CN103985081A - Digital image multiple scrambling method based on lifting wavelet transformation - Google Patents
Digital image multiple scrambling method based on lifting wavelet transformation Download PDFInfo
- Publication number
- CN103985081A CN103985081A CN201410231666.8A CN201410231666A CN103985081A CN 103985081 A CN103985081 A CN 103985081A CN 201410231666 A CN201410231666 A CN 201410231666A CN 103985081 A CN103985081 A CN 103985081A
- Authority
- CN
- China
- Prior art keywords
- image
- scramble
- component
- key
- scrambling
- 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
Links
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression Of Band Width Or Redundancy In Fax (AREA)
- Facsimile Transmission Control (AREA)
Abstract
The invention discloses a digital image multiple scrambling method based on lifting wavelet transformation and belongs to the field of digital image process and information security. The method includes the two parts of content: forward scrambling of an image and reverse scrambling of the image. In the algorithm, a designated lifting wavelet function is utilized, and the image is scrambled through multiple secrete key methods such as image blocking, random integer matrix disturbance and transporsition of low-frequency components and high-frequency diagonal components. Experimental data show that the algorithm is high in security and university and has the advantages of being simple and high in scrambling speed, the high image scrambling degree and good scrambling effect can be obtained only through less scrambling, and the recovered image after reverse scrambling is lossless.
Description
Technical field
The present invention relates to the multiple disorder method of a kind of image based on lifting wavelet transform, is a kind of new image encryption means and Information hiding preprocess method specifically, belongs to digital image processing field and information security field.
Background technology
Along with the web development of digital multimedia technology, the safety of digital image information and privacy problem are subject to the attention of country and the whole society.It is the focus of digital image processing field and information security field research that digital image information is encrypted, and Method for Digital Image Scrambling is one of important method in digital image information encryption technology.
Chaotic Technology of Digital Images both can be used as a kind of image encryption method, can be used as again the preprocess method of Information hiding.Digital Image Scrambling is through certain algorithmic transformation by digital picture, becoming another width changed beyond recognition does not have the chaotic image of clear meaning, its essence is exactly the enciphering hiding of image, with solving security and the privacy concerns of digital image information in Internet Transmission and storing process.
At present common Method for Digital Image Scrambling belongs to Space domain mostly, comprises Arnold conversion, Fibonacci conversion, affined transformation, Hilbert curve, the scramble of sampling, magic scrambling, knight cruise scramble, Life of Game's scramble, chaos sequence scramble etc.Spatial domain disorder method is a lot, but there is no definite method rule, no matter is that the location of pixels of digital picture is carried out to scramble, or pixel value is carried out to scramble, all in the defect existing aspect certain to a certain extent.Such as the most frequently used Arnold conversion and sample technique scramble, be all that the other side's system of battle formations looks like to carry out scramble, and the not change of the histogram of scramble result, poor in versatility and security.Arnold conversion also has obvious periodicity, and, after several times iterated transform, matrix is got back to original state, and its mapping algorithm is fixed, and reach satisfied scramble effect need be through iteration repeatedly.The problems such as that other similar algorithms also exist is large such as calculated amount, calculation of complex, periodicity, in the scramble effect that is difficult to reach desirable aspect the comprehensive evaluations such as scramble speed, scramble efficiency and scramble degree.
In fact, the scramble to digital picture, can carry out in spatial domain, also can on frequency domain, carry out.From the angle of digital image encryption, frequency domain encryption is encrypted safer than spatial domain, and can be compatible with international JPEG compression standard, has more superiority in performance.Common frequency domain scrambling method mainly contains Fourier transform (DFT), discrete cosine transform (DCT), Fourier conversion and wavelet transformation (DWT) etc., but not yet finds the image scrambling method based on lifting wavelet transform.
Lifting wavelet transform belongs to a kind of Second Generation Wavelet Transformation that does not rely on Fourier transform, have simple in structure, calculated amount is little, the direct feature such as upset of inverse transformation, in digital watermark technology, has at present indivedual application.The realization of traditional wavelet completes by convolution, its calculation of complex, arithmetic speed is slow, memory demand is large, and the coefficient producing is floating number, be subject to the restriction of computing machine limited wordlength and accurately reconstruct original signal after traditional wavelet.And Lifting Wavelet is by a wavelet, progressively construct a better small echo of character (basic meaning that Here it is promotes), it is not only Upgrade Problem simply, also have simple in structure, operand is little, former bit arithmetic, save memory headroom, the inverse transformation features such as realization of can directly overturning, both the advantage that had kept Traditional Wavelet, overcome again the limitation of Traditional Wavelet, be just in time applicable to the demand of Chaotic Technology of Digital Images.
Therefore consider, the Method for Digital Image Scrambling of utilizing lifting wavelet transform to realize, not only can carry out quick scramble to digital picture, can also give security to the versatility of image scrambling effect, integrality and security.
Summary of the invention
The object of the invention is to propose the multiple disorder method of a kind of digital picture based on lifting wavelet transform.This algorithm has adopted the innovative design method of the multiple keys such as image block, the disturbance of random integers matrix, low frequency component and high frequency diagonal component transposition, not only solve the safety problem of recovering such as periodically, also met the robustness requirement of digital image encryption and Information hiding.
The object of the invention is to be achieved through the following technical solutions: the multiple disorder method of a kind of digital picture based on lifting wavelet transform, is characterized in that: comprise disorderly two processes of positive scramble and inverted;
Described positive scramble process implementation step is as follows:
The initial pictures Image of scramble is treated in input, and reading its image size is M * N, sets scramble number of times key cycle ∈ [1, max (M, N)], divide block size key block (block ∈ [1, min (M, N)]) and selected wavelet function Praenomen key wname;
(1) loop iteration starts, cycling condition k=1:cycle;
(2) successively each piecemeal of image is reduced the operation of a times, and definite four component blocks coordinate positions, then by the Lifting Wavelet scheme of setting, convert, to form 4 component image arrays [CA, CH, CV, CD];
(3) set up random perturbation matrix key R, and R and low frequency component CA are superposeed: CA=CA+R;
(4) by low frequency component CA and high frequency diagonal component CD transposition after stack, then by inverse transformation, revert to the image fig of M * N, an iteration of positive scramble finishes;
(5) when k is not equal to cycle, i.e. circulation while not finishing, goes to step (2) and continues operation, until iteration k time, obtains image fig after scramble, and just scramble process finishes.
The random process implementation step of described inverted is as follows:
Image fig after input scramble, sets and scramble number of times key cycle, minute block size key block and wavelet function Praenomen key wname identical in positive scramble process;
(1) loop iteration starts, cycling condition k=1:cycle;
(2) successively each piecemeal of image is reduced the operation of a times, and definite four component blocks coordinate positions, then by the Lifting Wavelet scheme of setting, convert, to form 4 component image arrays [CA, CH, CV, CD];
(3) set up random perturbation matrix key R, and carry out subtraction operation: CA=CA-R with low frequency component CA;
(4) by low frequency component CA, i.e. former CD, with high frequency diagonal component CD, i.e. former CA, transposition, then carry out the inverse transformation of Lifting Wavelet, and result is kept in FIG, and iteration of algorithm finishes;
(5) when k is not equal to cycle, i.e. circulation while not finishing, goes to step (2) and continues operation, until iteration k time, the random Recovery image FIG of the inverted obtaining, inverted unrest process finishes.
Described Lifting Wavelet function and wavelet function Praenomen wname, its value can be to comprise other applicable wavelet function names such as " db3 ", " db6 ", " sym4 ", " sym8 ", " coif ".
Described low frequency component CA and high frequency diagonal component CD transposition, comprise the exchange of the optional position of component image array.
Beneficial effect of the present invention: the scramble image visual effect that uses the inventive method to obtain is good, image after scramble is more as white noise, and the image after inverted disorderly recovers is compared with original image without any pixel error (feature because of lifting wavelet transform with integer transform).
Accompanying drawing explanation
Fig. 1 (a)～Fig. 1 (c) has shown based on lifting wavelet transform disorder method scramble 256 * 512lena image effect figure.Wherein Fig. 1 (a) is original image, and Fig. 1 (b) is 16 image effect figure of scramble, and Fig. 1 (c) is scramble Recovery image figure.
Fig. 2 (a)～Fig. 2 (c) has shown based on lifting wavelet transform disorder method scramble 512 * 512lena image effect figure.Wherein Fig. 2 (a) is original image figure, and Fig. 2 (b) is 3 image effect figure of scramble, and Fig. 2 (c) is scramble Recovery image figure.
Fig. 3 (a)～Fig. 3 (c) has shown Arnold disorder method 512 * 512lena image effect figure.Wherein Fig. 3 (a) is original image figure, and Fig. 3 (b) is 1 image effect of scramble.Fig. 3 (c) is 135 image effect figure of scramble.
Fig. 4 (a)～Fig. 4 (c) has shown sampling disorder method scramble 512 * 512lena image effect figure.Wherein Fig. 4 (a) is original image figure.Fig. 4 (b) is 3 image effect figure of scramble.Fig. 4 (c) is 77 image effect figure of scramble.
Fig. 5 is the continuous scramble degree evaluation of the gray-scale value of lifting wavelet transform scramble curve map.
Fig. 6 is the continuous scramble degree evaluation of the gray-scale value of Arnold disorder method curve map.
Fig. 7 is the continuous scramble degree evaluation of the gray-scale value curve map of sampling disorder method.
Fig. 8 is the wavelet field Local standard deviation scramble degree evaluation curve map of lifting wavelet transform scramble.
Fig. 9 is the wavelet field Local standard deviation scramble degree evaluation curve map of Arnold disorder method.
Figure 10 is the wavelet field Local standard deviation scramble degree evaluation curve map of sampling disorder method.
Embodiment
Design philosophy of the present invention is as follows:
One, establish image size for M * N (M, N can be unequal, and the method is directly applied for rectangular image).In this algorithm, design quadruple scramble key, comprise cycle index cycle ∈ [1, max (M, N)], minute block size block ∈ [1, min (M, N)], wavelet function Praenomen wname ∈ [1,37], as ' db6' or ' sym4' etc., and random perturbation matrix R ∈ [1, max (M, N)].Take 512 * 512 images as example, have the selection of 211 kinds of keys of C41537 ≈, certainly, also comprise wherein applying in a flexible way of several cipher key combinations of selection, or adopt the scheme of Multi-stage lifting wavelet transformation, can greatly improve scramble security.
Two, the position of the low frequency component CA position after lifting wavelet transform and high frequency diagonal line component CD is put upside down and low frequency component CA is added to random integers disturbance.Because the major part that the low frequency component CA after lifting wavelet transform has comprised eye recognition image, high frequency diagonal line component CD is the most insensitive detail section for human eye.So low frequency component CA adds that random integers disturbance has made the major part of image obtain scramble, then through with high frequency diagonal line component CD reversed position after image after inverse transformation, its scramble degree is stronger.Certainly, can also also take reversed position or add the scheme of random integers disturbance other high fdrequency component, will further strengthen the difficulty of decoding like this.
Three, this disorder method is done respectively lifting wavelet transform to each piecemeal, not only be equivalent to former figure to do the repeatedly consideration of this aspect of lifting wavelet transform, and because the piece number of piecemeal changes, thereby the size that makes each component in each lifting wavelet transform is made corresponding change, CA and CD out of position and to the disturbance of low frequency component CA additional random integer in addition, thus reach the object of further increasing data degree at random.Certainly, minute block size for rectangular image needs to be divided exactly by rectangular dimension.
Now to implementation method of the present invention, be elaborated as follows:
First, the key step of positive scramble process is as follows:
The initial pictures Image of scramble is treated in input, reading its image size is M * N, set scramble number of times key cycle ∈ [1, max (M, N)], divide block size key block=128 (block ∈ [1, min (M, N)]) and selected wavelet function Praenomen key wname=' db6 ';
(1) loop iteration starts, cycling condition k=1:cycle;
(2) set basic lifting step els={'p', [0.125-k*0.0010.125+k*0.001], 0}, and set polynomial interpolator lifting scheme lsnewInt=addlift (liftwave (wname, ' int2int'), els);
(3) successively each piecemeal of image is reduced the operation of a times, and determine four component blocks coordinate positions, then by lsnewInt lifting scheme, convert, to form 4 component image arrays: [CA (X, Y), CH (X, Y), CV (X, Y), CD (X, Y)]=lwt2 (double (Image (x, y)), lsnewInt);
(4) set up random perturbation matrix key R=randperm (size (CA, 1) * size (CA, 2)), and R and low frequency component CA are superposeed: CA=CA+R;
(5) by low frequency component CA and high frequency diagonal component CD transposition after stack, by inverse transformation, revert to image fig (x, y)=ilwt2 (CD (X, Y) of M * N, CH (X, Y), CV (X, Y), CA (X, Y), lsnewInt), an iteration of positive scramble finishes;
When k is not equal to cycle (i.e. circulation do not finish), go to step (2) and continue operation, until iteration k time, image fig after the scramble obtaining, just scramble process finishes.Second portion, the key step of the random process of inverted is as follows:
Image fig after input scramble, sets and scramble number of times key cycle, minute block size key block and wavelet function Praenomen key wname identical in positive scramble process;
(1) loop iteration starts, cycling condition k=1:cycle;
(2) set basic lifting step els={'p', [0.125-k*0.0010.125+k*0.001], 0}, and set polynomial interpolator lifting scheme lsnewInt=addlift (liftwave (wname, ' int2int'), els);
(3) successively each piecemeal of image is reduced the operation of a times, and determine four component blocks coordinate positions, then by lsnewInt lifting scheme, convert, to form 4 component image arrays: [CA (X, Y), CH (X, Y), CV (X, Y), CD (X, Y)]=lwt2 (double (Image (x, y)), lsnewInt);
(4) set up random perturbation matrix key R=randperm (size (CA, 1) * size (CA, 2)), and R and low frequency component CA are carried out to subtraction operation: CA=CA-R;
(5) by low frequency component CA (being former CD) and high frequency diagonal component CD (being former CA) transposition, then carry out the inverse transformation of Lifting Wavelet, result is kept in FIG, and iteration of algorithm finishes;
(6) when k is not equal to cycle (i.e. circulation do not finish), go to step (2) and continue operation, until iteration k time, the scramble Recovery image FIG obtaining, inverted unrest process finishes.
Below in conjunction with accompanying drawing to the present invention
Accompanying drawing 1 has provided the scramble design sketch of this algorithm to 256 * 512lena image, to rectangular image only 16 scrambles can obtain satisfied image degree at random and good scramble effect.
Accompanying drawing 2 has provided the scramble design sketch of this algorithm to 512 * 512lena image, and the other side's system of battle formations only 3 scrambles can obtain satisfied image degree at random and good scramble effect.
Accompanying drawing 3 has provided Arnold disorder method 512 * 512lena image effect figure, and comparatively speaking, the visual effect of the method after 135 scrambles still has obvious texture.
Accompanying drawing 4 has provided sampling disorder method 512 * 512lena image effect figure, and comparatively speaking, the visual effect of the method after 77 scrambles also has obvious texture.
Visible, disorder method of the present invention, is all being far superior to traditional disorder method aspect scramble speed, scramble degree and scramble effect, especially conventional Arnold disorder method.
Accompanying drawing 5, accompanying drawing 6 and accompanying drawing 7 are the curve maps that respectively lifting wavelet transform disorder method of the present invention, Arnold disorder method and sampling disorder method carried out to the evaluation of scramble degree according to continuum image scrambling degree evaluation method (formula 1).
Contrast known: the curve map (accompanying drawing 7) that the curve map (accompanying drawing 6) that use Arnold algorithm iteration is 256 times and sampling algorithm iteration are 256 times, scramble degree has significantly variation between 0.2～0.8, the scramble degree that Arnold and sampling algorithm are described has obvious periodicity, and security is poor; And with the curve map (accompanying drawing 5) of algorithm scramble of the present invention 256 times, scramble degree remains unchanged substantially, illustrates that this algorithm is more stable, and can reach fast the effect of scramble.
In formula (1), F'
m * nthe number in continuity region in image array after expression scramble, F
m * nthe number that represents continuum property in original image matrix, the size that m * n is image.
According to wavelet field Local standard deviation image scrambling degree evaluation method (formula 2), respectively above-mentioned three kinds of algorithms are carried out to the evaluation of scramble degree, wherein with the scramble of the Arnold algorithm scramble of line (accompanying drawing 9) and the sampling algorithm line (accompanying drawing 10) of writing music of writing music, all can see the increase along with scramble number of times, scramble degree has significantly variation between 0～0.8, have obvious periodicity, security is poor; And the fluctuation to some extent in 5 times that line (accompanying drawing 8) only starts at scramble of writing music of the scramble of algorithm of the present invention, later scramble degree remains unchanged substantially, close to desired level 1.
In formula (2), lsd is the Local standard deviation of image wavelet coefficient before scramble, and LSD is the Local standard deviation of image wavelet coefficient after scramble, and u is normalized factor.
Above-mentioned two kinds of different evaluation methods, draw the same conclusions of mutual confirmation, and this algorithm can obtain gratifying result aspect scramble degree and scramble effect as seen.
Again according to the scramble degree evaluation method shown in formula (1), guarantee that scramble degree is stabilized in 0.8 situation, above-mentioned three kinds of disorder methods are carried out to scramble speed contrast experiment, correlation data is as shown in table 1.
In table 1, embodied scramble speed (second) correlation data of the lower three kinds of disorder methods of scramble degree (0.8).
Table 1:
As can be seen from Table 1, the multiple disorder method of lifting wavelet transform of the present invention is obviously being better than traditional Arnold algorithm and sampling algorithm aspect scramble speed, especially the large Capacity Plan of scramble as time performance advantage more obvious.
In sum, the multiple disorder method of the digital picture based on lifting wavelet transform proposing in the present invention, at quadruple key, guarantee under the prerequisite of security, only pass through the scramble of less number of times, can obtain satisfied scramble degree and scramble effect, there is very strong security and versatility advantage, the feature such as especially have that algorithm is simple, scramble speed fast, scramble degree is high and scramble is effective.
Claims (3)
1. the multiple disorder method of the digital picture based on lifting wavelet transform, is characterized in that: comprise disorderly two processes of positive scramble and inverted;
Described positive scramble process implementation step is as follows:
The initial pictures Image of scramble is treated in input, and reading its image size is M * N, sets scramble number of times key cycle ∈ [1, max (M, N)], divide block size key block (block ∈ [1, min (M, N)]) and selected wavelet function Praenomen key wname;
(1) loop iteration starts, cycling condition k=1:cycle;
(2) successively each piecemeal of image is reduced the operation of a times, and definite four component blocks coordinate positions, then by the Lifting Wavelet scheme of setting, convert, to form 4 component image arrays [CA, CH, CV, CD];
(3) set up random perturbation matrix key R, and R and low frequency component CA are superposeed: CA=CA+R;
(4) by low frequency component CA and high frequency diagonal component CD transposition after stack, then by inverse transformation, revert to the image fig of M * N, an iteration of positive scramble finishes;
(5) when k is not equal to cycle, i.e. circulation while not finishing, goes to step (2) and continues operation, until iteration k time, obtains image fig after scramble, and just scramble process finishes.
The random process implementation step of described inverted is as follows:
Image fig after input scramble, sets and scramble number of times key cycle, minute block size key block and wavelet function Praenomen key wname identical in positive scramble process;
(1) loop iteration starts, cycling condition k=1:cycle;
(2) successively each piecemeal of image is reduced the operation of a times, and definite four component blocks coordinate positions, then by the Lifting Wavelet scheme of setting, convert, to form 4 component image arrays [CA, CH, CV, CD];
(3) set up random perturbation matrix key R, and carry out subtraction operation: CA=CA-R with low frequency component CA;
(4) by low frequency component CA, i.e. former CD, with high frequency diagonal component CD, i.e. former CA, transposition, then carry out the inverse transformation of Lifting Wavelet, and result is kept in FIG, and iteration of algorithm finishes;
(5) when k is not equal to cycle, i.e. circulation while not finishing, goes to step (2) and continues operation, until iteration k time, the random Recovery image FIG of the inverted obtaining, inverted unrest process finishes.
2. the multiple disorder method of a kind of digital picture based on lifting wavelet transform according to claim 1, it is characterized in that: described Lifting Wavelet function and wavelet function Praenomen wname, its value can be to comprise " db3 ", " db6 ", " sym4 ", " sym8 ", " coif " other applicable wavelet function name.
3. the multiple disorder method of a kind of digital picture based on lifting wavelet transform according to claim 1, is characterized in that: described low frequency component CA and high frequency diagonal component CD transposition, comprise the exchange of the optional position of component image array.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410231666.8A CN103985081B (en) | 2014-05-27 | 2014-05-27 | A kind of multiple disorder method of digital picture based on lifting wavelet transform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410231666.8A CN103985081B (en) | 2014-05-27 | 2014-05-27 | A kind of multiple disorder method of digital picture based on lifting wavelet transform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103985081A true CN103985081A (en) | 2014-08-13 |
| CN103985081B CN103985081B (en) | 2017-07-28 |
Family
ID=51277039
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410231666.8A Active CN103985081B (en) | 2014-05-27 | 2014-05-27 | A kind of multiple disorder method of digital picture based on lifting wavelet transform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103985081B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105281894A (en) * | 2015-11-25 | 2016-01-27 | 深圳供电局有限公司 | Plaintext encryption method and system based on seven-order magic cube |
| CN109859092A (en) * | 2019-01-21 | 2019-06-07 | 深圳前海微众银行股份有限公司 | Information concealing method, device, equipment and computer readable storage medium |
| CN111079573A (en) * | 2019-11-29 | 2020-04-28 | 童勤业 | Anti-counterfeiting encryption method based on image random scrambling technology |
-
2014
- 2014-05-27 CN CN201410231666.8A patent/CN103985081B/en active Active
Non-Patent Citations (2)
| Title |
|---|
| HE PAN-LI ETAL.: "A Method of Audio Digital Watermark Based on Discrete Wavelet Transform and Quantization Index Modulation", 《THE INTERNATIONAL CONFERENCE ON RECENT TRENDS IN MATERIALS AND MECHANICAL ENGINEERING (ICRTMME 2011)》 * |
| 范铁生 等: "小波域局部标准差的图像置乱评价方法", <小型微型计算机系统> * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105281894A (en) * | 2015-11-25 | 2016-01-27 | 深圳供电局有限公司 | Plaintext encryption method and system based on seven-order magic cube |
| CN105281894B (en) * | 2015-11-25 | 2018-10-23 | 深圳供电局有限公司 | Plaintext encryption method and system based on seven-order magic cube |
| CN109859092A (en) * | 2019-01-21 | 2019-06-07 | 深圳前海微众银行股份有限公司 | Information concealing method, device, equipment and computer readable storage medium |
| CN109859092B (en) * | 2019-01-21 | 2023-05-12 | 深圳前海微众银行股份有限公司 | Information hiding method, device, equipment and computer readable storage medium |
| CN111079573A (en) * | 2019-11-29 | 2020-04-28 | 童勤业 | Anti-counterfeiting encryption method based on image random scrambling technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103985081B (en) | 2017-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103023633B (en) | Digital image hiding method based on chaotic random phase and coherence stack principle | |
| WO2015103932A1 (en) | Hypersphere-based multivariable public key signature/verification system and method | |
| Xin et al. | 3D variable-structure chaotic system and its application in color image encryption with new Rubik’s Cube-like permutation | |
| Lv et al. | A novel scheme for simultaneous image compression and encryption based on wavelet packet transform and multi-chaotic systems | |
| Mhala et al. | A secure visual secret sharing (VSS) scheme with CNN-based image enhancement for underwater images | |
| CN103778593A (en) | Quantum image encryption method based on image association dissection | |
| CN114493991A (en) | Attention circulation countermeasure network-based style migration system, method and device | |
| CN103985081A (en) | Digital image multiple scrambling method based on lifting wavelet transformation | |
| Qiu et al. | Rich-information reversible watermarking scheme of vector maps | |
| Zhang et al. | Multiple-image encryption algorithm based on Sarrus rule and 3D Fibonacci matrix | |
| Ahuja et al. | Bifold-crypto-chaotic steganography for visual data security | |
| Rahim et al. | Exploiting de-noising convolutional neural networks DnCNNs for an efficient watermarking scheme: a case for information retrieval | |
| Jia et al. | Dual-complementary convolution network for remote-sensing image denoising | |
| Huang et al. | Atrous pyramid transformer with spectral convolution for image inpainting | |
| CN104143174A (en) | Image disguising and reconstructing method based on Tangram algorithm and two-dimensional double-scale rectangular mapping | |
| CN106408621A (en) | Image compression encryption and decryption method based on Lyapunov exponents | |
| Shi et al. | A Blind Watermarking Technique for Color Image based on SVD with Circulation. | |
| Wang et al. | Color image encryption based on discrete trinion Fourier transform and compressive sensing | |
| Manikandan et al. | Let wavelet authenticate and tent-map encrypt: a sacred connect against a secret nexus | |
| Yang et al. | Research on multiple-image encryption scheme based on joint power spectral division multiplexing and ghost imaging | |
| Qian et al. | Constructive steganography using texture synthesis | |
| CN104835107A (en) | Digital image disguising and reconstructing method based on block sequences | |
| CN110569945A (en) | QR code encryption method based on single-intensity recording phase recovery process | |
| Wang et al. | A novel (n, n) secret image sharing scheme based on Sudoku | |
| Chai et al. | Exploiting 2D-SDMCHM and matching embedding driven by flag-shaped hexagon prediction for visually meaningful medical image cryptosystem |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |