CN114078073A - anti-LED (light emitting diode) reproduction copying method and system based on light field conversion - Google Patents
anti-LED (light emitting diode) reproduction copying method and system based on light field conversion Download PDFInfo
- Publication number
- CN114078073A CN114078073A CN202210062916.4A CN202210062916A CN114078073A CN 114078073 A CN114078073 A CN 114078073A CN 202210062916 A CN202210062916 A CN 202210062916A CN 114078073 A CN114078073 A CN 114078073A
- Authority
- CN
- China
- Prior art keywords
- map
- copy
- image
- embedding
- vector
- 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
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000006243 chemical reaction Methods 0.000 title claims abstract description 38
- 239000013598 vector Substances 0.000 claims abstract description 316
- 238000004364 calculation method Methods 0.000 claims abstract description 16
- 239000011159 matrix material Substances 0.000 claims description 71
- 238000012545 processing Methods 0.000 claims description 55
- 230000006870 function Effects 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 9
- 238000003491 array Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 6
- 238000001514 detection method Methods 0.000 abstract description 4
- 230000008859 change Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 230000008929 regeneration Effects 0.000 description 3
- 238000011069 regeneration method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000001413 cellular effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/0021—Image watermarking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/10—Protecting distributed programs or content, e.g. vending or licensing of copyrighted material ; Digital rights management [DRM]
- G06F21/106—Enforcing content protection by specific content processing
- G06F21/1063—Personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/10—Protecting distributed programs or content, e.g. vending or licensing of copyrighted material ; Digital rights management [DRM]
- G06F21/16—Program or content traceability, e.g. by watermarking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2201/00—General purpose image data processing
- G06T2201/005—Image watermarking
- G06T2201/0064—Image watermarking for copy protection or copy management, e.g. CGMS, copy only once, one-time copy
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2201/00—General purpose image data processing
- G06T2201/005—Image watermarking
- G06T2201/0065—Extraction of an embedded watermark; Reliable detection
Abstract
The invention provides an anti-LED (light emitting diode) reproduction copy method and system based on light field conversion.A Transformer encoder is used for encoding each copy map in a copy map set respectively to obtain corresponding copy map embedding vectors, and the set of each copy map embedding vector is used as a copy map embedding vector set; respectively encoding each native map in the native map set by using a Transformer encoder to obtain corresponding native map embedding vectors, and taking the set of the native map embedding vectors as a native map embedding vector set; the distinguishing array is obtained by calculating the copy map embedding vector set and the original map embedding vector set, whether the comparison original map embedding vector is consistent with the embedding vector of the image to be detected or not is judged by the distinguishing array, if yes, the copy image to be detected is judged to be a new original image, quick detection and identification of the copy image and the original image are achieved, and the beneficial effect of reducing the calculation time cost of image identification is achieved.
Description
Technical Field
The invention belongs to the technical field of information security and image processing, and particularly relates to an LED (light-emitting diode) reproduction copy resisting method and system based on light field conversion.
Background
In recent years, with rapid development of technologies such as mobile internet and 5G mobile communication and wide spread of high-performance smart terminals, multimedia applications represented by images and videos have been increased explosively. The wide application of multimedia generates huge economic benefit, but the copyright dispute accompanied with the huge concern, rampant piracy seriously damages the economic benefit of multimedia content providers, and hinders the further popularization and application of multimedia technology. Secure and efficient multimedia content protection schemes are becoming the most urgent technical requirements in current multimedia applications. According to the research report of the Chinese industry information network, the information security market in China keeps 20% of growth every year in the future, and 700 billion yuan of market scale is broken through in 2020. Wherein, the multimedia content security product can at least contribute 25% of the market share, which reaches about 125 billion yuan.
The information hiding technology provides copyright authentication and traceability of images on the premise of not influencing the visual effect of media by embedding digital copyright watermarks in a certain form in multimedia contents such as images, and is one of key technologies in the field of multimedia content security. The traditional information hiding technology, especially the digital watermarking technology, often emphasizes the robustness of copyright watermarks, so that watermark information still exists in the watermarks generated by shooting after secondary shooting. Pirates often exploit the robustness of watermarks to illicitly pirate media images by means of reproduction copying, i.e. the copyright owner has a media image of the copyright information, and then uses the copied image media, thereby gaining profit by impersonating the original media.
Disclosure of Invention
The present invention is directed to a method and system for LED-resistant reproduction copy based on light field conversion, which solves one or more of the problems of the prior art and provides at least one of the advantages.
The invention provides an anti-LED (light emitting diode) reproduction copy method and system based on light field conversion.A Transformer encoder is used for encoding each copy map in a copy map set respectively to obtain corresponding copy map embedding vectors, and the set of each copy map embedding vector is used as a copy map embedding vector set; respectively encoding each native map in the native map set by using a Transformer encoder to obtain corresponding native map embedding vectors, and taking the set of the native map embedding vectors as a native map embedding vector set; and calculating by using the copy map embedding vector set and the original map embedding vector set to obtain a distinguishing array, judging whether the comparison original map embedding vector is consistent with the embedding vector of the to-be-detected map by using the distinguishing array, and if so, judging that the to-be-detected copy image is a new original image.
To achieve the above object, according to an aspect of the present invention, there is provided an anti-LED reproduction copy method based on light field conversion, the method comprising the steps of:
s100, shooting a plurality of different images by using an image pickup device, and taking the plurality of different images as a raw image set;
s200, respectively displaying each image in the original image set on an LED display, respectively shooting each image in the original image set by using a camera of the mobile equipment to obtain corresponding copy images, and taking the set of the copy images as a copy image set;
s300, respectively encoding each copy map in the copy map set by using a Transformer encoder to obtain corresponding copy map embedding vectors, and taking the set of each copy map embedding vector as a copy map embedding vector set;
s400, respectively encoding each protogram in the protogram set by using a Transformer encoder to obtain corresponding protogram embedding vectors, and taking the set of the protogram embedding vectors as a protogram embedding vector set;
s500, calculating by using the copy map embedded vector set and the native map embedded vector set to obtain a distinguishing array;
s600, obtaining a corresponding original image embedding vector as a comparison original image embedding vector for the image newly added into the original image set by using the method in S400, obtaining a corresponding copy image embedding vector as an embedding vector for the to-be-detected image for the to-be-detected copy image by using the method in S400, judging whether the comparison original image embedding vector is consistent with the embedding vector for the to-be-detected image by using a distinguishing array, and if so, judging that the to-be-detected copy image is the original image of the image newly added into the original image set.
Further, in S100, a plurality of different images are captured using the imaging apparatus, and a method of setting the plurality of different images as a raw image set is:
the method comprises the steps of shooting a plurality of groups of different images of the same object under a natural light source by utilizing a camera device, taking the plurality of different images as a protograph set, wherein the protographs correspond to copy graphs one by one, and the camera device comprises a CCD camera and a CMOS sensor.
Further, in S300, a method for encoding each copy map in the copy map set by using a Transformer encoder to obtain a corresponding copy map embedding vector, and using the set of copy map embedding vectors as the copy map embedding vector set includes:
the Transformer is a Pre-Trained Image Processing Transformer IPT (see Chen H, Wang Y, Guo T, et al, Pre-transformed Image Processing Transformer [ J ] 2020.), the Encoder of the Transformer is an Encoder in an IPT model, the output of the Encoder of the Transformer is a tensor after an image matrix is input into the Encoder of the Transformer, the number of elements in the tensor output by the coder of the transform is recorded as k, the elements in the tensor output by the coder of the transform are converted into a k-dimensional array (namely, each element in the tensor output by the coder of the transform is taken as one dimension in the array), the copy map embedding vector is the k-dimensional array converted by the elements in the tensor output after the image matrix of the copy map is input into the coder of the transform, and the set of each copy map embedding vector is taken as the copy map embedding vector set.
Further, in S400, a method for encoding each of the protographs in the set of protographs by using a Transformer encoder to obtain a corresponding protograph embedding vector, and using the set of protograph embedding vectors as a set of protograph embedding vectors includes:
the primitive map embedding vectors are k-dimensional arrays converted from elements in a tensor output after an image matrix of the primitive map is input to the encoder of the Transformer, and a set of the primitive map embedding vectors is used as a primitive map embedding vector set.
Further, in S500, the method for calculating the distinguishing array by using the copy map embedding vector set and the native map embedding vector set includes:
recording the number of elements in the copy map embedding vector set as n, recording the serial numbers of the elements in the copy map embedding vector set as i, i belongs to [1, n ], recording the copy map embedding vector set as a set Copset, and recording the elements with the serial numbers of i in the copy map embedding vector set as the Copset (i);
the number of elements in the copy map embedded vector set is the same as that of elements in the native map embedded vector set, the number of elements in the native map embedded vector set is the same as n, the sequence numbers of the elements in the native map embedded vector set are the same as i, the native map embedded vector set is recorded as a set Orgset, and the element with the sequence number of i in the native map embedded vector set is recorded as Orgset (i);
the number of dimensions in the copy map embedding vector is k, the serial number of the dimensions in the copy map embedding vector is v, v belongs to [1, k ], and the numerical value of the dimension with the serial number of v in the copset (i) is copset (i) v;
the number of dimensions in the native map embedding vector is equal to k, the serial number of the dimensions in the native map embedding vector is equal to v, and the numerical value of the dimension with the serial number of v in Orgset (i) is Orgset (i) [ v ];
converting the copy map embedding vector set into a copy map embedding matrix: the copy map embedding matrix is a matrix with n rows and k columns, the serial number of the rows in the copy map embedding matrix is i, the ith row in the copy map embedding matrix is copset (i), the serial number of the columns in the copy map embedding matrix is v, and the element of the vth column in the ith row in the copy map embedding matrix is copset (i) v;
converting the set of protogram embedding vectors into a protogram embedding matrix: the native map embedded matrix is a matrix with n rows and k columns, the serial number of a row in the native map embedded matrix is i, the ith row in the native map embedded matrix is Orgset (i), the serial number of a column in the native map embedded matrix is v, and the element of the vth column in the ith row in the native map embedded matrix is Orgset (i) [ v ];
recording the copy map embedding matrix as Copmat, the element of the ith row and the vth column in the Copmat as Copmat (i, v), recording the native map embedding matrix as Orgmat, and the element of the ith row and the vth column in the Orgmat as Orgmat (i, v);
the specific steps of calculating to obtain the distinguishing array are as follows:
s501, defining a native map processing vector as a native map processing vector obtained by processing the native map embedding matrix, where the method for obtaining the native map processing vector includes: the number of dimensions in the native graph processing vector is equal to the number of rows in the native graph embedding matrix, the native graph processing vector is Orgvec, the number of dimensions in Orgvec is n, the number of dimensions in Orgvec is i, the value of the dimension with number i in Orgvec is Orgvec (i), and the calculation formula of Orgvec (i) is:
thus obtaining a native graph processing vector; wherein the function exp is an exponential function with a natural constant e as a base; the native map processing vector has the advantages that the inverse change of the pixel value of the native map embedding matrix does not generate overlarge change in numerical value along with the sharp change or reduction of the pixel value of the image;
s502, defining a copy map processing vector as a copy map processing vector obtained by processing the copy map embedding matrix, and the method for obtaining the copy map processing vector includes: the number of dimensions in the copy map processing vector is equal to the number of rows in the copy map embedding matrix, the copy map processing vector is denoted as Copvec, the number of dimensions in the Copvec is denoted as n, the serial number of the dimensions in the Copvec is denoted as i, the numerical value of the dimension with the serial number of i in the Copvec is Copvec (i), and the calculation formula of Copvec (i) is as follows:
thereby obtaining a copy map processing vector; the copy map processing vector achieves the beneficial effects of making the inverse change of pixel values of the copy map embedding matrix not generate too large change in numerical value with the sharp change or reduction of the copy map pixel values, and making the derivative of the pixel values thereof continuously derivable;
s503, defining a distinguishing array as a group of numerical values for measuring the distinguishing degree between the native map processing vector and the copy map processing vector, wherein the number of dimensions in the distinguishing array is n, the serial number of the dimensions in the distinguishing array is i, the distinguishing array is Distvec, the numerical value with the serial number of i in the distinguishing array is Distvec (i), and the calculation formula of Distvec (i) is as follows:
the function ln is a function for calculating a natural logarithm, and a distinguishing array is obtained.
Further, in S600, for the image newly added to the raw image set, the method in S300 is used to obtain a corresponding raw image embedding vector as a comparison raw image embedding vector, the method in S400 is used to obtain a corresponding copy image embedding vector as an embedding vector of the to-be-detected image, and a distinguishing array is used to determine whether the comparison raw image embedding vector is consistent with the embedding vector of the to-be-detected image, if so, the method for determining that the to-be-detected copy image is the original image of the image newly added to the raw image set is as follows:
recording the number of dimensions in the embedding vector of the control native map as Orgvec1, recording the embedding vector of the to-be-detected map as Copvec1, the number of dimensions in the embedding vector of the control native map as well as the number of dimensions in the embedding vector of the to-be-detected map as n, the sequence number of dimensions in the embedding vector of the control native map as well as the sequence number of dimensions in the embedding vector of the to-be-detected map as i, i belongs to [1, n ], the value of the dimension with sequence number i in Orgvec1 as Orgvec1(i), and the value of the dimension with sequence number i in Copvec1 as Copvec1 (i);
the distinguishing array is marked as Distvec, the dimension number in Distvec is n, the dimension sequence number in Distvec is i, and the dimension with sequence number in Distvec is marked as Distvec (i);
furthermore, the method for judging whether the reference protogram embedding vector is consistent with the embedding vector of the image to be detected by using the distinguishing array specifically comprises the following steps:
s601, setting a threshold value as r, and representing copy distribution between the reference native map embedding vector and the embedding vector of the image to be detected by using r, wherein the copy distribution refers to a corresponding distribution state between pixel values of the reference native map embedding vector and the embedding vector of the image to be detected, and a calculation formula of r is as follows:
thereby obtaining r;
s602, using r and the distinguishing array to judge whether the following constraint conditions are met:
if yes, the contrast native map embedding vector is inconsistent with the embedding vector of the to-be-detected map, and if not, the contrast native map embedding vector is consistent with the embedding vector of the to-be-detected map; the copy distribution has the advantages that the consistency of the trend of the dimension value between the embedding vector of the control original image and the embedding vector of the image to be detected is detected with high sensitivity, so that the consistency between the embedding vector of the control original image and the embedding vector of the image to be detected is detected;
therefore, whether the embedding vector of the comparison native image is consistent with the embedding vector of the image to be detected is judged, and if yes, the copied image to be detected is judged to be the original image of the image newly added into the native image set.
The invention also provides an anti-LED reproduction copy system based on light field conversion, which comprises: the system can be operated in computing equipment such as desktop computers, notebooks, palm computers, cloud data centers and the like, and the operable system can include, but is not limited to, a processor, a memory and a server cluster, and the processor executes the computer program to operate in the following units of the system:
the image pickup unit is used for shooting a plurality of different images by using the image pickup equipment, and taking the plurality of different images as a raw image set;
the copy map unit is used for respectively displaying each image in the original map set on the LED display, respectively shooting each image in the original map set by using a camera of the mobile equipment to obtain corresponding copy maps, and taking the set of the copy maps as a copy map set;
the copy map embedding unit is used for respectively coding each copy map in the copy map set by using a Transformer encoder to obtain corresponding copy map embedding vectors, and taking the set of each copy map embedding vector as a copy map embedding vector set;
the native map embedding unit is used for respectively coding each native map in the native map set by using a Transformer encoder to obtain corresponding native map embedding vectors, and taking the set of the native map embedding vectors as a native map embedding vector set;
the distinguishing array computing unit is used for computing a distinguishing array by using the copy map embedded vector set and the native map embedded vector set;
and the judging and detecting unit is used for obtaining a corresponding original image embedding vector as a comparison original image embedding vector for the image newly added into the original image set by using the method in the S400, obtaining a corresponding copy image embedding vector as an embedding vector of the image to be detected by using the copy image to be detected by using the method in the S400, judging whether the comparison original image embedding vector is consistent with the embedding vector of the image to be detected by using the distinguishing array, and if so, judging that the copy image to be detected is the original image of the image newly added into the original image set.
The invention has the beneficial effects that: the invention provides an anti-LED (light emitting diode) reproduction copy method and system based on light field conversion.A Transformer encoder is used for encoding each copy map in a copy map set respectively to obtain corresponding copy map embedding vectors, and the set of each copy map embedding vector is used as a copy map embedding vector set; respectively encoding each native map in the native map set by using a Transformer encoder to obtain corresponding native map embedding vectors, and taking the set of the native map embedding vectors as a native map embedding vector set; the distinguishing array is obtained by calculating the copy map embedding vector set and the original map embedding vector set, whether the comparison original map embedding vector is consistent with the embedding vector of the image to be detected or not is judged by the distinguishing array, if yes, the copy image to be detected is judged to be a new original image, quick detection and identification of the copy image and the original image are achieved, and the beneficial effect of reducing the calculation time cost of image identification is achieved.
Drawings
The above and other features of the present invention will become more apparent by describing in detail embodiments thereof with reference to the attached drawings in which like reference numerals designate the same or similar elements, it being apparent that the drawings in the following description are merely exemplary of the present invention and other drawings can be obtained by those skilled in the art without inventive effort, wherein:
FIG. 1 is a flow chart of a method for anti-LED reproduction copy based on light field conversion;
fig. 2 is a system configuration diagram of an anti-LED reproduction copy system based on light field conversion.
Detailed Description
The conception, the specific structure and the technical effects of the present invention will be clearly and completely described in conjunction with the embodiments and the accompanying drawings to fully understand the objects, the schemes and the effects of the present invention. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
In the description of the present invention, the meaning of a plurality of means is one or more, the meaning of a plurality of means is two or more, and larger, smaller, larger, etc. are understood as excluding the number, and larger, smaller, inner, etc. are understood as including the number. If the first and second are described for the purpose of distinguishing technical features, they are not to be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated or implicitly indicating the precedence of the technical features indicated.
Referring to fig. 1, a flow chart of an LED copy protection method based on light field conversion according to the present invention is shown, and a method and a system for LED copy protection based on light field conversion according to an embodiment of the present invention are described below with reference to fig. 1.
The invention provides an LED (light-emitting diode) reproduction copy resisting method based on light field conversion, which specifically comprises the following steps of:
s100, shooting a plurality of different images by using an image pickup device, and taking the plurality of different images as a raw image set;
s200, respectively displaying each image in the original image set on an LED display, respectively shooting each image in the original image set by using a camera of the mobile equipment to obtain corresponding copy images, and taking the set of the copy images as a copy image set;
s300, respectively encoding each copy map in the copy map set by using a Transformer encoder to obtain corresponding copy map embedding vectors, and taking the set of each copy map embedding vector as a copy map embedding vector set;
s400, respectively encoding each protogram in the protogram set by using a Transformer encoder to obtain corresponding protogram embedding vectors, and taking the set of the protogram embedding vectors as a protogram embedding vector set;
s500, calculating by using the copy map embedded vector set and the native map embedded vector set to obtain a distinguishing array;
s600, obtaining a corresponding original image embedding vector as a comparison original image embedding vector for the image newly added into the original image set by using the method in S400, obtaining a corresponding copy image embedding vector as an embedding vector for the to-be-detected image for the to-be-detected copy image by using the method in S400, judging whether the comparison original image embedding vector is consistent with the embedding vector for the to-be-detected image by using a distinguishing array, and if so, judging that the to-be-detected copy image is the original image of the image newly added into the original image set.
Further, in S100, a plurality of different images are captured using the imaging apparatus, and a method of setting the plurality of different images as a raw image set is:
the method comprises the steps of shooting a plurality of groups of different images of the same object under a natural light source by utilizing a camera, and taking the plurality of different images as a raw image set, wherein the raw images correspond to copy images one by one.
Further, in S300, a method for encoding each copy map in the copy map set by using a Transformer encoder to obtain a corresponding copy map embedding vector, and using the set of copy map embedding vectors as the copy map embedding vector set includes:
the Transformer is a Pre-Trained Image Processing Transformer IPT (see Chen H, Wang Y, Guo T, et al, Pre-transformed Image Processing Transformer [ J ] 2020.), the Encoder of the Transformer is an Encoder in an IPT model, the output of the Encoder of the Transformer is a tensor after an image matrix is input into the Encoder of the Transformer, the number of elements in the tensor output by the coder of the transform is recorded as k, the elements in the tensor output by the coder of the transform are converted into a k-dimensional array (namely, each element in the tensor output by the coder of the transform is taken as one dimension in the array), the copy map embedding vector is the k-dimensional array converted by the elements in the tensor output after the image matrix of the copy map is input into the coder of the transform, and the set of each copy map embedding vector is taken as the copy map embedding vector set.
Further, in S400, a method for encoding each of the protographs in the set of protographs by using a Transformer encoder to obtain a corresponding protograph embedding vector, and using the set of protograph embedding vectors as a set of protograph embedding vectors includes:
the primitive map embedding vectors are k-dimensional arrays converted from elements in a tensor output after an image matrix of the primitive map is input to the encoder of the Transformer, and a set of the primitive map embedding vectors is used as a primitive map embedding vector set.
Further, in S500, the method for calculating the distinguishing array by using the copy map embedding vector set and the native map embedding vector set includes:
recording the number of elements in the copy map embedding vector set as n, recording the serial numbers of the elements in the copy map embedding vector set as i, i belongs to [1, n ], recording the copy map embedding vector set as a set Copset, and recording the elements with the serial numbers of i in the copy map embedding vector set as the Copset (i);
the number of elements in the copy map embedded vector set is the same as that of elements in the native map embedded vector set, the number of elements in the native map embedded vector set is the same as n, the sequence numbers of the elements in the native map embedded vector set are the same as i, the native map embedded vector set is recorded as a set Orgset, and the element with the sequence number of i in the native map embedded vector set is recorded as Orgset (i);
the number of dimensions in the copy map embedding vector is k, the serial number of the dimensions in the copy map embedding vector is v, v belongs to [1, k ], and the numerical value of the dimension with the serial number of v in the copset (i) is copset (i) v;
the number of dimensions in the native map embedding vector is equal to k, the serial number of the dimensions in the native map embedding vector is equal to v, and the numerical value of the dimension with the serial number of v in Orgset (i) is Orgset (i) [ v ];
converting the copy map embedding vector set into a copy map embedding matrix: the copy map embedding matrix is a matrix with n rows and k columns, the serial number of the rows in the copy map embedding matrix is i, the ith row in the copy map embedding matrix is copset (i), the serial number of the columns in the copy map embedding matrix is v, and the element of the vth column in the ith row in the copy map embedding matrix is copset (i) v;
converting the set of protogram embedding vectors into a protogram embedding matrix: the native map embedded matrix is a matrix with n rows and k columns, the serial number of a row in the native map embedded matrix is i, the ith row in the native map embedded matrix is Orgset (i), the serial number of a column in the native map embedded matrix is v, and the element of the vth column in the ith row in the native map embedded matrix is Orgset (i) [ v ];
recording the copy map embedding matrix as Copmat, the element of the ith row and the vth column in the Copmat as Copmat (i, v), recording the native map embedding matrix as Orgmat, and the element of the ith row and the vth column in the Orgmat as Orgmat (i, v);
the specific steps of calculating to obtain the distinguishing array are as follows:
s501, defining a native map processing vector as a native map processing vector obtained by processing the native map embedding matrix, where the method for obtaining the native map processing vector includes: the number of dimensions in the native graph processing vector is equal to the number of rows in the native graph embedding matrix, the native graph processing vector is Orgvec, the number of dimensions in Orgvec is n, the number of dimensions in Orgvec is i, the value of the dimension with number i in Orgvec is Orgvec (i), and the calculation formula of Orgvec (i) is:
thus obtaining a native graph processing vector; wherein the function exp is an exponential function with a natural constant e as a base;
s502, defining a copy map processing vector as a copy map processing vector obtained by processing the copy map embedding matrix, and the method for obtaining the copy map processing vector includes: the number of dimensions in the copy map processing vector is equal to the number of rows in the copy map embedding matrix, the copy map processing vector is denoted as Copvec, the number of dimensions in the Copvec is denoted as n, the serial number of the dimensions in the Copvec is denoted as i, the numerical value of the dimension with the serial number of i in the Copvec is Copvec (i), and the calculation formula of Copvec (i) is as follows:
thereby obtaining a copy map processing vector;
s503, defining a distinguishing array as a group of numerical values for measuring the distinguishing degree between the native map processing vector and the copy map processing vector, wherein the number of dimensions in the distinguishing array is n, the serial number of the dimensions in the distinguishing array is i, the distinguishing array is Distvec, the numerical value with the serial number of i in the distinguishing array is Distvec (i), and the calculation formula of Distvec (i) is as follows:
the function ln is a function for calculating a natural logarithm, and a distinguishing array is obtained.
Further, in S600, for the image newly added to the raw image set, the method in S300 is used to obtain a corresponding raw image embedding vector as a comparison raw image embedding vector, the method in S400 is used to obtain a corresponding copy image embedding vector as an embedding vector of the to-be-detected image, and a distinguishing array is used to determine whether the comparison raw image embedding vector is consistent with the embedding vector of the to-be-detected image, if so, the method for determining that the to-be-detected copy image is the original image of the image newly added to the raw image set is as follows:
recording the number of dimensions in the embedding vector of the control native map as Orgvec1, recording the embedding vector of the to-be-detected map as Copvec1, the number of dimensions in the embedding vector of the control native map as well as the number of dimensions in the embedding vector of the to-be-detected map as n, the sequence number of dimensions in the embedding vector of the control native map as well as the sequence number of dimensions in the embedding vector of the to-be-detected map as i, i belongs to [1, n ], the value of the dimension with sequence number i in Orgvec1 as Orgvec1(i), and the value of the dimension with sequence number i in Copvec1 as Copvec1 (i);
the distinguishing array is marked as Distvec, the dimension number in Distvec is n, the dimension sequence number in Distvec is i, and the dimension with sequence number in Distvec is marked as Distvec (i);
furthermore, the method for judging whether the reference protogram embedding vector is consistent with the embedding vector of the image to be detected by using the distinguishing array specifically comprises the following steps:
s601, setting a threshold value as r, and representing copy distribution between the embedding vector of the control original map and the embedding vector of the to-be-detected map by using r, wherein the calculation formula of r is as follows:
thereby obtaining r;
s602, using r and the distinguishing array to judge whether the following constraint conditions are met:
if yes, the contrast native map embedding vector is inconsistent with the embedding vector of the to-be-detected map, and if not, the contrast native map embedding vector is consistent with the embedding vector of the to-be-detected map;
therefore, whether the contrast native image embedding vector is consistent with the embedding vector of the image to be detected or not is judged, if yes, the copy image to be detected is judged to be the original image of the image newly added into the native image set, the quick detection and identification of the copy image and the native image are achieved, and the beneficial effect of reducing the calculation time cost of image identification is achieved.
The anti-LED reproduction copying system based on light field conversion comprises: the processor executes the computer program to implement the steps in the above embodiment of the light field conversion-based anti-LED regeneration copying method, and the system for anti-LED regeneration copying based on light field conversion may be run in a desktop computer, a notebook, a palm computer, a cloud data center, and other computing devices, and the system that may be run may include, but is not limited to, a processor, a memory, and a server cluster.
An embodiment of the present invention provides an LED-resistant reproduction copy system based on light field conversion, as shown in fig. 2, the LED-resistant reproduction copy system based on light field conversion of the embodiment includes: a processor, a memory and a computer program stored in the memory and executable on the processor, the processor implementing the steps in one of the above-mentioned embodiments of the light field conversion based anti-LED reproduction copy method, when executing the computer program, the processor executing the computer program to run in the units of the following system:
the image pickup unit is used for shooting a plurality of different images by using the image pickup equipment, and taking the plurality of different images as a raw image set;
the copy map unit is used for respectively displaying each image in the original map set on the LED display, respectively shooting each image in the original map set by using a camera of the mobile equipment to obtain corresponding copy maps, and taking the set of the copy maps as a copy map set;
the copy map embedding unit is used for respectively coding each copy map in the copy map set by using a Transformer encoder to obtain corresponding copy map embedding vectors, and taking the set of each copy map embedding vector as a copy map embedding vector set;
the native map embedding unit is used for respectively coding each native map in the native map set by using a Transformer encoder to obtain corresponding native map embedding vectors, and taking the set of the native map embedding vectors as a native map embedding vector set;
the distinguishing array computing unit is used for computing a distinguishing array by using the copy map embedded vector set and the native map embedded vector set;
and the judging and detecting unit is used for obtaining a corresponding original image embedding vector as a comparison original image embedding vector for the image newly added into the original image set by using the method in the S400, obtaining a corresponding copy image embedding vector as an embedding vector of the image to be detected by using the copy image to be detected by using the method in the S400, judging whether the comparison original image embedding vector is consistent with the embedding vector of the image to be detected by using the distinguishing array, and if so, judging that the copy image to be detected is the original image of the image newly added into the original image set.
The anti-LED regeneration copying system based on light field conversion can be operated in computing equipment such as desktop computers, notebooks, palm computers and cloud data centers. The anti-LED reproduction copying system based on light field conversion comprises, but is not limited to, a processor and a memory. It will be understood by those skilled in the art that the example is only an example of the LED reproduction copy resisting method and system based on the light field conversion, and does not constitute a limitation to the LED reproduction copy resisting method and system based on the light field conversion, and may include more or less components than the light field conversion, or combine some components, or different components, for example, the LED reproduction copy resisting system based on the light field conversion may further include an input and output device, a network access device, a bus, and the like.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete component Gate or transistor logic, discrete hardware components, etc. The general purpose processor may be a microprocessor or the processor may be any conventional processor, etc., the processor is the control center of the LED anti-copy system based on light field conversion, and various interfaces and lines are used to connect the various sub-regions of the whole LED anti-copy system based on light field conversion.
The memory can be used for storing the computer program and/or the module, and the processor realizes various functions of the anti-LED reproduction copying method and system based on light field conversion by running or executing the computer program and/or the module stored in the memory and calling data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
The invention provides an anti-LED (light emitting diode) reproduction copy method and system based on light field conversion.A Transformer encoder is used for encoding each copy map in a copy map set respectively to obtain corresponding copy map embedding vectors, and the set of each copy map embedding vector is used as a copy map embedding vector set; respectively encoding each native map in the native map set by using a Transformer encoder to obtain corresponding native map embedding vectors, and taking the set of the native map embedding vectors as a native map embedding vector set; the distinguishing array is obtained by calculating the copy map embedding vector set and the original map embedding vector set, whether the comparison original map embedding vector is consistent with the embedding vector of the image to be detected or not is judged by the distinguishing array, if yes, the copy image to be detected is judged to be a new original image, quick detection and identification of the copy image and the original image are achieved, and the beneficial effect of reducing the calculation time cost of image identification is achieved.
Although the present invention has been described in considerable detail and with reference to certain illustrated embodiments, it is not intended to be limited to any such details or embodiments or any particular embodiment, so as to effectively encompass the intended scope of the invention. Furthermore, the foregoing describes the invention in terms of embodiments foreseen by the inventor for which an enabling description was available, notwithstanding that insubstantial modifications of the invention, not presently foreseen, may nonetheless represent equivalent modifications thereto.
Claims (7)
1. An anti-LED reproduction copying method based on light field conversion is characterized by comprising the following steps:
s100, shooting a plurality of different images by using an image pickup device, and taking the plurality of different images as a raw image set;
s200, respectively displaying each image in the original image set on an LED display, respectively shooting each image in the original image set by using a camera of the mobile equipment to obtain corresponding copy images, and taking the set of the copy images as a copy image set;
s300, respectively encoding each copy map in the copy map set by using a Transformer encoder to obtain corresponding copy map embedding vectors, and taking the set of each copy map embedding vector as a copy map embedding vector set;
s400, respectively encoding each protogram in the protogram set by using a Transformer encoder to obtain corresponding protogram embedding vectors, and taking the set of the protogram embedding vectors as a protogram embedding vector set;
s500, calculating by using the copy map embedded vector set and the native map embedded vector set to obtain a distinguishing array;
s600, obtaining a corresponding original image embedding vector as a comparison original image embedding vector for the image newly added into the original image set by using the method in S400, obtaining a corresponding copy image embedding vector as an embedding vector for the to-be-detected image for the to-be-detected copy image by using the method in S400, judging whether the comparison original image embedding vector is consistent with the embedding vector for the to-be-detected image by using a distinguishing array, and if so, judging that the to-be-detected copy image is the original image of the image newly added into the original image set.
2. The method for anti-LED reproduction copy based on light field conversion according to claim 1, wherein in S100, a plurality of different images are captured by using an image capturing device, and the method for using the plurality of different images as a raw image set is as follows:
the method comprises the steps of shooting a plurality of groups of different images of the same object under a natural light source by utilizing a camera, and taking the plurality of different images as a raw image set, wherein the raw images correspond to copy images one by one.
3. The method for anti-LED reproduction copy based on light field conversion according to claim 1, wherein in S300, the method for encoding each copy map in the set of copy maps by using a transform encoder to obtain a corresponding copy map embedding vector, and using the set of copy map embedding vectors as the set of copy map embedding vectors comprises:
the method comprises the steps that a Transformer is pre-trained image processing Transformer IPT for short, an Encoder of the Transformer is an Encoder in an IPT model, the output of the Encoder of the Transformer is a tensor after an image matrix is input into the Encoder of the Transformer, the number of elements in the tensor output by the Encoder of the Transformer is k, the elements in the tensor output by the Encoder of the Transformer are converted into a k-dimensional array, a copy map embedding vector is a k-dimensional array converted from the elements in the tensor output after the image matrix of a copy map is input into the Encoder of the Transformer, and a set of all copy map embedding vectors is used as a copy map embedding vector set.
4. The method of claim 3, wherein in step S400, a transform encoder is used to encode each of the raw graphs in the raw graph set to obtain a corresponding raw graph embedding vector, and the method for using the set of raw graph embedding vectors as the set of raw graph embedding vectors is:
the primitive map embedding vectors are k-dimensional arrays converted from elements in a tensor output after an image matrix of the primitive map is input to the encoder of the Transformer, and a set of the primitive map embedding vectors is used as a primitive map embedding vector set.
5. The method for anti-LED reproduction copy based on light field conversion according to claim 4, wherein in S500, the method for obtaining the distinguishing array by using the copy map embedding vector set and the raw map embedding vector set comprises:
recording the number of elements in the copy map embedding vector set as n, recording the serial numbers of the elements in the copy map embedding vector set as i, i belongs to [1, n ], recording the copy map embedding vector set as a set Copset, and recording the elements with the serial numbers of i in the copy map embedding vector set as the Copset (i);
the number of elements in the copy map embedded vector set is the same as that of elements in the native map embedded vector set, the number of elements in the native map embedded vector set is the same as n, the sequence numbers of the elements in the native map embedded vector set are the same as i, the native map embedded vector set is recorded as a set Orgset, and the element with the sequence number of i in the native map embedded vector set is recorded as Orgset (i);
the number of dimensions in the copy map embedding vector is k, the serial number of the dimensions in the copy map embedding vector is v, v belongs to [1, k ], and the numerical value of the dimension with the serial number of v in the copset (i) is copset (i) v;
the number of dimensions in the native map embedding vector is equal to k, the serial number of the dimensions in the native map embedding vector is equal to v, and the numerical value of the dimension with the serial number of v in Orgset (i) is Orgset (i) [ v ];
converting the copy map embedding vector set into a copy map embedding matrix: the copy map embedding matrix is a matrix with n rows and k columns, the serial number of the rows in the copy map embedding matrix is i, the ith row in the copy map embedding matrix is copset (i), the serial number of the columns in the copy map embedding matrix is v, and the element of the vth column in the ith row in the copy map embedding matrix is copset (i) v;
converting the set of protogram embedding vectors into a protogram embedding matrix: the native map embedded matrix is a matrix with n rows and k columns, the serial number of a row in the native map embedded matrix is i, the ith row in the native map embedded matrix is Orgset (i), the serial number of a column in the native map embedded matrix is v, and the element of the vth column in the ith row in the native map embedded matrix is Orgset (i) [ v ];
recording the copy map embedding matrix as Copmat, the element of the ith row and the vth column in the Copmat as Copmat (i, v), recording the native map embedding matrix as Orgmat, and the element of the ith row and the vth column in the Orgmat as Orgmat (i, v);
the specific steps of calculating to obtain the distinguishing array are as follows:
s501, defining a native graph processing vector as a vector obtained by processing the native graph embedding matrix, where the method for obtaining the native graph processing vector includes: the number of dimensions in the native graph processing vector is equal to the number of rows in the native graph embedding matrix, the native graph processing vector is Orgvec, the number of dimensions in Orgvec is n, the number of dimensions in Orgvec is i, the value of the dimension with number i in Orgvec is Orgvec (i), and the calculation formula of Orgvec (i) is:
thus obtaining a native graph processing vector; wherein the function exp is an exponential function with a natural constant e as a base;
s502, defining a copy map processing vector as a vector obtained by processing the copy map embedding matrix, wherein the method for obtaining the copy map processing vector comprises the following steps: the number of dimensions in the copy map processing vector is equal to the number of rows in the copy map embedding matrix, the copy map processing vector is denoted as Copvec, the number of dimensions in the Copvec is denoted as n, the serial number of the dimensions in the Copvec is denoted as i, the numerical value of the dimension with the serial number of i in the Copvec is Copvec (i), and the calculation formula of Copvec (i) is as follows:
thereby obtaining a copy map processing vector;
s503, defining a distinguishing array as a group of numerical values for measuring the distinguishing degree between the native map processing vector and the copy map processing vector, wherein the number of dimensions in the distinguishing array is n, the serial number of the dimensions in the distinguishing array is i, the distinguishing array is Distvec, the numerical value with the serial number of i in the distinguishing array is Distvec (i), and the calculation formula of Distvec (i) is as follows:
the function ln is a function for calculating a natural logarithm, and a distinguishing array is obtained.
6. The method of claim 4, wherein in step S600, the method in S300 is used for the image newly added to the primitive image set to obtain the corresponding primitive image embedding vector as the comparison primitive image embedding vector, the method in S400 is used for the copy image to be detected to obtain the corresponding copy image embedding vector as the embedding vector for the image to be detected, the distinguishing array is used to determine whether the comparison primitive image embedding vector is consistent with the embedding vector for the image to be detected, and if so, the method for determining that the copy image to be detected is the original image of the image newly added to the primitive image set is that:
recording the number of dimensions in the embedding vector of the control native map as Orgvec1, recording the embedding vector of the to-be-detected map as Copvec1, the number of dimensions in the embedding vector of the control native map as well as the number of dimensions in the embedding vector of the to-be-detected map as n, the sequence number of dimensions in the embedding vector of the control native map as well as the sequence number of dimensions in the embedding vector of the to-be-detected map as i, i belongs to [1, n ], the value of the dimension with sequence number i in Orgvec1 as Orgvec1(i), and the value of the dimension with sequence number i in Copvec1 as Copvec1 (i);
the distinguishing array is marked as Distvec, the dimension number in Distvec is n, the dimension sequence number in Distvec is i, and the dimension with sequence number in Distvec is marked as Distvec (i);
furthermore, the method for judging whether the reference protogram embedding vector is consistent with the embedding vector of the image to be detected by using the distinguishing array specifically comprises the following steps:
s601, setting a threshold value as r, and representing copy distribution between the embedding vector of the control original map and the embedding vector of the to-be-detected map by using r, wherein the calculation formula of r is as follows:
thereby obtaining r;
s602, using r and the distinguishing array to judge whether the following constraint conditions are met:
if yes, the contrast native map embedding vector is inconsistent with the embedding vector of the to-be-detected map, and if not, the contrast native map embedding vector is consistent with the embedding vector of the to-be-detected map;
therefore, whether the embedding vector of the comparison native image is consistent with the embedding vector of the image to be detected is judged, and if yes, the copied image to be detected is judged to be the original image of the image newly added into the native image set.
7. An anti-LED reproduction copy system based on light field conversion, characterized in that the anti-LED reproduction copy system based on light field conversion comprises: the system comprises a processor, a memory and a computer program stored in the memory and running on the processor, wherein the processor implements the steps in the light field conversion-based anti-LED reproduction copying method in claim 1 when executing the computer program, and the light field conversion-based anti-LED reproduction copying system runs in computing equipment of desktop computers, notebooks, palmtops and cloud data centers.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210062916.4A CN114078073B (en) | 2022-01-20 | 2022-01-20 | anti-LED (light emitting diode) reproduction copying method and system based on light field conversion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210062916.4A CN114078073B (en) | 2022-01-20 | 2022-01-20 | anti-LED (light emitting diode) reproduction copying method and system based on light field conversion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114078073A true CN114078073A (en) | 2022-02-22 |
| CN114078073B CN114078073B (en) | 2022-04-08 |
Family
ID=80284689
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210062916.4A Active CN114078073B (en) | 2022-01-20 | 2022-01-20 | anti-LED (light emitting diode) reproduction copying method and system based on light field conversion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114078073B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090116686A1 (en) * | 2007-10-05 | 2009-05-07 | Rajan Samtani | Content Serialization by Varying Content Properties, Including Varying Master Copy Watermark Properties |
| CN101460973A (en) * | 2006-06-06 | 2009-06-17 | 索尼英国有限公司 | Encoding and detecting apparatus |
| CN101504655A (en) * | 2009-03-06 | 2009-08-12 | 中山大学 | Color relationship characteristic based image approximate copy detection method |
| CN101719275A (en) * | 2009-11-23 | 2010-06-02 | 中国科学院计算技术研究所 | Image feature point extracting and realizing method, image copying and detecting method and system thereof |
| CN103092935A (en) * | 2013-01-08 | 2013-05-08 | 杭州电子科技大学 | Approximate copy image detection method based on scale invariant feature transform (SIFT) quantization |
| CN109543674A (en) * | 2018-10-19 | 2019-03-29 | 天津大学 | A kind of image copy detection method based on generation confrontation network |
| CN110378871A (en) * | 2019-06-06 | 2019-10-25 | 绍兴聚量数据技术有限公司 | Game charater original painting copy detection method based on posture feature |
| CN110892484A (en) * | 2018-07-11 | 2020-03-17 | 因美纳有限公司 | Deep learning-based framework for identifying sequence patterns causing sequence-specific errors (SSEs) |
| CN112527938A (en) * | 2020-12-17 | 2021-03-19 | 安徽迪科数金科技有限公司 | Chinese POI matching method based on natural language understanding |
-
2022
- 2022-01-20 CN CN202210062916.4A patent/CN114078073B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101460973A (en) * | 2006-06-06 | 2009-06-17 | 索尼英国有限公司 | Encoding and detecting apparatus |
| US20090116686A1 (en) * | 2007-10-05 | 2009-05-07 | Rajan Samtani | Content Serialization by Varying Content Properties, Including Varying Master Copy Watermark Properties |
| CN101504655A (en) * | 2009-03-06 | 2009-08-12 | 中山大学 | Color relationship characteristic based image approximate copy detection method |
| CN101719275A (en) * | 2009-11-23 | 2010-06-02 | 中国科学院计算技术研究所 | Image feature point extracting and realizing method, image copying and detecting method and system thereof |
| CN103092935A (en) * | 2013-01-08 | 2013-05-08 | 杭州电子科技大学 | Approximate copy image detection method based on scale invariant feature transform (SIFT) quantization |
| CN110892484A (en) * | 2018-07-11 | 2020-03-17 | 因美纳有限公司 | Deep learning-based framework for identifying sequence patterns causing sequence-specific errors (SSEs) |
| CN109543674A (en) * | 2018-10-19 | 2019-03-29 | 天津大学 | A kind of image copy detection method based on generation confrontation network |
| CN110378871A (en) * | 2019-06-06 | 2019-10-25 | 绍兴聚量数据技术有限公司 | Game charater original painting copy detection method based on posture feature |
| CN112527938A (en) * | 2020-12-17 | 2021-03-19 | 安徽迪科数金科技有限公司 | Chinese POI matching method based on natural language understanding |
Non-Patent Citations (1)
| Title |
|---|
| HANTING CHEN 等: "Pre-Trained Image Processing Transformer", 《PROCEEDINGS OF THE IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114078073B (en) | 2022-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11238556B2 (en) | Embedding signals in a raster image processor | |
| US9443277B2 (en) | Method for embedding and extracting multi-scale space based watermark | |
| US8855420B2 (en) | Descriptor determination in a multimedia content | |
| US8103054B2 (en) | Method for identifying marked images using statistical moments based at least in part on a JPEG array | |
| Hsu et al. | Image tamper detection and recovery using adaptive embedding rules | |
| Lei et al. | Video sequence matching based on the invariance of color correlation | |
| WO2009158691A2 (en) | Patch-based texture histogram coding for fast image similarity search | |
| US20130034337A1 (en) | Copy detection | |
| JP2011508323A (en) | Permanent visual scene and object recognition | |
| WO2013036086A2 (en) | Apparatus and method for robust low-complexity video fingerprinting | |
| WO2021187776A1 (en) | Methods and systems for grouping of media based on similarities between features of the media | |
| CN103325081B (en) | Insertion and extracting method based on spatial domain image digital watermark | |
| US20190311744A1 (en) | Comparing frame data to generate a textless version of a multimedia production | |
| Haviana et al. | Average hashing for perceptual image similarity in mobile phone application | |
| Liu et al. | Detection of JPEG double compression and identification of smartphone image source and post-capture manipulation | |
| Reed et al. | Closed form non-iterative watermark embedding | |
| CN114078073B (en) | anti-LED (light emitting diode) reproduction copying method and system based on light field conversion | |
| Darwish et al. | Improved color image watermarking using logistic maps and quaternion Legendre-Fourier moments | |
| US9264584B2 (en) | Video synchronization | |
| Mehta et al. | Near-duplicate detection for LCD screen acquired images using edge histogram descriptor | |
| Roopalakshmi et al. | Recent trends in content-based video copy detection | |
| Hashmi et al. | Real time copyright protection and implementation of image and video processing on android and embedded platforms | |
| US20230325961A1 (en) | Zoom agnostic watermark extraction | |
| Lo et al. | Probability-based image authentication scheme for indexed color images | |
| CN108921767B (en) | Image digital watermark processing system and processing method based on FPGA |
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 |