA0003818706880000032
where r is the subscript, λ, of the current slider r The number of different types of HMSB in the current slider.
The compression method of the rough block in the step 4 comprises the following steps: first, a corrected image I is calculated m The prediction value of the pixels of the medium coarse block is as follows:
(1) first row and first column of pixels I m (1,1) as predicted;
(2) if the current pixel is positioned on the first row, the value of the left adjacent pixel is the predicted value of the current pixel;
(3) if the current pixel is positioned in the first column, the value of the pixel adjacent to the current pixel above the current pixel is the predicted value of the current pixel;
(4) for other coarse blocks of pixels I m (i, j) the predicted value is calculated by using a Median Edge Detection (MED) method, and the calculation formula is as follows:
Figure BDA0003818706880000041
wherein a = I m (i-1,j-1),b＝I m (i,j-1),c＝I m (i-1,j)，
Figure BDA0003818706880000042
Is a 1 m (i, j) corresponding to the predicted value. Next, the original and predicted values are converted into two 8-bit binary sequences, respectively, using the following equations:
Figure BDA0003818706880000043
wherein P is k (i, j) and
Figure BDA0003818706880000044
comparing the two sequences from Most Significant Bit (MSB) to Least Significant Bit (LSB) for the binary sequences corresponding to the original value and the predicted value respectively until a bit is different, and marking how many bits are the same from MSB to LSB between the original value and the predicted value sequence by a tag value epsilon, the calculation formula is as follows:
Figure BDA0003818706880000051
taking the first 7 bits of the two sequences to process, correcting the first row and the first column of pixels I in the image m (1,1) will be stored as the only reference pixel.
Scanning all rough block pixels, and obtaining the label images tau corresponding to all rough block pixels through the processing process.
In step 4, the specific steps of performing adaptive compression coding on the sliding block are as follows: each smooth block pixel
Figure BDA0003818706880000052
The adaptive coding result of (2) is defined as
Figure BDA0003818706880000053
Figure BDA0003818706880000054
In the formula
Figure BDA0003818706880000055
Indicates a plurality of prefix bits, which are composed of a plurality of '0' and a termination bit '1' which are continuous, and is the adaptive coding result of the HMSB part of the current smooth pixel,
Figure BDA0003818706880000056
is the unmodified portion after HMSB;
(1) performing self-adaptive coding on different HMSB in the sliding block according to the type number of the lambda and obtaining a code word of the HMSB;
(2) adding successive '0's and one stop bit '1', i.e. adding
Figure BDA0003818706880000057
Before its code word to satisfy the coding result in formula (6);
(3) after the completion of the two steps, all n in the flat sliding block 2 The HMSB of each pixel is performed by adaptive compression coding.
In step 4, the specific steps of performing adaptive compression coding on the rough block are as follows: defining 8 Huffman codes to represent eight label values epsilon, namely { "00", "01", "100", "101", "1100", "1101", "1110", "1111" }, calculating the number of different types of label values according to the label graph tau, sorting the eight labels according to the occurrence frequency from high to low, and sequentially allocating 8 Huffman codes, wherein after the compression process is realized, the length of the label graph tau after Huffman coding can be calculated by the following formula:
Figure BDA0003818706880000061
in the formula, the label value in the coarse block is represented as the number of pixels, and the length of the corresponding huffman code is represented.
In step 4, the image I is encrypted e The calculation formula of (c) is:
Figure BDA0003818706880000062
in step 5, the length Len of the auxiliary data aux Is 20 bits in length and is placed at the head of the entire auxiliary data sequence, the long binary auxiliary data sequence.
In step 5, the whole encrypted image I e Three embeddable spaces are totally arranged, the least significant bit plane, the embeddable space in the smooth block and the embeddable space in the rough block are embedded into the encrypted image I according to the sequence of the three embeddable spaces by a bit replacement mode e In (1).
The technical scheme adopted by the invention is also that the method for decrypting the reversible data of the encrypted image based on the self-adaptive compression coding comprises the following steps:
step 1, carrying out secret image I ew All auxiliary data in the least significant bit plane of (a) are extracted and sequentially restored to obtain: auxiliary data length Len aux Huffman coding rule, reference pixel, position map L, recovery sequence Rq, label map tau and original binary sequence L sb ；
Step 2, comparing the auxiliary data length Len aux And a secret-carrying image I ew Determining secret data d by using minimum-significant bit-plane embeddable quantity MxN e And extracting the part of the secret data; distinguishing secret-carrying images I from location maps ew The flat sliding block is extracted from the secret data part embedded in the flat sliding block by depending on the self-adaptive coding rule table; distinguishing secret-carrying images I from location maps ew The residual secret data embedded in the rough block is extracted by the rough block depending on the Huffman coding rule and the label graph tau;
step 3, connecting the secret data segments extracted in the step 2 in sequence to obtain complete secret data d e And hiding the secret data d with the data hiding key e Decrypting to obtain decrypted embedded data d;
step 4, recovering the data in all the smooth blocks according to the recovery sequence Rq and the position diagram L; restoring data in all rough blocks according to a Huffman coding rule, reference pixels, a label graph and a position graph;
step 5, generating a binary matrix S by utilizing the encryption key in a pseudo-random manner, and carrying out bit XOR on the binary matrix S and the image processed in the step 4;
6, binary sequence L of original image sb The data in the step (5) are correspondingly put back to the least significant bit plane of the image processed in the step (5) one by one, and finally the original image I which is recovered without damage is obtained o 。
The invention has the following beneficial effects:
the invention relates to an encrypted image reversible data encryption and decryption method based on self-adaptive compression coding, which provides a self-adaptive compression coding scheme, firstly, an original image is preprocessed, then, the original image is divided into a smooth block and a rough block which are respectively compressed, image redundancy and high pixel correlation between adjacent pixels are fully utilized, a median edge detection prediction method is innovatively introduced, a larger embeddable space is created for embedding secret data, and the data embedding rate is greatly improved; by applying the bit XOR operation between the pixel and the pseudo-random matrix generated by using the encryption key, the safety of secret data is fully ensured, and the robustness of the algorithm is higher; the necessary auxiliary data generated during the adaptive compression coding ensures that the original image can be recovered intact at the receiver side.
Drawings
FIG. 1 is a flow chart of the reversible data encryption and decryption method for encrypted images based on adaptive compression coding according to the present invention;
FIG. 2 is a schematic diagram of an exemplary block of 4 × 4 size for the adaptive compression coding based encrypted image reversible data encryption and decryption method of the present invention;
FIG. 3 is a schematic diagram of a binary form of an exemplary block of the encrypted image reversible data encryption and decryption method based on adaptive compression coding according to the present invention;
FIG. 4 is a schematic diagram of four HMSB occurrence frequencies and auxiliary data generation processes in an exemplary smooth block of the reversible data encryption and decryption method for encrypted images based on adaptive compression coding according to the present invention;
FIG. 5 is a schematic diagram of the process of generating label values for the coarse block pixels of the reversible data encryption and decryption method for encrypted images based on adaptive compression coding according to the present invention;
FIG. 6 is a schematic diagram of the adaptive encoding rule of the smooth block of the reversible data encryption and decryption method for encrypted images based on adaptive compression encoding according to the present invention;
FIG. 7 is a schematic diagram of an example of adaptive encoding of a smooth block in the reversible data encryption and decryption method for encrypted images based on adaptive compression encoding according to the present invention;
FIG. 8 is a standard gray scale Lena chart of 512 × 512 size for the reversible data encryption and decryption method for encrypted images based on adaptive compression coding according to the present invention;
FIG. 9 is an encrypted Lena diagram of the reversible data encryption and decryption method for encrypted images based on adaptive compression coding according to the present invention;
FIG. 10 is a block diagram of the auxiliary data sequence composition of the reversible data encryption/decryption method for encrypted images based on adaptive compression coding according to the present invention;
FIG. 11 is a diagram illustrating the histogram analysis result of a standard gray Lena graph of the reversible data encryption and decryption method for encrypted images based on adaptive compression coding according to the present invention;
FIG. 12 is a diagram showing the histogram analysis result of the encrypted image according to the reversible data encryption/decryption method for encrypted images based on adaptive compression coding of the present invention;
FIG. 13 is a diagram showing the histogram analysis result of the encrypted image according to the reversible data encryption/decryption method for encrypted images based on adaptive compression coding according to the present invention;
fig. 14 is a schematic diagram of comparison and analysis of experimental results of the encryption and decryption method of reversible data of encrypted images based on adaptive compression coding according to the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples.
The overall flow chart of the encryption image reversible data encryption method based on the adaptive compression coding is shown in fig. 1, and the method comprises the following steps:
step 1, original image I with size of M multiplied by N o Require pre-processing, i.e. modifying I with "0 o Then get the corrected image I m . Since a part of the subsequently calculated auxiliary data is embedded in the least significant bits of the encrypted pixels, in order to restore the original image I losslessly in the future o Must preserve I o The least significant bit of all pixels in the array. Thus, a binary sequence L is used sb Storage I o The least significant bit of all pixels in the array.
Step 2, correcting the image I m Is divided into M non-overlapping blocks of size N × N, where M = M × N/N 2 。
And 3, in order to obtain more embeddable spaces, dividing the m non-overlapped blocks obtained in the step 2 into a smooth block and a rough block according to the correlation between pixels, and then processing the blocks by using two different compression coding modes. For I m For each block in the block, the first H e {4,5,6,7} Most Significant bits (Most Significant Bit), i.e., H MSBs, of each pixel in the block are retrieved, and the same H MSBs in the block are classified as one. Defining: if the block has lambda kinds of H MSBs and lambda is not more than a preset threshold T epsilon {3,4,5,6,7}, the current block is determined as a smooth block, otherwise, the current block is determined as a rough block.
For example, the following steps are carried out: let H =5 and T =4, which means that 5 MSBs of all pixels in the current block are used. A 4 x 4 block is shown in fig. 2, and its binary form is given in fig. 3. It can be found that there are four different 5 MSBs in the block, i.e. λ =4. In this example, the threshold T is set to 4, so the current block satisfies λ ≦ T, and is therefore determined to be a slider.
According to the above determination method, the m non-overlapping blocks are sequentially determined to generate the position map L, and each block is marked as a smooth block or a rough block. Each block is marked with a binary bit, with a "1" for smooth blocks and a "0" for rough blocks. The binary position diagram L is generated according to the following formula:
Figure BDA0003818706880000101
and 4, respectively compressing the pixels in the smooth block and the rough block by adopting self-adaptive coding, and simultaneously generating necessary auxiliary data: the method comprises the steps of a Huffman coding rule, reference pixels, a recovery sequence Rq and a label graph.
Step 4.1, for the smooth block compression, three bits (α) are used since the maximum value of the threshold T is 7 1 ,α 2 ,α 3 ) To denote lambda of the slider. I.e., (alpha) 1 ,α 2 ,α 3 ) Which is used to indicate how many H MSBs are in common in a flat slider. Specifically, when λ equals 1,2,3,4,5,6 and 7, the corresponding (α) 1 ,α 2 ,α 3 ) Respectively as follows: "000","001","010","011","100","101","110". And then according to the sequence of the occurrence frequency of the lambda kinds of H MSBs in the block from high to low, the original data of the lambda kinds of H MSBs are connected in series to form a binary sequence of lambda multiplied by H. Of three bits in series (alpha) 1 ,α 2 ,α 3 ) And λ × H bits of original H MSB data to generate the auxiliary data required for the current slider. The number of sliders is denoted as u (≦ m), these sliders are scanned and their auxiliary data are calculated. Subsequently, they are connected in sequence to give a recovered sequence Rq whose length, η, is:
Figure BDA0003818706880000111
where r is the subscript, λ, of the current slider r The number of different types of H MSBs in the current smooth block. Continuing with the 4 x 4 block example of fig. 2, fig. 4 gives an example of the auxiliary data generation for this smooth block. From high to low in frequency of occurrence, 5 MSBs in a block: "00111", "01000", "00110", "01001" appeared 9, 4, 2 and 1 times, respectively. Therefore, the auxiliary data of the current flat slider, i.e., { "011", "00111", "01000", "00110", "01001" }, can be generated by connecting 3+4 × 5=23 bits.
Step 4.2, for the compression of the coarse block, firstly, the corrected image I is calculated m The predicted value of the pixel of the medium coarse block. The prediction rule is as follows:
(1) first row and first column of pixels I m The predictor of (1,1) is itself.
(2) If the current pixel is in the first row, the value of its left-hand neighboring pixel is its predicted value.
(3) If the current pixel is in the first column, the value of its upper neighbor is its predicted value.
(4) For other coarse blocks of pixels I m (i, j), the predicted value is calculated by using a Median Edge Detection (MED) method, and the calculation formula is as follows:
Figure BDA0003818706880000121
wherein a = I m (i-1,j-1),b＝I m (i,j-1),c＝I m (i-1,j)，
Figure BDA0003818706880000122
Is I m (i, j) corresponding to the predicted value. Next, the original and predicted values are converted into two 8-bit binary sequences, respectively, using the following equations:
Figure BDA0003818706880000123
wherein P is k (i, j) and
Figure BDA0003818706880000124
the binary sequences are respectively corresponding to the original value and the predicted value. The two sequences are compared from the Most Significant Bit (MSB) to the Least Significant Bit (LSB) until one of the bits is different. The tag value epsilon is then used to mark how many bits from MSB to LSB are the same between its original value and the predicted value sequence, and the calculation formula is as follows:
Figure BDA0003818706880000125
although the two sequences are 8 bits in length, only the first 7 bits of the two sequences are taken for processing because the modified image I m Has previously been zeroed out and will be used as part of the embedding space. Thus, epsilon is an integer no greater than 7, which means that epsilon has a maximum of eight values. During the embedding phase, the pixels in each coarse block may provide an embedding space of ε +1 bits. It is noted that the first row and column of pixels I in the modified image m (1,1) will be stored as the only reference pixel since the subsequent image recovery stage needs to use the reference pixel to recover all the coarse block pixels.
For example, as shown in fig. 5, if the original pixel value of a rough block is 162 and the predicted value is 168 after prediction calculation, the two values can be converted into eight-bit binary forms "10100010" and "10101000", respectively. The first 4 bits are found to be the same from the MSB to LSB comparison, so the label value of the coarse block pixel, e =4, also represents that 5 bits are available for embedding extra data for this coarse block pixel.
And 4.3, scanning all rough block pixels, and obtaining the label maps tau corresponding to all the rough block pixels through the processing process of the step 4.2. A large amount of embeddable space can then be created in these coarse block pixels to accommodate the additional data according to the tag map.
Step 4.4, the sliding block is subjected to self-adaptive compression coding, and each sliding block pixel is subjected to self-adaptive compression coding
Figure BDA0003818706880000131
The adaptive coding result of (2) is defined as
Figure BDA0003818706880000132
Figure BDA0003818706880000133
Wherein
Figure BDA0003818706880000134
It represents a plurality of prefix bits, and is composed of a plurality of continuous '0' and a termination bit '1'.
Figure BDA0003818706880000135
Is the current smooth pixel
Figure BDA0003818706880000136
The adaptive coding result of the HMSB part of (1).
Figure BDA0003818706880000137
Is the unmodified part after HMSB. The detailed steps of the slider-slider pixel encoding are given below:
(1) as shown in fig. 6, different H MSBs in the flat sliding block are adaptively encoded according to the number of types of λ and the codeword thereof is obtained. Obviously, H MSBs that occur more frequently will be encoded using shorter codewords.
(2) Adding successive '0's and one stop bit '1', i.e. adding
Figure BDA0003818706880000138
Before its code word to satisfy the coding result in equation (6).
(3) After the completion of the two steps, all n in the flat sliding block 2 The H MSBs of each pixel are all done by adaptive compression coding. However, it is noted that (8-H) LSB, i.e., LSB
Figure BDA0003818706880000141
And all the time, each encoded slider pixel still consists of 8 bits.
The encoding results for the smooth block pixel H MSB for different λ cases are listed in fig. 6. In detail, for a coded smooth block, if λ =1, it indicates that there is only one type of adaptive codeword, i.e., "1"; if λ =2, two types of adaptive codewords are indicated, namely "1" and "0"; if λ =3, three types of adaptive codewords are indicated, i.e. "1", "10", "00", and so on. Note that for each smooth block, the adaptive compression coding result corresponding to the λ H MSB corresponds to the H MSB frequency from high to low. The H MSB with the highest frequency of occurrence in the block is encoded with the shortest codeword, whereas the H MSB with the longest codeword is encoded with the opposite.
Continuing with the example of the block in fig. 2, the left side of fig. 7 shows four adaptive codewords of 5 MSBs within the block: {"1","10","100","000"}. The right side of fig. 7 is the adaptive compression coding result { "00011", "00110", "01100", "01000" } obtained by adding a number of prefix bits "0" and a termination bit "1".
And 4.5, carrying out self-adaptive compression coding on the rough block. As described in step 4.2, the label value epsilon has eight possibilities of values, namely {0,1,2,3,4,5,6,7}. Accordingly, we define a set of 8 huffman codes to represent the eight label values, i.e., { "00", "01", "100", "101", "1100", "1101", "1110", "1111" }. Firstly, the number of different types of label values is calculated through the label graph obtained in the step 5.3, then the eight types of labels are sorted according to the occurrence frequency from high to low, and 8 types of Huffman codes are sequentially distributed. That is, where a "00" huffman code would represent the most numerous tag values, and a "1111" huffman code would represent the least numerous tag values. The use of huffman codes will help to reduce the amount of side data and thus to obtain better embedding capacity. It is noted that the huffman coding rule consisting of 8 huffman codes should be considered as a part of the auxiliary data to be saved.
As shown in fig. 8, taking the most classical 512 × 512-sized Lena graph in digital image processing as an example, in our method, when the correlation parameters H =5, t =4, n =4, there are 4282 coarse blocks, that is, there are 68512 coarse block pixels. The number of 8 label values is calculated, which corresponds to the compression rule as shown in the following table:
Label 0 1 2 3 4 5 6 7
Distribution 4152 5857 7066 13183 14123 10483 6535 7113
Code 1111 1110 1100 01 00 100 1101 101
it should be noted that the number of bits occupied by the compression rule is fixed, i.e. 26 bits. After the compression process is implemented, the length of the label graph τ after huffman coding can be calculated by the following formula:
Figure BDA0003818706880000151
wherein n is t Representing the number of pixels with tag value t in the coarse block, c t Indicating the length of the corresponding huffman code.
Then, a binary matrix S is generated by utilizing the encryption key in a pseudo-random manner, and the binary matrix S and the corrected image I are combined m Performing bit XOR to generate an encrypted image I e As shown in fig. 9, the calculation formula is as follows:
Figure BDA0003818706880000152
step 5, calculating the lengths of all auxiliary data and recording the lengths as Len aux Then all the auxiliary data, namely: huffman coding rule, reference pixel, position diagram L, recovery sequence Rq, label diagram and original image binary sequence L sb Sequentially storing to encrypted image I e In the least significant bit plane of (a).
Step 5.1, calculating the length Len of all auxiliary data aux Stored with 20 bits and placed at the head of the entire auxiliary data sequence, forming a long binary auxiliary data sequence as shown in fig. 10.
Step 5.2, it should be noted that the entire encrypted image I e There are three embeddable spaces, namely: (1) a least significant bit plane, (2) an embeddable space in a smooth block, and (3) an embeddable space in a rough block. Embedding the whole binary auxiliary data sequence into the encrypted image I in the sequence of the three embeddable spaces by means of bit replacement e In (1).
Step 6, encrypting the embedded data d by using the data hidden secret key to obtain secret data d e (ii) a Secret data d e Stored in the encrypted image I e The rest of the space can be embedded into the space to obtain a secret image I carrying secret data ew 。
The method for decrypting the reversible data of the encrypted image based on the self-adaptive compression coding comprises the following steps:
step 1, carrying out secret image I ew All auxiliary data in the least significant bit plane of (a) are extracted and sequentially restored to obtain: auxiliary data length Len aux Huffman coding rule, reference pixel, position map L, recovery sequence Rq, label map tau and original binary sequence L sb ；
Step 2, comparing the auxiliary data length Len aux And a secret-carrying image I ew Determining secret data d by using minimum-significant bit-plane embeddable quantity MxN e And extracting the part of the secret data; distinguishing secret-carrying images I from location maps ew The flat sliding block is extracted from the secret data part embedded in the flat sliding block by depending on a self-adaptive coding rule table; then distinguish the secret-carrying image I according to the position diagram ew The residual secret data embedded in the rough block is extracted from the rough block by relying on the Huffman coding rule and the label graph tau;
step 3, connecting the secret data segments extracted in the step 2 in sequence to obtain complete secret data d e And hiding the secret data d with the data hiding key e Decrypting to obtain decrypted embedded data d;
and 4, restoring the data in all the smooth blocks according to the restoring sequence Rq and the position diagram L. Since the type number of the H MSB in each flat block and the original data are sequentially stored in the recovery sequence Rq, the original data of each flat block can be directly extracted to recover the original data of the H MSB in the flat block. Note that the data of other bit planes, except for the H MSB, are not moved throughout the entire process, so no recovery operation is required; and recovering the data in all the rough blocks according to the Huffman coding rule, the reference pixels, the label graph tau and the position graph L.
The method comprises the following specific steps: first, the original pixel data of the first row and the first column are calculated by referring to the pixels, and then the predicted values of all the pixels are calculated by a median detection Method (MED). Then, according to the label value corresponding to each rough pixel in the label map, it can be known that the front epsilon +1 bit of the rough pixel is modified. And because the front epsilon bit of the coarse pixel is completely the same as the front epsilon bit of the predicted value, the front epsilon bit of the predicted value can be directly covered and recovered. Similarly, the ε +1 of a coarse pixel may be recovered by inverting the ε +1 bit of its predicted value. Until the data in the rough block is completely recovered;
step 5, generating a binary matrix S by utilizing the encryption key in a pseudo-random manner, and carrying out bit XOR on the binary matrix S and the image processed in the step 4;
step 6, because in the initial preprocessing, the original image I is processed o All the least significant bit planes of the original image are set to zero, so that the original image binary sequence L is set to zero sb The data in the step (5) are put back to the least significant bit plane of the image processed in the step (5) in a one-to-one correspondence way, and finally the original image I o Can be recovered without loss.
The beneficial effects of the invention are verified as follows:
by analyzing the histogram of the Lena chart, the encrypted Lena histogram and the secret-carrying Lena histogram containing the secret data, as shown in fig. 11, 12 and 13, it can be seen by comparison that the distribution of the histogram is more uniform in the encrypted image compared with the original image, and after the secret data is embedded in the encrypted image, the histogram thereof is relatively uniformly distributed in the overall view despite slight fluctuation, and it is difficult to observe any effective information therefrom.
The indexes for judging and measuring the image encryption algorithm are NPCR and UACI, wherein the corresponding ideal values are 99.61% and 33.46% respectively, and the formula is as follows:
Figure BDA0003818706880000181
Figure BDA0003818706880000182
Figure BDA0003818706880000183
in the above equation, the rows and columns of the image are denoted by M and N, respectively, and the plaintext image q 1 The pixel value at the (i, j) position is represented by q 1 (i, j) representing the ciphertext image q 2 The pixel value at the (i, j) position is represented by q 2 (i, j) represents, and q represents 1 (i, j) and q 2 The difference in pixel values between (i, j) is only 1. Let C (i, j) be a binary matrix, and the size of C (i, j) and q 1 、q 1 Same if q 1 (i, j) and q 2 (i, j) are equal, i.e. q 1 (i,j)＝q 2 (i, j), then C (i, j) =1; if q is 1 (i, j) and q 2 (i, j) are not equal, i.e. q 1 (i,j)≠q 2 (i, j), then C (i, j) =1.
The results of encrypting 10 classical test images by the method of the invention and performing NPCR and UACI calculation on the encrypted images are shown in the following table, and it can be seen that the algorithm provided by the invention has extremely strong differential attack resistance.
Figure BDA0003818706880000191
Because high correlation exists between adjacent pixels of the image, information of surrounding pixels of one pixel is often leaked, and an attacker can use the characteristic to estimate the gray value of the next pixel, so that the recovery of the whole plaintext image is realized. The stronger the image encryption algorithm is against attacks, the smaller the correlation of the image before and after encryption should be. To test the encryption algorithm of the present invention, the correlation between the original image and the encrypted image was calculated:
Figure BDA0003818706880000192
wherein, cov (I, I ″) ew ) Is the covariance between the original image and the encrypted image, σ (I) and σ (I ″) ew ) Is the standard deviation. In addition, information entropy is commonly used to evaluate randomness of an encrypted image, and the formula is:
Figure BDA0003818706880000193
wherein x i Is the gray value, P (x) i ) Is the frequency at which it correspondingly occurs. For encrypted images, the ideal information entropy is 8, the higher the information entropy, the more uniform the distribution of the image.
For 10 classical test images, the correlation coefficient and information entropy are calculated, and the results are shown in the following table:
Figure BDA0003818706880000194
as can be seen from the test data results in the table above, the correlation coefficients of the 10 test images are all close to 0, which indicates that there is almost no correlation between the original image and the corresponding encrypted image; the information entropy of the images is close to 8, which shows that the distribution of the encrypted images is uniform. This effectively indicates that the inventive encryption algorithm is resistant to attack by a lawbreaker.
The data embedding rate is also an important index for measuring the reversible data hiding method of the encrypted image. As shown in fig. 14, comparing the data embedding rates of the present invention and the conventional excellent reversible data hiding method for encrypted images, it can be seen that the data embedding rate of the present invention is the highest compared to the test image.
The experiment is not limited to these images, but extends to all images in three databases, UCID, BOWS2 and BOSSBase, to demonstrate the universal embedding capability of the method. As shown in the following table, for the database UCID, the average embedding rate of the method is 2.8223bpp, and the average increase amounts of the other five methods [ Puteaux ], [ BBE ], [ YIn2], [ INS ] and [ YIn ] are 1.9293, 0.9957, 0.5540, 0.7025 and 0.1347, respectively. For the larger database BOWS2, the average embedding rate increases slightly, but still outperforms the other databases. Also, the BOSSBase database may conclude the same, i.e. that the proposed method on average always has the best embedding results.
Figure BDA0003818706880000201
In conclusion, through the analysis, the superiority of the method in embedding capacity and the safety of the algorithm are verified experimentally. The invention discloses an encrypted image reversible data hiding method based on self-adaptive compression coding, which provides a self-adaptive compression coding mode that a leveling block and a rough block are respectively processed, fully utilizes image redundancy, creates larger space for embedding secret data, and provides more space for embedding the secret data. By using bit exclusive or calculation between pseudo random matrices generated by the encryption key, the security of the secret data and the whole algorithm can be ensured, so that the robustness of the algorithm is higher.

Claims (10)

1. The method for encrypting the reversible data of the encrypted image based on the self-adaptive compression coding is characterized by comprising the following steps:
step 1, firstly, an original image I with the size of M multiplied by N o Binary sequence L for original data of least significant bit plane of all pixels in the image sb Storing, and then setting all of them to zero to obtain corrected image I m ；
Step 2, correcting the image I obtained in the step 1 m Dividing into M non-overlapping blocks of size N × N, where M = M × N/N 2 ；
Step 3, setting a threshold value T for the m non-overlapped blocks obtained in the step 2, judging the first H most significant bits of each block independently from the threshold value T, marking the m non-overlapped blocks with two types of smooth blocks and rough blocks, and generating a position map L according to the judgment result;
step 4, adopting self-adaptive weavingThe code compresses the pixels in the smooth block and the rough block in the step 3 respectively to generate necessary auxiliary data: huffman coding rule, reference pixel and recovery sequence R q A label graph; simultaneously, a binary matrix S is generated by utilizing an encryption key in a pseudo-random manner, and the binary matrix S and the corrected image I generated in the step 1 are combined m Performing bit XOR to generate an encrypted image I e ；
Step 5, calculating the length of the auxiliary data generated in the step 4 and recording the length as Len aux Then let Len aux With Huffman coding rule, reference pixel, position diagram L and recovery sequence R q Binary sequence L of label graph and original graph sb Sequentially storing to the encrypted image I generated in step 4 e In the least significant bit plane of; the embedded data d is encrypted by using the data hidden secret key to obtain secret data d e ；
Step 6, secret data d obtained in step 5 e Storing the encrypted image I generated in step 4 e The middle flat sliding block and the rough block are compressed to provide an embeddable space to obtain a secret image I carrying secret data ew 。
2. The encryption method according to claim 1, wherein the specific method for marking non-overlapped blocks as smooth blocks and rough blocks in step 3 is as follows: for I m For each block in the block, retrieving the first H e {4,5,6,7} most significant bits, i.e. HMSB, of each pixel in the block, and classifying the same HMSB in the block into one, if there are λ kinds of HMSB in the block and not greater than a preset threshold T e {3,4,5,6,7}, then the current block is marked as a flat block, otherwise, the current block is marked as a rough block;
the method for generating the position map L in the step 3 comprises the following steps: each block is marked with a binary bit, with a "1" for smooth blocks and a "0" for rough blocks, the formula being:
Figure FDA0003818706870000021
3. the encryption method according to claim 1, wherein the compression method of the slider block in the step 4 is: set up (alpha) 1 ,α 2 ,α 3 ) For λ of the flat slider, i.e. (α) 1 ,α 2 ,α 3 ) Representing how many HMSB are in common in the flat slider, and connecting the original data of the HMSB in series according to the order of the HMSB from high to low in the block to form a binary sequence, connecting three bits of (alpha) 1 ,α 2 ,α 3 ) And original HMSB data of λ H bits to generate the auxiliary data required for the current slider, representing the number of sliders as u (≦ m), scanning these sliders and calculating their auxiliary data, and concatenating them in sequence to obtain a recovery sequence Rq, the length η of which is:
Figure FDA0003818706870000031
where r is the subscript, λ, of the current slider r The number of different types of HMSB in the current slider.
4. The encryption method according to claim 1, wherein the compression method of the rough block in the step 4 is: first, a corrected image I is calculated m The prediction value of the pixels of the medium coarse block is as follows:
(1) first row and first column of pixels I m (1,1) as predicted;
(2) if the current pixel is positioned on the first row, the value of the left adjacent pixel is the predicted value of the current pixel;
(3) if the current pixel is positioned in the first column, the value of the pixel adjacent to the current pixel above the current pixel is the predicted value of the current pixel;
(4) for other coarse blocks of pixels I m (i, j), the predicted value is calculated by using a Median Edge Detection (MED) method, and the calculation formula is as follows:
Figure FDA0003818706870000032
wherein a = I m (i-1,j-1),b＝I m (i,j-1),c＝I m (i-1,j)，
Figure FDA0003818706870000033
Is I m (i, j) corresponding to the predicted value. Next, the original and predicted values are converted into two binary sequences of 8 bits, respectively, using the following formula:
Figure FDA0003818706870000034
wherein P is k (i, j) and
Figure FDA0003818706870000041
comparing the two sequences from Most Significant Bit (MSB) to Least Significant Bit (LSB) for the binary sequences corresponding to the original value and the predicted value respectively until a bit is different, and marking how many bits are the same from MSB to LSB between the original value and the predicted value sequence by a tag value epsilon, the calculation formula is as follows:
Figure FDA0003818706870000042
taking the first 7 bits of the two sequences to process, correcting the first row and the first column of pixels I in the image m (1,1) will be stored as the only reference pixel;
scanning all rough block pixels, and obtaining the label images tau corresponding to all rough block pixels through the processing process.
5. The encryption method according to claim 1, wherein in the step 4, the adaptive compression encoding of the smooth block comprises the following specific steps: each smooth block pixel
Figure FDA0003818706870000043
The adaptive coding result of (2) is defined as
Figure FDA0003818706870000044
Figure FDA0003818706870000045
In the formula
Figure FDA0003818706870000046
Indicates a plurality of prefix bits, which are composed of a plurality of '0' and a termination bit '1' which are continuous, and is the adaptive coding result of the HMSB part of the current smooth pixel,
Figure FDA0003818706870000047
is the unmodified portion after HMSB;
(1) performing self-adaptive coding on different HMSB in the sliding block according to the type number of the lambda and obtaining a code word of the HMSB;
(2) adding successive '0's and one stop bit '1', i.e. adding
Figure FDA0003818706870000048
Before its code word to satisfy the coding result in formula (6);
(3) after the two steps are completed, all n in the flat sliding block are totally n 2 The HMSB of each pixel is performed by adaptive compression coding.
6. The encryption method according to claim 1, wherein in the step 4, the adaptive compression encoding of the coarse block comprises the following specific steps: defining 8 Huffman codes to represent eight label values epsilon, namely { "00", "01", "100", "101", "1100", "1101", "1110", "1111" }, calculating the number of different types of label values according to the label graph tau, sorting the eight labels according to the occurrence frequency from high to low, and sequentially allocating 8 Huffman codes, wherein after the compression process is realized, the length of the label graph tau after Huffman coding can be calculated by the following formula:
Figure FDA0003818706870000051
in the formula, the label value in the coarse block is represented as the number of pixels, and the length of the corresponding huffman code is represented.
7. The encryption method according to claim 1, wherein in said step 4, the image I is encrypted e The calculation formula of (2) is as follows:
Figure FDA0003818706870000052
8. the encryption method according to claim 1, wherein in said step 5, the length Len of the auxiliary data aux Is 20 bits in length and is placed at the head of the entire auxiliary data sequence, the long binary auxiliary data sequence.
9. The encryption method according to claim 1, wherein in said step 5, the whole encrypted image I e Three embeddable spaces are totally arranged, the least significant bit plane, the embeddable space in the smooth block and the embeddable space in the rough block are embedded into the encrypted image I according to the sequence of the three embeddable spaces by a bit replacement mode e In (1).
10. An encrypted image reversible data decryption method based on adaptive compression coding, characterized in that, the decryption method is used for the encrypted image obtained by the encryption method of claims 1-9, and comprises the following steps:
step 1, secret-carrying pictureLike I ew All auxiliary data in the least significant bit plane of (a) are extracted and sequentially restored to obtain: auxiliary data length Len aux Huffman coding rule, reference pixel, position map L, recovery sequence Rq, label map tau and original binary sequence L sb ；
Step 2, comparing the auxiliary data length Len aux And a secret-carrying image I ew Determining secret data d by using minimum-significant bit-plane embeddable quantity MxN e And extracting the part of the secret data; distinguishing secret-carrying images I from location maps ew The flat sliding block is extracted from the secret data part embedded in the flat sliding block by depending on the self-adaptive coding rule table; distinguishing secret-carrying images I from location maps ew The residual secret data embedded in the rough block is extracted by the rough block depending on the Huffman coding rule and the label graph tau;
step 3, connecting the secret data segments extracted in the step 2 in sequence to obtain complete secret data d e And using the data hiding key to the secret data d e Decrypting to obtain decrypted embedded data d;
step 4, recovering the data in all the smooth blocks according to the recovery sequence Rq and the position diagram L; restoring data in all rough blocks according to a Huffman coding rule, reference pixels, a label graph and a position graph;
step 5, generating a binary matrix S by utilizing the encryption key in a pseudo-random manner, and carrying out bit XOR on the binary matrix S and the image processed in the step 4;
6, binary sequence L of original image sb The data in the step (5) are correspondingly put back to the least significant bit plane of the image processed in the step (5) one by one, and finally the original image I which is recovered without damage is obtained o 。
CN202211034832.6A 2022-08-26 2022-08-26 Encryption image reversible data encryption and decryption method based on self-adaptive compression coding Active CN115379066B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211034832.6A CN115379066B (en) 2022-08-26 2022-08-26 Encryption image reversible data encryption and decryption method based on self-adaptive compression coding

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211034832.6A CN115379066B (en) 2022-08-26 2022-08-26 Encryption image reversible data encryption and decryption method based on self-adaptive compression coding

Publications (2)

Publication Number Publication Date
CN115379066A true CN115379066A (en) 2022-11-22
CN115379066B CN115379066B (en) 2023-07-25

Family

ID=84067943

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211034832.6A Active CN115379066B (en) 2022-08-26 2022-08-26 Encryption image reversible data encryption and decryption method based on self-adaptive compression coding

Country Status (1)

Country Link
CN (1) CN115379066B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117082154A (en) * 2023-10-16 2023-11-17 长沙瑞腾信息技术有限公司 Big data-based double-path server data storage system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005051793A (en) * 1995-05-08 2005-02-24 Digimarc Corp Steganographic system
JP2006222788A (en) * 2005-02-10 2006-08-24 Oki Electric Ind Co Ltd Image verification device, image embedding device, image detection device, image embedding method, image detection method, computer program, and printed matter manufacturing method
CN110913092A (en) * 2019-11-26 2020-03-24 西南交通大学 Reversible information hiding method for encrypted image
CN111898138A (en) * 2020-07-06 2020-11-06 西北工业大学 Separable ciphertext domain reversible data hiding method
CN112311954A (en) * 2020-10-29 2021-02-02 杭州电子科技大学 Ciphertext domain reversible information hiding method based on complementary code mark and bitmap embedding
CN114567710A (en) * 2021-12-03 2022-05-31 湖北工业大学 Reversible data steganography method and system based on ridge regression prediction

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005051793A (en) * 1995-05-08 2005-02-24 Digimarc Corp Steganographic system
JP2006222788A (en) * 2005-02-10 2006-08-24 Oki Electric Ind Co Ltd Image verification device, image embedding device, image detection device, image embedding method, image detection method, computer program, and printed matter manufacturing method
CN110913092A (en) * 2019-11-26 2020-03-24 西南交通大学 Reversible information hiding method for encrypted image
CN111898138A (en) * 2020-07-06 2020-11-06 西北工业大学 Separable ciphertext domain reversible data hiding method
CN112311954A (en) * 2020-10-29 2021-02-02 杭州电子科技大学 Ciphertext domain reversible information hiding method based on complementary code mark and bitmap embedding
CN114567710A (en) * 2021-12-03 2022-05-31 湖北工业大学 Reversible data steganography method and system based on ridge regression prediction

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
ZHAOXIA YIN: "Reversible Data Hiding in Encrypted Images Based on Multi-MSB Prediction and Huffman Coding", IEEE TRANSACTIONS ON MULTIMEDIA *
张秋余;冯玉春;: "自适应的密文彩色图像可逆数据隐藏算法", 计算机应用与软件, no. 02 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117082154A (en) * 2023-10-16 2023-11-17 长沙瑞腾信息技术有限公司 Big data-based double-path server data storage system
CN117082154B (en) * 2023-10-16 2023-12-15 长沙瑞腾信息技术有限公司 Big data-based double-path server data storage system

Also Published As

Publication number Publication date
CN115379066B (en) 2023-07-25

Similar Documents

Publication Publication Date Title
Qin et al. Separable reversible data hiding in encrypted images via adaptive embedding strategy with block selection
Cao et al. High capacity reversible data hiding in encrypted images by patch-level sparse representation
CN112311954B (en) Ciphertext domain reversible information hiding method based on complementary code mark and bitmap embedding
CN107578452B (en) JPEG image encryption method with compatible format and invariable size
CN113194213A (en) PNG image information hiding and recovering method based on secret sharing and chaotic mapping
Wang et al. A novel high-performance reversible data hiding scheme using SMVQ and improved locally adaptive coding method
CN111898138B (en) Separable ciphertext domain reversible data hiding method
CN111277831A (en) Image compression method and system based on classified coding
CN114745475B (en) Robust reversible information hiding method for encrypted image
CN115379066A (en) Encrypted image reversible data encryption and decryption method based on self-adaptive compression coding
Chang et al. A difference expansion based reversible information hiding scheme with high stego image visual quality
CN108305298B (en) Image recoverable information hiding method based on Sudoku matrix index
Li et al. Attack and improvement of the joint fingerprinting and decryption method for vector quantization images
CN115190216B (en) Encrypted image reversible data hiding and decrypting method based on prediction error map
Chang et al. A novel data-hiding and compression scheme based on block classification of SMVQ indices
CN115297218B (en) Reversible data hiding method based on Huffman coding rule and position diagram compression
CN113382127B (en) Encrypted image reversible information hiding method with elastic selection and run-length coding
Wang et al. A novel reversible data hiding scheme for VQ codebooks
CN114374774A (en) Separable encryption domain reversible information hiding method based on nearest neighbor pixel prediction
CN115348361B (en) Encryption image reversible data encryption and decryption method based on multi-MSB block coding
CN115103081B (en) Encrypted image reversible data hiding method based on mixed prediction and Huffman coding
Vaish et al. A systematic review on various reversible data hiding techniques in digital images
Ji et al. Reversible Data Hiding in Encrypted Images with Adaptive Huffman Code based on Dynamic Prediction Axes
Wu et al. Image reversible visual transformation based on MSB replacement and histogram bin mapping
Chidirala et al. Run length encoding based reversible data hiding scheme in encrypted images

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant