CN107948461A - More images hiding in QR codes is realized based on compressed sensing and orthogonal modulation - Google Patents
More images hiding in QR codes is realized based on compressed sensing and orthogonal modulation Download PDFInfo
- Publication number
- CN107948461A CN107948461A CN201711256205.6A CN201711256205A CN107948461A CN 107948461 A CN107948461 A CN 107948461A CN 201711256205 A CN201711256205 A CN 201711256205A CN 107948461 A CN107948461 A CN 107948461A
- Authority
- CN
- China
- Prior art keywords
- image
- orthogonal
- data
- codes
- measurement
- 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.)
- Pending
Links
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
-
- 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
Abstract
The invention discloses a kind of method for realizing that based on compressed sensing and orthogonal modulation more images hide in QR codes, belong to optical imagery and hide and field of encryption.The present invention mainly utilizes compressed sensing and orthogonal modulation technique, obtains the overall measurement value that multiple image is represented with decimal number, and preserved in the form of QR codes；Decoding when, barcode scanning is carried out to QR codes with photoelectric devices such as mobile phones, regains overall measurement value, then realize as needed entire image piecemeal reconstruction or multiple image in partial reconstitution.From the point of view of optical encryption angle, compressed sensing of the invention and quadrature modulation process can also be considered as two security levels, it is necessary to be correctly only possible to obtain reconstruction image completely in two levels, thus have good encryption performance.
Description
Technical field
Hidden the invention belongs to optical imagery and field of encryption, specifically a kind of Method of Steganography based on QR codes.
Background technology
With the development of information age, the disposal ability and security of information become more and more important, how to handle sea
It is the problem of modern society needs concern to measure data and how to prevent that data from being obtained by unauthorized user, thus various
Compress technique and optical encryption technology have obtained extensive concern.QR codes, i.e. quick response matrix code (Quick
Response code), it is a kind of matrix two-dimensional barcode that Japanese Denso companies research and develop in September, 1994, extensively
The field such as the processing applied to real time data and optical encryption generally.Have benefited from the widely available of smart mobile phone, QR codes are being made
Make, sell, logistics, storage, the field such as transport and consumer products gradually have more wide application prospect.But, although it
With powerful fault-tolerant and error correcting capability, but still suffer from some limitations and challenge.According to the coding standard of QR codes, it is only
Code conversion (i.e. figure pattern, alphanumeric pattern, byte mode and chinese character) can be carried out to the data of four types,
And image information is usually not suitable for being converted directly into QR codes since data volume is too big, only those extremely simple, data volumes are very
Small image is possible to.Therefore, storage of the image information in QR codes is realized, it is necessary first to reduce the data volume of image.
2006, Donoho proposed theory (the Donoho D L.Compressed sensing [J] of compressed sensing
.IEEE Transaction on Information Theory.2006,52(4):1289-1306.), with compressed sensing (or
Claim directly to perceive) mode eliminate the information after redundant data, realize the process compressed in collection.By one group with
The fractional-sample pixel of machine can reconstruct original image, that is to say, that can be recovered using less significant data original
Signal.Up to the present, the most important application of compressive sensing theory is the single pixel camera that rice university of the U.S. works out, profit
Line sampling measurement is carried out with digital micromirror array (DMD), measured value vector is obtained, then carries out Accurate Reconstruction.Fig. 1 is
The single pixel camera schematic diagram that rice university works out.The imaging system can be divided into three modules：Optical path modulation module, number
According to acquisition module, image reconstruction module.The system belongs to passive-type imaging system, having some limitations property, its reconstructed image
Resolution ratio is set by calculation matrix, but since pattern of rows and columns of DMD is limited, can only carry out limited resolution
The reconstruction of rate image.
In order to increase the flexibility of image reconstruction resolution ratio, and the needs of active imaging system are adapted to, can adopted
The random measurement picture of generation is projected to the mode on object with liquid crystal projection apparatus, to replace the function of above-mentioned DMD, is realized
The single pixel imaging of dynamic lighting.A series of pictures that will be generated at random according to certain probability distribution, are thrown successively by projecting apparatus
Penetrate on testee, then carry out data receiver and sampling with single pixel photodetector, here it is single image is carried out
Compress the process of collection.
The content of the invention
The present invention is directed to multiple image, carries out orthogonal modulation respectively by the sampled value to multiple image, is then added
To total measurement Value Data, the method for being converted to QR codes is re-encoded, realizes multiple image hiding in QR codes.The present invention exists
The compressed sensing of single image, collection have following innovation on the basis of rebuilding：First, although the data capacity of QR codes does not increase
Add, but due to the application of compressed sensing, QR codes is concealed more information with same capacity, that is, add and be stored in QR
Information content in code, condition is created to convert the image into QR codes；Secondly, which can meet wanting for active imaging system
Ask, by setting the size of random measurement matrix, can easily change the resolution ratio of reconstructed image, so as to fulfill more points
The reconstruct of resolution image；Further, since the application of orthogonal modulation principle, it can be achieved that in the multiple image of record only reconstruction portion
Partial image, or required partial information is only extracted in huge image, the flexibility of data reconstruction is increased, and drop
The low operand rebuild；Finally, which also has preferable data security, because from the point of view of angle is encrypted, the present invention
Compressed sensing and quadrature modulation process, two security levels can be considered to be, it is necessary to correctly just may be used completely in two levels
Reconstruction image can be obtained；The potentiality of the invention that are widely used in life, such as unit or personal much information centralized management
Deng.
Fig. 2 is the structure principle chart of the present invention, its target is to realize to carry out larger entire image piecemeal storage and again
Build or the restoration and reconstruction wherein piece image from the multiple image of storage.By taking the reconstruction of entire image piecemeal as an example, its basic work
It is that original image is subjected to uniform piecemeal first as flow, and measures processing respectively, equivalent to the compression sense to multiple image
Primary data gathers；Then orthogonal basic matrix is multiplied by the sampled value of each block image respectively, and be added and obtain by positive intermodulation
Overall measurement value after system；It is again QR codes by overall measurement value code conversion with QR codes compiler.When needing to rebuild certain blocking information
When, barcode scanning is carried out to QR codes with common Quick Response Code optoelectronic scanning device (such as mobile phone) and retrieves overall measurement value, multiplied by with
The transposition of the orthogonal basic matrix of partial response to be restored, can extract the view data for needing to recover.
It is of the present invention based on compressed sensing and orthogonal modulation realize more images in QR codes hide method, including with
Lower step：
1) by the pending uniform piecemeal of entire image and be sequentially placed in the first lens rear, by by calculation matrix generate with
Machine picture is cast out after being sent into liquid crystal projection apparatus, and projection light field is converged on block image through the first lens；From block image
The light transmitted is collected by the second lens again, and is focused on silicon optical detector；The electric signal of explorer response is adopted through data
Storage obtains data, is then converted into digital quantity and preserves in a computer.Projecting apparatus often exports a width projection image, data
Collector records an electric signal amount corresponding with transmission total light intensity, can after projecting M times for each block image
The size for obtaining measured value y, M that M × 1 is tieed up is determined by the sample rate set；Adopted as long as a block image completes measurement
Sample, another block image then repeat the process, until all block images complete measurement gatherer process；
2) orthogonal modulation is carried out respectively to these measurement results, by orthogonal basic matrix with each block diagram as corresponding measurement result
It is multiplied, you can obtain the corresponding measurement data after orthogonal modulation；Then each block image measurement data is summed to obtain total
Measurement Value Data；It is finally QR codes by overall measurement Value Data code conversion, realizes more images hiding in Quick Response Code；
3) if it is desired to rebuilding the parts of images information in entire image, then retrieved always with photoelectric device barcode scanning decoding QR codes
Measurement Value Data, the measured value y collected is then multiplied by with the transposition of orthogonal basic matrix corresponding with the part, you can carry
The corresponding data of the parts of images are taken out, the parts of images is then recovered from the data extracted with algorithm for reconstructing again.
Wherein, above-mentioned steps 1) the specific implementation process is as follows：
Uniform piecemeal 1a) is carried out to entire image as required, if multiple image, then all picture sizes are pressed into the maximum
It is extended to by way of mending zero value pixels consistent.As illustrated in fig. 2, it is assumed that entire image is uniformly divided into m blocks (X1,X2,...,
Xm), they are also considered as the different independent image of m width, are sequentially placed in order after the first lens and carry out each measurement step
Suddenly.If necessary to recover to the parts of images of reconstruction according to the resolution ratio of N=n × n, then the sampling of each block image
Number is (i.e. the quantity of measured value y) M=N × sample rate %.According to compressed sensing principle, the measured value of each block image
It is represented by y=ΦM×NXN×1, wherein Φ is calculation matrix generate at random, that size is M × N, and image X also need to through
The sparse matrix progress for crossing proper transformation is sparse.Understood according to limited isometry (RIP) property and accurate reconstruction principle (ERP),
Useful information can be extracted from compression sampling, it is necessary to meet calculation matrix and the uncorrelated condition of sparse matrix.
1b) since entire image has been uniformly divided into m blocks, the random measurement that corresponding m size of generation is M × N is also required to
Matrix (Φ1,Φ2,...,Φm) (1), a block image is corresponded to respectively.Generated with every a line of each random measurement matrix
One width random pictures, are so generated after one group of random pictures is sent into liquid crystal projection apparatus (2) by m calculation matrix and cast out, thrown
Light field is penetrated to converge on block image (4) through the first lens (3)；The light transmitted from block image (4) is again by the second lens
(5) collect and focus on silicon optical detector (6)；The electric signal of explorer response is through data collector (7) gathered data, so
After be converted to digital quantity and preserve in a computer.Projecting apparatus often exports a width projection image, data collector record one with
Transmit the corresponding electric signal amount of total light intensity.For each block image, the measurement that M × 1 is tieed up is can obtain after projecting M times
Value y.1c) when a block image completes measurement sampling, another block image replaces it and repeats above procedure, until institute
There is block image to complete measurement sampling.It is not difficult to know, after all be measured, we can obtain one group of measurement result
(y1,y2,...,ym) (8), each measured value is the column vector for including M element.
The step 2) is implemented as follows：
Each block image measured value result 2a) obtained to the gatherer process Jing Guo compressed sensing carries out orthogonal modulation respectively,
The orthogonal basic matrix used in the present invention by Hadamard orthogonal matrixes Column vector groups into.Hadamard orthogonal matrixes are one
A square formation, its column vector are orthogonal.If the Hadamard orthogonal matrixes that one size of generation is mM × mM, and by it per M
Row propose one new orthogonal basic matrix of composition, then can generate and corresponding m orthogonal basic matrixs of m group measured values
(O1,O2,...,Om)(9).The orthogonal basic matrix O of mM × M element is so included by eachiWith each block image
Corresponding measurement result yiIt is multiplied, you can obtain the corresponding measurement data D after orthogonal modulationi, its mathematic(al) representation is：
Di=Oiyi=OiΦiXi,i∈[1,m]. (1)
2b) measurement Value Data of each block image after orthogonal modulation is summed, obtains total measurement Value Data D (10)：
2c) total measurement Value Data D contains the information of all block images.By more image informations in the present invention through overcompression
Perception and orthogonal modulation, obtain one group of metric measurement Value Data, then with QR code compilers, are advised according to the coding of QR codes
Then it is encoded to QR codes.
The step 3) is implemented as follows：
3a) according to step 2a) structure orthogonal basic matrix (O1,O2,...,Om) (9), it is by the row of Hadamard orthogonal matrixes
Vector forms, and is mutually orthogonal between each column vector, so meeting：
Wherein, OTIt is the transposition of orthogonal basic matrix O, E is a unit matrix, and c is a constant (mould of column vector).
3b) Fig. 3 is the schematic diagram that part block image is rebuild from the QR codes for store multiple image information.Utilize the light such as mobile phone
Electrical part carries out barcode scanning to QR codes, can decode and regain total measurement Value Data.Such as need to rebuild wherein any one width piecemeal
Image, then with the transposition of corresponding orthogonal basic matrixCorresponding jth width block image is extracted from overall measurement Value Data D
Measurement Value Data yj：
Therefore, can select to rebuild any one block image as needed, without first rebuilding entire image, this is just significantly
Reduce the calculation amount of redundancy.(4) formula is the measurement Value Data corresponding to the block image that the needs extracted are rebuild, then profit
With 3c) in algorithm for reconstructing realize image reconstruction recover.The partial information for only extracting needs is thereby realized, without
Algorithm reconstruction is carried out to all data.
3c) restructuring procedure of signal needs the conventional image reconstruction algorithm using compressive sensing theory, mainly there is greedy tracing algorithm
(OMP), convex this three classes of relaxed algorithm and combinational algorithm.And OMP algorithms are main flow algorithm, it is to utilize iteration thought, each
In iterative process select a locally optimal solution come progressively approach original signal：First with the principle of correlation, select
With the most matched atom of iteration surplus, the atom is then subjected to Gram-Schmidt orthogonalization process, then signal is projected to
By these orthogonalization atomic buildings spatially, so that obtaining signal has selected component and iteration surplus on atom at these,
Then repeat the decomposition that the process carries out surplus.
Advantages of the present invention can be summarized as follows：
1) present invention is projected the random measurement picture of generation on object with liquid crystal projection apparatus, is suitable for active dynamic and is shone
Bright single pixel imaging, it is possible to achieve the reconstruction of a variety of image in different resolution, equipment are simple；
2) present invention can reduce the treating capacity of data based on compressive sensing theory, offer convenience for post-processing so that in QR
Storage multiple image information becomes possible in code；
3) for the larger image of single width, the piecemeal for realizing image using quadrature modulation method is rebuild, and can not only be reduced superfluous
Remaining calculation amount, and total calculation amount also significantly reduces, in a particular application, can only recover required parts of images and nothing
Whole huge image need to be rebuild；
4) this scheme as optical encryption in application, there is two security levels of compressed sensing and orthogonal modulation, only when measurement square
Battle array and orthogonal basic matrix could correctly be rebuild when all correctly matching needed for image, thus encryption is good；Need to rebuild certain width figure
During picture, it is only necessary to carry out corresponding extraction, be independent of each other between each view data, efficiency greatly improves；
5) processing method of the invention can be divided into two parts：Rebuild on line under collection and line., can be into one according to this principle
Step develops the system that medical treatment and aviation etc. need to transmit mass data that can be applied to, and mitigates the pressure of data transfer.
Brief description of the drawings
Attached drawing 1 is the single pixel camera schematic diagram that rice university works out；
Attached drawing 2 is the method for the present invention coding stage systematic schematic diagram；
Attached drawing 3 is the method for the present invention decoding stage systematic schematic diagram；
Attached drawing 4 is the square objects that four pierced patterns are included in present example；
Attached drawing 5 has been illustrated the process that projector projects random pictures are transmitted through object and is equivalent to random pictures and is weighed
Build the product between image；
Attached drawing 6 is the QR codes for including four width hollow out image sampling values in present example；
Attached drawing 7 is the reconstruction experimental result picture of each block image in present example；
Attached drawing 8 is original image to be reconstructed in present example；
Attached drawing 9 is to rebuild some point using the orthogonal basic matrix of mistake or the calculation matrix of mistake respectively in present example
The design sketch of block diagram picture；
Shown by reference numeral in above-mentioned attached drawing is：
The random measurement picture of 1 calculation matrix generation, 2 liquid crystal projection apparatus, 3 first lens, 4 represent the object of block image, and 5 the
Two lens, 6 photodetectors, 7 data collectors, 8 measured values, 9 orthogonal basic matrixs, 10 overall measurement values, 11 orthogonal basic matrixs
Transposition.
Embodiment
The invention will be further described below in conjunction with the accompanying drawings, and the specific embodiment of the present invention is described.
The liquid crystal projection apparatus model EPSON CB-S03 used in the present embodiment, data collector model 7IDA3, light
For electric explorer using silicon photocell as probe, preposition amplification, active filter circuit are that the software such as self-control, system control is used
Python is worked out.Block image as shown in Figure 4 includes the square objects of four pierced patterns, and photodetector is positioned over
On the back focal plane of mirror, four pieces are carried out to the object, and collection and recovery, each block image carry out weight with the resolution ratio of 64*64 respectively
Build.
The overall measurement value of 4 width block images is obtained by compressed sensing and orthogonal modulation.The object of four hollow out patterns, quilt
After being placed sequentially in the first lens.The random pictures generated by calculation matrix are projected to the every of hollow out object by liquid crystal projection apparatus
On a block.Often project M times, recorded one group of measured value of one of block diagram picture, next block diagram picture repeats the process.Fig. 5
The process that projector projects random pictures are transmitted through object has been illustrated to be equivalent between random pictures and reconstructed image
Product.Therefore, each output voltage of photodiode amplifier circuit is random measurement picture and treats between reconstructed image
The expression of inner product.The hits M of each block image is by by wanting the resolution ratio of reconstruction image and sample rate to determine.In this example
In, it is that the original is divided into 4 pieces to be acquired and recover respectively, is rebuild per block diagram picture with resolution ratio 64*64, and with
30% sample rate is sampled, i.e., needs to carry out M=1024 accidental projection for each block diagram picture, that is to say, that, it is necessary to
The measured value that the Hadamard orthogonal matrixes O of one 4096 × 4096 obtains each block diagram picture after compressed sensing carries out
Orthogonal modulation, every 1024 row form a new orthogonal basic matrix, so as to generate and four block measured values corresponding four
A orthogonal basic matrix (O1,O2,O3,O4).Sum after being multiplied by the measured value of each image respectively using each orthogonal basic matrix, you can
One group of total measurement Value Data is obtained, is represented by：
Realize multiple image information hiding in QR codes.The electric signal of explorer response, first passes through amplifier and filtering
Device processing, is then fed into data collector and carries out A/D conversions, become digital signal and preserved, what is so finally obtained is total
Measured value is one group of decimal number, can be easily QR codes with QR code compiler code conversions, realize the hidden of multiple image
Hide.Fig. 6 is the QR codes for including four width hollow out image sampling values.
The restoration and reconstruction of each block image are realized in decoding.Barcode scanning decoding QR codes can retrieve total measurement Value Data, so
Orthogonal modulation principle is based on afterwards, uses the figure needed for the transposition from total measured value extracting data of corresponding orthogonal basic matrix
As data.If for example, recovering the first block diagram picture, useCorresponding first width block diagram is extracted from overall measurement Value Data D
The measurement Value Data y of picture1：
Then the image is reconstructed using orthogonal matching pursuit algorithm (OMP).The recovery of other each block images similarly, each piecemeal
The reconstruction experimental result of image is as shown in fig. 7, only could be accurate under the conditions of calculation matrix and orthogonal basic matrix are matched at the same time
Rebuild, illustrate that encryption is good.Certainly, the image of reconstruction also includes certain noise, this is inevitable for compressed sensing
, but can be improved to a certain extent by increasing the number of measurement or improving reconstruction resolution ratio.
Rebuild respectively with resolution ratio 32*32 as shown in figure 8, the image is divided into four pieces, Fig. 9 is illustrated with mistake
Orthogonal basic matrix extraction data or the design sketch that testee is rebuild with the calculation matrix of mistake.Wherein, Fig. 9 (a)-(d) is use
The image that the orthogonal basic matrix of mistake and correct calculation matrix are rebuild, Fig. 9 (e)-(h) is with the calculation matrix of mistake and just
The image that true orthogonal basic matrix is rebuild.As can be seen that from the perspective of encryption, the present invention has good encryption, only
Could accurate reconstruction image under the conditions of calculation matrix and orthogonal basic matrix are matched at the same time.
Claims (2)
1. a kind of realize multiple image method hiding in QR codes based on compressed sensing and orthogonal modulation technique, its feature exists
In：One group of decimal number is collected by multiple image is measured, and is converted to the preservation of QR codes form；In decoding, use
The photoelectric devices such as mobile phone carry out barcode scanning to QR codes, so that it may the overall measurement value of multiple image is reacquired, then can be right as needed
Which part image is individually rebuild；Without carrying out computing reconstruction to all data, operand greatly reduces；From optical encryption
From the point of view of angle, the compressed sensing and quadrature modulation process of this method can be considered to be two security levels, only when measurement square
Battle array and orthogonal basic matrix could correctly be rebuild when all correctly matching needed for image, thus have good encryption performance.
2. one kind according to claim 1 realizes that multiple image is hidden in QR codes based on compressed sensing and orthogonal modulation technique
The method of Tibetan, it is characterised in that comprising data below acquisition phase, QR coding preservation stages and data extraction phase of regeneration three
Step：
1) data acquisition phase：
Uniform piecemeal 1a) is carried out to entire image as required, if multiple image, then all picture sizes are pressed into the maximum
It is extended to by way of mending zero value pixels consistent；Assuming that entire image is uniformly divided into m blocks (X1,X2,...,Xm), they
It can be considered as the different independent image of m width, be sequentially placed in order after the first lens and carry out each measuring process；If desired
The parts of images of reconstruction is recovered according to the resolution ratio of N=n × n, then the sampling number of each block diagram picture is (i.e. measured value
The quantity of y) M=N × sample rate %；According to compressed sensing principle, the measured value of each block diagram picture is represented by y=ΦM× NXN×1, wherein Φ is calculation matrix generate at random, that size is M × N, and image X is also needed to by the sparse of proper transformation
Matrix carries out sparse；
M blocks 1b) have been uniformly divided into according to entire image, it is corresponding to generate the random measurement matrix (Φ that m size is M × N1,
Φ2,...,Φm) (1), a block image is corresponded to respectively；A width random pictures are generated with every a line of each calculation matrix, this
M calculation matrix of sample is generated after one group of random pictures is sent into liquid crystal projection apparatus (2) and cast out, and projection light field is through the first lens
(3) converge on block image (4)；The light transmitted from block image (4) is collected by the second lens (5) again, and is focused on
On silicon optical detector (6)；The electric signal of explorer response is then converted into digital quantity simultaneously through data collector (7) gathered data
Preserve in a computer；Projecting apparatus often exports a width projection image, and data collector record one is corresponding with transmission total light intensity
Electric signal amount, for each block image, can obtain the measured value y that M × 1 is tieed up after projecting M times；
After 1c) block image completes measurement sampling, another block image replaces it and then repeats above step, until institute
There is block image to complete measurement gatherer process；After all be measured, one group of measurement result (y is obtained1,y2,...,ym) (8),
Each measured value is the column vector for including M element；
2) QR encodes the preservation stage：
Each block image measured value result 2a) obtained to the gatherer process Jing Guo compressed sensing carries out orthogonal modulation respectively；
The orthogonal basic matrix used in the present invention is the Column vector groups by Hadamard orthogonal matrixes into Hadamard orthogonal matrixes are one
A square formation, its column vector are orthogonal；If the Hadamard orthogonal matrixes that one size of generation is mM × mM, and by it per M
Row propose one new orthogonal basic matrix of composition, then can generate and corresponding m orthogonal basic matrix (O of m group measured values1,
O2,...,Om)(9)；The orthogonal basic matrix O of mM × M element is so included by eachiIt is corresponding with each block image
Measurement result yiIt is multiplied, you can obtain the corresponding measurement data D after orthogonal modulationi, its mathematic(al) representation is：
Di=Oiyi=OiΦiXi,i∈[1,m]. (1)
2b) measurement Value Data of each block image after orthogonal modulation is summed, obtains total measurement Value Data D (10)：
2c) total measurement Value Data D contains the information of all block images；These overall measurement values are one group of decimal numbers, so
QR code compilers are utilized afterwards, and QR codes are encoded to according to QR coding rules；
3) data extraction phase of regeneration：
3a) according to step 2a) structure orthogonal basic matrix (O1,O2,...,Om) (9), be from Hadamard orthogonal matrixes row to
Amount forms, and is mutually orthogonal between its each column vector, so meeting：
Wherein, OTIt is the transposition of orthogonal basic matrix O, E is a unit matrix, and c is a constant (mould of column vector)；
3b) to storing the QR codes of multiple image information, barcode scanning decoding can be carried out by photoelectric devices such as mobile phones, regained
Total measurement Value Data D；Such as need to rebuild wherein any one width block image, then with the transposition of corresponding orthogonal basic matrixFrom
The measurement Value Data y of corresponding jth width block image is extracted in overall measurement Value Data Dj：
Then compressed sensing reconstruction algorithm is used again from measurement Value Data yjIn recover the parts of images.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711256205.6A CN107948461A (en) | 2017-12-04 | 2017-12-04 | More images hiding in QR codes is realized based on compressed sensing and orthogonal modulation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711256205.6A CN107948461A (en) | 2017-12-04 | 2017-12-04 | More images hiding in QR codes is realized based on compressed sensing and orthogonal modulation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107948461A true CN107948461A (en) | 2018-04-20 |
Family
ID=61948406
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711256205.6A Pending CN107948461A (en) | 2017-12-04 | 2017-12-04 | More images hiding in QR codes is realized based on compressed sensing and orthogonal modulation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107948461A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110489989A (en) * | 2019-08-27 | 2019-11-22 | 深圳市道通智能航空技术有限公司 | Data encrypting/de-encrypling method, its encryption/deciphering chip and its unmanned plane |
| CN110536041A (en) * | 2019-08-27 | 2019-12-03 | 天津大学 | Quick response coding and several color image optical encryption methods of CMYK color space |
| CN110942413A (en) * | 2019-12-03 | 2020-03-31 | 中国科学院大学 | Method and system for hiding and extracting large-capacity optical information |
| CN111582098A (en) * | 2020-04-28 | 2020-08-25 | 北京集创北方科技股份有限公司 | Fingerprint image encryption, identification and decryption methods, storage medium and terminal |
| CN113485081A (en) * | 2021-07-09 | 2021-10-08 | 北京航空航天大学 | Non-cascade optical scanning holographic multi-image parallel encryption method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1438603A (en) * | 2003-03-25 | 2003-08-27 | 西安理工大学 | 2-D bar-code automatic reading method based on general office apparatus |
| CN102054209A (en) * | 2009-10-27 | 2011-05-11 | 镇江精英软件科技有限公司 | Method for transmitting driver pre-entered information data by using two-dimensional bar codes |
| CN103530669A (en) * | 2013-11-08 | 2014-01-22 | 成都博格科技有限公司 | Data information two-dimensional code encoding method |
| CN106887026A (en) * | 2017-01-18 | 2017-06-23 | 四川大学 | Many compression of images and the method rebuild are realized based on compressed sensing and orthogonal modulation |
-
2017
- 2017-12-04 CN CN201711256205.6A patent/CN107948461A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1438603A (en) * | 2003-03-25 | 2003-08-27 | 西安理工大学 | 2-D bar-code automatic reading method based on general office apparatus |
| CN102054209A (en) * | 2009-10-27 | 2011-05-11 | 镇江精英软件科技有限公司 | Method for transmitting driver pre-entered information data by using two-dimensional bar codes |
| CN103530669A (en) * | 2013-11-08 | 2014-01-22 | 成都博格科技有限公司 | Data information two-dimensional code encoding method |
| CN106887026A (en) * | 2017-01-18 | 2017-06-23 | 四川大学 | Many compression of images and the method rebuild are realized based on compressed sensing and orthogonal modulation |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110489989A (en) * | 2019-08-27 | 2019-11-22 | 深圳市道通智能航空技术有限公司 | Data encrypting/de-encrypling method, its encryption/deciphering chip and its unmanned plane |
| CN110536041A (en) * | 2019-08-27 | 2019-12-03 | 天津大学 | Quick response coding and several color image optical encryption methods of CMYK color space |
| CN110536041B (en) * | 2019-08-27 | 2021-11-16 | 天津大学 | Quick response coding and CMYK color space multi-color image optical encryption method |
| CN110942413A (en) * | 2019-12-03 | 2020-03-31 | 中国科学院大学 | Method and system for hiding and extracting large-capacity optical information |
| CN110942413B (en) * | 2019-12-03 | 2022-04-29 | 中国科学院大学 | Method and system for hiding and extracting large-capacity optical information |
| CN111582098A (en) * | 2020-04-28 | 2020-08-25 | 北京集创北方科技股份有限公司 | Fingerprint image encryption, identification and decryption methods, storage medium and terminal |
| CN113485081A (en) * | 2021-07-09 | 2021-10-08 | 北京航空航天大学 | Non-cascade optical scanning holographic multi-image parallel encryption method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107948461A (en) | More images hiding in QR codes is realized based on compressed sensing and orthogonal modulation | |
| Wang et al. | Hyperreconnet: Joint coded aperture optimization and image reconstruction for compressive hyperspectral imaging | |
| Shi et al. | Hscnn+: Advanced cnn-based hyperspectral recovery from rgb images | |
| Shi et al. | Scalable convolutional neural network for image compressed sensing | |
| Arad et al. | Ntire 2022 spectral recovery challenge and data set | |
| Willett et al. | Sparsity and structure in hyperspectral imaging: Sensing, reconstruction, and target detection | |
| US8948540B2 (en) | Optimized orthonormal system and method for reducing dimensionality of hyperspectral images | |
| Yang et al. | Compressive hyperspectral imaging via sparse tensor and nonlinear compressed sensing | |
| Wang et al. | High-speed hyperspectral video acquisition by combining nyquist and compressive sampling | |
| Peng et al. | Residual pixel attention network for spectral reconstruction from RGB images | |
| Xu et al. | Compressed domain image classification using a dynamic-rate neural network | |
| CN106887026B (en) | The method for realizing more compression of images and reconstruction based on compressed sensing and orthogonal modulation | |
| Yang et al. | Image reconstruction via manifold constrained convolutional sparse coding for image sets | |
| Hsu et al. | DCSN: Deep compressed sensing network for efficient hyperspectral data transmission of miniaturized satellite | |
| Arad et al. | NTIRE 2022 spectral demosaicing challenge and data set | |
| CN107817641A (en) | Single pixel camera system based on multimode fibre | |
| Li et al. | State of the art and prospects of structured sensing matrices in compressed sensing | |
| CN107622515A (en) | The physical re-organization method of squeezed light field | |
| Jalali et al. | Compressive imaging via one-shot measurements | |
| Ince | Hyperspectral image denoising using group low-rank and spatial-spectral total variation | |
| Shin et al. | Efficient spatially-variant single-pixel imaging using block-based compressed sensing | |
| Wakin et al. | Compressive sensing fundamentals | |
| Fan et al. | Global sensing and measurements reuse for image compressed sensing | |
| Le Guludec et al. | Deep light field acquisition using learned coded mask distributions for color filter array sensors | |
| Diaz et al. | Multispectral filter array design by optimal sphere packing |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180420 |
|
| RJ01 | Rejection of invention patent application after publication |