CN105844638A - Method and system for discriminating genuine or counterfeit photos through camera noise - Google Patents
Method and system for discriminating genuine or counterfeit photos through camera noise Download PDFInfo
- Publication number
- CN105844638A CN105844638A CN201610168248.8A CN201610168248A CN105844638A CN 105844638 A CN105844638 A CN 105844638A CN 201610168248 A CN201610168248 A CN 201610168248A CN 105844638 A CN105844638 A CN 105844638A
- Authority
- CN
- China
- Prior art keywords
- photo
- noise
- piecemeal
- camera
- correlation
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0006—Industrial image inspection using a design-rule based approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
Abstract
The invention relates to the technical field of electronics and especially relates to a method and system for discriminating genuine or counterfeit photos through camera noise. The system comprises a photo segmentation module, a photo noise information extraction module, a noise information comparison and matching module and a display which are successively connected; the photo segmentation module sends information of reference photos and tested photo divided blocks to the photo noise information extraction module; the photo noise information extraction module sends extracted photon response non-uniform noise data of each divided block of the reference photos and each corresponding divided block of the tested photos to the noise information comparison and matching module; the noise information comparison and matching module performs correlation coefficient calculation and transfers the matching results to the display. By means of the system, not only whether or not photos taken by certain types of cameras are counterfeit can be determined, but also the specific processed parts can be determined, which refines the traditional authentic identification method and provides possibility for restoring genuine photo from points to surfaces.
Description
Technical field
The present invention relates to electronic technology field, particularly relate to a kind of method and system realizing differentiating the photo true and false by camera noise.
Background technology
In the use of mobile phone and camera, digital picture is ubiquitous, and along with development and the appearance of numerous image processing software of digital image processing techniques, any domestic consumer possesses certain software and uses ability, can modify digital photos.When some amendments no longer keep image verity originally, the when of starting society is threatened, these are distorted the method that operation effectively detects and becomes more and more important.
Summary of the invention
It is an object of the invention to provide the user a kind of method and system by the camera noise identification photo true and false, be possible not only to detect the photo captured by a certain camera the most modified, and can determine the region that photo is modified.
For achieving the above object, the technical solution used in the present invention is:
A kind of method being realized by camera noise differentiating the photo true and false, is comprised the following steps:
S1. use with reference to camera installation shooting photo as with reference to photo, extract the uneven noise data of photo response with reference to photo, as device-fingerprint；
Extract the uneven noise data of photo response of test photo the most again, compare with device-fingerprint, and carry out the calculating of correlation coefficient with the noise data as device-fingerprint；And carry out matching judgment, and if the match is successful, showing that testing photo comes from reference to camera installation, photo was not revised；
S3. identification result is sent to display show.
Preferably, correlation coefficient described in step S2 is to reflect the statistical indicator of dependency relation level of intimate between variable, it is to calculate by product moment method, based on two variablees deviation with respective meansigma methods, it is multiplied by two deviations and reflects degree of correlation between two variablees, utilize correlation peak energy than calculating degree of association.
Preferably, the judgement principle that described in step S2, the match is successful is: according to correlation detection algorithm return value, whether differential test photo is modified, and when matching judgment operates, compares correlation peak energy and i.e. can be determined whether coupling than with threshold size；If correlation coefficient is more than threshold value, then the match is successful.
It is further preferred that described correlation detection algorithm refers to carry out uneven for photo response noise as device-fingerprint the algorithm of image source discriminating.
For realizing the method realizing the discriminating photo true and false above by camera noise, by the following technical solutions:
A kind of system being realized the discriminating photo true and false by camera noise, including the picture noise information extraction modules being sequentially connected with, noise information comparison and matching module and display；Described picture noise information extraction modules passes to described noise information comparison and matching module by extracting the uneven noise data of photo response with reference to photo and test photo；Described noise information comparison carries out Calculation of correlation factor with matching module and mates, and is resent to display.
For realizing can determine the region that photo is modified, have employed techniques below scheme:
A kind of method being realized by camera noise differentiating the photographic region true and false, is comprised the following steps:
(1) use with reference to camera installation shooting photo, as with reference to photo, the grid (n is positive integer) of n*n, n will be divided into reference to photo2Individual an equal amount of piecemeal, is also divided into the grid of n*n, n simultaneously by test photo2Individual an equal amount of piecemeal；
(2) extract with reference to n in photo2The uneven noise data of the photo response of each piecemeal in individual piecemeal, as noise template；
(3) n in test photo is extracted2The uneven noise data of the photo response of each piecemeal in individual piecemeal, compares one by one with each piecemeal of noise template correspondence position, and carries out the calculating of correlation coefficient as the noise data of noise template with correspondence position；Carrying out matching judgment one by one, if the match is successful, the piecemeal of this position of differential test photo comes from reference to camera installation；
(4) identification result is sent to display and shows.
Preferably, step (3) calculates n in test photo2In individual piecemeal, the uneven noise data of the photo response of each piecemeal and correspondence position are as the correlation coefficient of the noise data of noise template, described correlation coefficient is to reflect the statistical indicator of dependency relation level of intimate between variable, it is to calculate by product moment method, based on two variablees deviation with respective meansigma methods, it is multiplied by two deviations and reflects degree of correlation between two variablees, utilize correlation peak energy than calculating degree of association.
Preferably, the resolution principle that described in step (3), the match is successful is: the uneven noise data of photo response of the piecemeal of the test a certain position of photo calculates with the uneven noise data of photo response of the piecemeal of correspondence position in noise template, according to correlation detection algorithm return value, whether the piecemeal of this position of differential test photo was forged, when matching judgment operates, compare correlation peak energy than i.e. can be determined whether coupling with threshold size；If correlation coefficient is more than threshold value, then the match is successful.
It is further preferred that described correlation detection algorithm refers to carry out uneven for photo response noise as device-fingerprint the algorithm of image source discriminating.
For realizing realizing, by camera noise, the method that differentiates the photographic region true and false, have employed techniques below scheme:
A kind of system being realized the discriminating photographic region true and false by camera noise, including the picture segmentation module being sequentially connected with, picture noise information extraction modules, noise information comparison and matching module and display；The information dividing piecemeal with reference to photo and test photo is passed to described picture noise information extraction modules by described picture segmentation module；Described picture noise information extraction modules, by each piecemeal of reference photo extracted respectively and the uneven noise data of photo response of test photo relevant position piecemeal, passes to described noise information comparison and matching module；Described noise information comparison and matching module will carry out Calculation of correlation factor and the result mated sends display to.
Different from the method that tradition differentiates image true-false, the uneven noise of photo response (Photo Response Non-Uniformity Noise, hereinafter referred to as PRNU) it is a kind of multiplicative noise with change in signal strength, it is that the defect in the inhomogeneities by silicon chip and sensor production process causes, even if the PRNU noise that the camera of same model produces also can differ.The PRNU noise of photo and photo content are unrelated, are only determined by camera apparatus itself, and therefore we can utilize PRNU noise to differentiate the camera photos true and false.By extracting test photo and the PRNU noise with reference to photo, according to correlation detection algorithm return value, it is determined that whether test photo was forged, if correlation peak energy ratio (the Peak to Correlation
Energyratio, calls PCE in the following text) value then tests photo more than threshold value and carrys out self-reference camera installation, i.e. and test photo was not forged, if PCE value is less than threshold value, tests photo and was forged.Wherein correlation detection algorithm refers to PRNU noise as the algorithm carrying out image source discriminating for device-fingerprint.
By photo subregion being carried out detection and the matching primitives of PRNU noise, identifying the region that test photo is modified, feedback information being sent to display and shows.Image processing techniques presents with forms such as image block effect, distortion and statistical properties, and its visual impact provides certification clue.The device class certification finally drawn according to these clues determines model or the production firm producing vision facilities.
The invention has the beneficial effects as follows: different from the method that tradition differentiates image true-false, utilize the characteristic that PRNU noise is retained because being difficult to eliminate, total steady noise that there is residual on photo, it as " identity card " of photo, extract the reference model noise of camera again, comparing with photo, the traditional method that compares accuracy rate obtains the biggest lifting.Image is carried out piecemeal, each piecemeal is carried out feature extraction, carry out similarity analysis and consistency detection, simultaneously positioning tampering region and determine the means of distorting.Can not only judge that whether the photo captured by the camera of certain model is through forging, it is also possible to judging treated concrete part, tradition false distinguishing mode refined, the true figure that for spreading over a whole area from one point reduces provides possibility.
Accompanying drawing explanation
Fig. 1 is the system block diagram of the embodiment of the present invention 1；
Fig. 2 is the system block diagram of the embodiment of the present invention 2；
Fig. 3 is one embodiment of the present of invention noise matching schematic diagram.
Detailed description of the invention
Below in conjunction with the accompanying drawings embodiments of the present invention are described in detail.
Whether embodiment 1, photograph are forged
PRNU is a kind of multiplicative noise with change in signal strength, and it is that the defect in the inhomogeneities by silicon chip and sensor production process causes, even if the PRNU noise that the camera of same model produces also can differ.The PRNU noise of photo and photo content are unrelated, are only determined by camera itself, and therefore we can utilize PRNU noise to differentiate the camera photos true and false.Utilize the characteristic that PRNU noise is retained because being difficult to eliminate, total steady noise that there is residual on digital photograph, it as " identity card " of photo, then extracts the reference model noise of camera, comparing with photo, the traditional method that compares accuracy rate obtains the biggest lifting.
Specific embodiment technical scheme is as follows: a kind of method being realized by camera noise differentiating the photo true and false, comprises the following steps:
S1. use with reference to camera installation shooting photo as with reference to photo, extract the uneven noise data of photo response with reference to photo, as device-fingerprint；
Extract the uneven noise data of photo response of test photo the most again, compare with device-fingerprint, and carry out the calculating of correlation coefficient with the noise data as device-fingerprint；And carry out matching judgment, and if the match is successful, showing that testing photo comes from reference to camera installation, photo was not revised；
S3. identification result is sent to display show.
Correlation coefficient described in step S2 is to reflect the statistical indicator of dependency relation level of intimate between variable, it is to calculate by product moment method, based on two variablees deviation with respective meansigma methods, it is multiplied by two deviations and reflects degree of correlation between two variablees, utilize correlation peak energy than calculating degree of association.
The judgement principle that described in step S2, the match is successful is: according to correlation detection algorithm return value, whether differential test photo is modified, and when matching judgment operates, compares correlation peak energy and i.e. can be determined whether coupling than with threshold size；If correlation coefficient is more than threshold value, then the match is successful.
Described correlation detection algorithm refers to carry out uneven for photo response noise as device-fingerprint the algorithm of image source discriminating.
More than differentiate that the correlation coefficient described in the method for the photo true and false is in order to reflect the statistical indicator of dependency relation level of intimate between variable, correlation coefficient is to calculate by product moment method, based on two variablees deviation with respective meansigma methods, it is multiplied by two deviations and reflects degree of correlation between two variablees；Correlation coefficient herein reflects the degree of correlation between the uneven noise data of photo response and the KI noise data of photo, the most relevant, have many Important Relations.
As it is shown in figure 1, judge whether photo is forged by the above method differentiating the photo true and false, it is achieved through the following technical solutions:
A kind of system being realized the discriminating photo true and false by camera noise, including the picture noise information extraction modules 2 being sequentially connected with, noise information comparison and matching module 1 and display 3；Described picture noise information extraction modules 2 passes to described noise information comparison and matching module 1 by extracting the uneven noise data of photo response with reference to photo and test photo；Described noise information comparison carries out Calculation of correlation factor with matching module 1 and mates, and is resent to display 3.
Picture noise information extraction modules 2 is in order to extract the PRNU noise template with reference to photo and test photo PRNU noise.
Noise information comparison and matching module 1 are in order to according to correlation detection algorithm return value, it is determined that whether photo was forged.If PCE is more than threshold value, photo is from reference camera, i.e. photo was not forged, if PCE value is less than threshold value, photo was forged.Wherein correlation detection algorithm refers to carry out PRNU noise as device-fingerprint the algorithm of image source discriminating.
As shown in Figure 3, the present embodiment noise matching schematic diagram, with reference to camera installation shooting photo as with reference to photo, picture noise information extraction modules obtains with reference to photo and the PRNU noise of photo to be measured, carries out Calculation of correlation factor by noise information comparison and matching module, calculates PCE, mate, more than threshold value, the match is successful, unsuccessful less than threshold value coupling.
By photo subregion being carried out detection and the matching primitives of PRNU noise, identifying the region that photo is modified, feedback information is sent into display.
Embodiment 2, judge the region that photo is forged
Can interpolate that and be forged region that a kind of method being realized by camera noise differentiating the photographic region true and false is comprised the following steps by the following technical solutions:
(1) use with reference to camera installation shooting photo, as with reference to photo, the grid (n is positive integer) of n*n, n will be divided into reference to photo2Individual an equal amount of piecemeal, is also divided into the grid of n*n, n simultaneously by test photo2Individual an equal amount of piecemeal；
(2) extract with reference to n in photo2The uneven noise data of the photo response of each piecemeal in individual piecemeal, as noise template；
(3) n in test photo is extracted2The uneven noise data of the photo response of each piecemeal in individual piecemeal, compares one by one with each piecemeal of noise template correspondence position, and carries out the calculating of correlation coefficient as the noise data of noise template with correspondence position；Carrying out matching judgment one by one, if the match is successful, the piecemeal of this position of differential test photo comes from reference to camera installation；
(4) identification result is sent to display and shows.
Step (3) calculates n in test photo2In individual piecemeal, the uneven noise data of the photo response of each piecemeal and correspondence position are as the correlation coefficient of the noise data of noise template, described correlation coefficient is to reflect the statistical indicator of dependency relation level of intimate between variable, it is to calculate by product moment method, based on two variablees deviation with respective meansigma methods, it is multiplied by two deviations and reflects degree of correlation between two variablees, utilize correlation peak energy than calculating degree of association.
The resolution principle that described in step (3), the match is successful is: the uneven noise data of photo response of the piecemeal of the test a certain position of photo calculates with the uneven noise data of photo response of the piecemeal of correspondence position in noise template, according to correlation detection algorithm return value, whether the piecemeal of this position of differential test photo was forged, when matching judgment operates, compare correlation peak energy than i.e. can be determined whether coupling with threshold size；If correlation coefficient is more than threshold value, then the match is successful.
Described correlation detection algorithm refers to carry out uneven for photo response noise as device-fingerprint the algorithm of image source discriminating.
By following system, embodiment 2 realizes judging which region of photo is forged, as shown in Figure 2, a kind of system being realized the discriminating photographic region true and false by camera noise, including the picture segmentation module 3 being sequentially connected with, picture noise information extraction modules 2, noise information comparison and matching module 1 and display 4；The information dividing piecemeal with reference to photo and test photo is passed to described picture noise information extraction modules 2 by described picture segmentation module 3；Described picture noise information extraction modules 2, by each piecemeal of reference photo extracted respectively and the uneven noise data of photo response of test photo relevant position piecemeal, passes to described noise information comparison and matching module 1；Described noise information comparison and matching module 1 will carry out Calculation of correlation factor and the result mated sends display 4 to.
Reference photo captured by reference camera is divided into n by picture segmentation module 32Individual an equal amount of piecemeal, i.e. n*n grid, test photo is also divided into n simultaneously2Individual an equal amount of piecemeal.
Picture noise information extraction modules 2 is in order to extract this n2The PRNU noise figure of each piecemeal in individual piecemeal, as noise template.The most also will test photo n2In individual piecemeal, the PRNU noise figure of each piecemeal extracts.
Noise information comparison and matching module 1 are in order to calculate correlation coefficient and the PCE value of each piecemeal of noise template and noise image to be measured.The noise template of the noise information of first piecemeal and first piecemeal calculates, and the noise information of the n-th piecemeal and the noise template of the n-th piecemeal calculate, i.e. the noise information from same position calculates with noise template.
Display is forged the position of part in order to display photos.
Embodiment 2 can not only judge whether the photo captured by the camera of certain model passes through and forge, but also may determine that specifically which piecemeal was forged.Tradition false distinguishing mode being refined, the true figure that for spreading over a whole area from one point reduces provides possibility.
It should be appreciated that the part that this specification does not elaborates belongs to prior art.
Although describe the detailed description of the invention of the present invention above in association with accompanying drawing, but it should be understood by one skilled in the art that these are merely illustrative of, these embodiments can be made various deformation or amendment, without departing from principle and the essence of the present invention.The scope of the present invention is only limited by the claims that follow.
Claims (10)
1. the method realizing differentiating the photo true and false by camera noise, it is characterised in that comprise the following steps:
S1. use with reference to camera installation shooting photo as with reference to photo, extract the photo response with reference to photo uneven
Noise data, as device-fingerprint；
The most again extract test photo the uneven noise data of photo response, compare with device-fingerprint, and with work
Noise data for device-fingerprint carries out the calculating of correlation coefficient；And carry out matching judgment, if the match is successful,
Showing that testing photo comes from reference to camera installation, photo was not revised；
S3. identification result is sent to display show.
The method realizing differentiating the photo true and false by camera noise the most according to claim 1, it is characterised in that
Correlation coefficient described in step S2 is to reflect the statistical indicator of dependency relation level of intimate between variable,
It is to calculate by product moment method, based on two variablees deviation with respective meansigma methods, is multiplied by two deviations
Reflect degree of correlation between two variablees, utilize correlation peak energy than calculating degree of association.
The method realizing identifying the photo true and false by camera noise the most according to claim 1, it is characterised in that
The judgement principle that described in step S2, the match is successful is: according to correlation detection algorithm return value, differential test
Whether photo is modified, and when matching judgment operates, compares correlation peak energy ratio and threshold size
Determine whether coupling；If correlation coefficient is more than threshold value, then the match is successful.
The method realizing differentiating the photo true and false by camera noise the most according to claim 3, it is characterised in that
Described correlation detection algorithm refers to carry out uneven for photo response noise as device-fingerprint the calculation of image source discriminating
Method.
5. one kind realizes differentiating the system of the photo true and false by camera noise, it is characterised in that: include being sequentially connected with
Picture noise information extraction modules, noise information comparison and matching module and display；Described picture noise is believed
Breath extraction module is made an uproar extracting with reference to described in the uneven noise data of photo response of photo and test photo is passed to
Acoustic intelligence comparison and matching module；Described noise information comparison and matching module carry out Calculation of correlation factor and
Join, be resent to display.
6. the method realizing differentiating the photographic region true and false by camera noise, it is characterised in that include following step
Rapid:
(1) use with reference to camera installation shooting photo, as with reference to photo, the net of n*n will be divided into reference to photo
Lattice (n is positive integer), n2Individual an equal amount of piecemeal, is also divided into the grid of n*n simultaneously by test photo,
n2Individual an equal amount of piecemeal；
(2) extract with reference to n in photo2The uneven noise data of the photo response of each piecemeal in individual piecemeal, by it
As noise template；
(3) n in test photo is extracted2The uneven noise data of the photo response of each piecemeal in individual piecemeal, and makes an uproar
Each piecemeal of acoustic mode plate correspondence position is compared one by one, and with correspondence position making an uproar as noise template
Sound data carry out the calculating of correlation coefficient；Carry out matching judgment, if the match is successful, differential test photo one by one
The piecemeal of this position comes from reference to camera installation；
(4) identification result is sent to display and shows.
The method being realized by camera noise differentiating the photographic region true and false the most according to claim 6, its feature
It is, step (3) calculates n in test photo2In individual piecemeal, the photo response of each piecemeal is uneven makes an uproar
Sound data and correspondence position as the correlation coefficient of the noise data of noise template, described correlation coefficient be in order to
The statistical indicator of dependency relation level of intimate between reflection variable, is to calculate by product moment method, with two variablees with
Each based on the deviation of meansigma methods, it is multiplied by two deviations and reflects degree of correlation between two variablees, profit
With correlation peak energy than calculating degree of association.
The method being realized by camera noise differentiating the photographic region true and false the most according to claim 6, its feature
Being, the resolution principle that described in step (3), the match is successful is: the light of the piecemeal of the test a certain position of photo
Sub-Non-uniform responsivity noise data and the photo response uneven noise number of the piecemeal of correspondence position in noise template
According to calculating, according to correlation detection algorithm return value, the piecemeal of this position of differential test photo is pseudo-
Made, when matching judgment operates, compared correlation peak energy than i.e. can be determined whether coupling with threshold size；
If correlation coefficient is more than threshold value, then the match is successful.
The method being realized by camera noise differentiating the photographic region true and false the most according to claim 8, its feature
Being, described correlation detection algorithm refers to as device-fingerprint, uneven for photo response noise is carried out image source mirror
Other algorithm.
10. the system being realized the discriminating photographic region true and false by camera noise, it is characterised in that: include successively
The picture segmentation module that connects, picture noise information extraction modules, noise information comparison and matching module and aobvious
Show device；The information dividing piecemeal with reference to photo and test photo is passed to described figure by described picture segmentation module
Sheet noise information extraction module；Described picture noise information extraction modules is by each for the reference photo extracted respectively
Piecemeal and the uneven noise data of photo response of test photo relevant position piecemeal, pass to described noise information
Comparison and matching module；Described noise information comparison and matching module will carry out Calculation of correlation factor and mate
Result sends display to.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610168248.8A CN105844638A (en) | 2016-03-23 | 2016-03-23 | Method and system for discriminating genuine or counterfeit photos through camera noise |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610168248.8A CN105844638A (en) | 2016-03-23 | 2016-03-23 | Method and system for discriminating genuine or counterfeit photos through camera noise |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105844638A true CN105844638A (en) | 2016-08-10 |
Family
ID=56583068
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610168248.8A Pending CN105844638A (en) | 2016-03-23 | 2016-03-23 | Method and system for discriminating genuine or counterfeit photos through camera noise |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105844638A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109214781A (en) * | 2018-09-12 | 2019-01-15 | 医倍思特（北京）医疗信息技术有限公司 | A kind of people hurts the check method of Claims Resolution setting loss expense |
| CN110097132A (en) * | 2019-05-07 | 2019-08-06 | 电子科技大学 | A method of identification digital photos and shooting camera |
| CN111027573A (en) * | 2019-12-09 | 2020-04-17 | 广西荔玉高速公路有限公司 | Image authenticity identification method based on blind evidence obtaining technology |
| CN112367457A (en) * | 2020-04-08 | 2021-02-12 | 齐鲁工业大学 | Video PRNU noise extraction method and camera source detection method |
| CN112991345A (en) * | 2021-05-11 | 2021-06-18 | 腾讯科技（深圳）有限公司 | Image authenticity detection method and device, computer equipment and storage medium |
| CN113052836A (en) * | 2021-04-21 | 2021-06-29 | 深圳壹账通智能科技有限公司 | Electronic identity photo detection method and device, electronic equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102819831A (en) * | 2012-08-16 | 2012-12-12 | 江南大学 | Camera source evidence obtaining method based on mode noise big component |
| WO2012168239A1 (en) * | 2011-06-06 | 2012-12-13 | Carl Zeiss Meditec Ag | Systems and methods for improved balanced detection in optical coherence tomography imaging |
-
2016
- 2016-03-23 CN CN201610168248.8A patent/CN105844638A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012168239A1 (en) * | 2011-06-06 | 2012-12-13 | Carl Zeiss Meditec Ag | Systems and methods for improved balanced detection in optical coherence tomography imaging |
| CN102819831A (en) * | 2012-08-16 | 2012-12-12 | 江南大学 | Camera source evidence obtaining method based on mode noise big component |
Non-Patent Citations (1)
| Title |
|---|
| 谭莉玲: "利用相关性检测的成像设备辨识及篡改检测技术", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109214781A (en) * | 2018-09-12 | 2019-01-15 | 医倍思特（北京）医疗信息技术有限公司 | A kind of people hurts the check method of Claims Resolution setting loss expense |
| CN110097132A (en) * | 2019-05-07 | 2019-08-06 | 电子科技大学 | A method of identification digital photos and shooting camera |
| CN111027573A (en) * | 2019-12-09 | 2020-04-17 | 广西荔玉高速公路有限公司 | Image authenticity identification method based on blind evidence obtaining technology |
| CN111027573B (en) * | 2019-12-09 | 2023-05-12 | 广西荔玉高速公路有限公司 | Image authenticity identification method based on blind evidence obtaining technology |
| CN112367457A (en) * | 2020-04-08 | 2021-02-12 | 齐鲁工业大学 | Video PRNU noise extraction method and camera source detection method |
| CN112367457B (en) * | 2020-04-08 | 2022-03-22 | 齐鲁工业大学 | Video PRNU noise extraction method and camera source detection method |
| CN113052836A (en) * | 2021-04-21 | 2021-06-29 | 深圳壹账通智能科技有限公司 | Electronic identity photo detection method and device, electronic equipment and storage medium |
| CN112991345A (en) * | 2021-05-11 | 2021-06-18 | 腾讯科技（深圳）有限公司 | Image authenticity detection method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105844638A (en) | Method and system for discriminating genuine or counterfeit photos through camera noise | |
| Wu et al. | Busternet: Detecting copy-move image forgery with source/target localization | |
| CN110728209B (en) | Gesture recognition method and device, electronic equipment and storage medium | |
| CN108171184B (en) | Method for re-identifying pedestrians based on Simese network | |
| CN105740775B (en) | Three-dimensional face living body identification method and device | |
| CN111931758B (en) | Face recognition method and device combining facial veins | |
| CN109190475B (en) | Face recognition network and pedestrian re-recognition network collaborative training method | |
| CN110263603B (en) | Face recognition method and device based on central loss and residual error visual simulation network | |
| CN103942777A (en) | Mobile phone glass cover plate defect detecting method based on principal component analysis | |
| CN106991448A (en) | A kind of portrait compares processing method | |
| CN112818722B (en) | Modular dynamic configurable living body face recognition system | |
| CN110942450A (en) | Multi-production-line real-time defect detection method based on deep learning | |
| CN108446687B (en) | Self-adaptive face vision authentication method based on interconnection of mobile terminal and background | |
| CN109389105B (en) | Multitask-based iris detection and visual angle classification method | |
| CN112150450A (en) | Image tampering detection method and device based on dual-channel U-Net model | |
| CN105224903B (en) | A kind of reading device of the method for anti-counterfeit and Quick Response Code of Quick Response Code | |
| Hou et al. | Detection of hue modification using photo response nonuniformity | |
| CN110222486A (en) | User ID authentication method, device, equipment and computer readable storage medium | |
| CN110457996A (en) | Moving Objects in Video Sequences based on VGG-11 convolutional neural networks distorts evidence collecting method | |
| CN111582118A (en) | Face recognition method and device | |
| CN110121109A (en) | Towards the real-time source tracing method of monitoring system digital video, city video monitoring system | |
| CN111223078B (en) | Method for determining flaw level and storage medium | |
| CN105120294A (en) | JPEG format image source identification method | |
| CN112819756A (en) | PCB surface defect detection device and method | |
| KR20080079798A (en) | Method of face detection and recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160810 |
|
| WD01 | Invention patent application deemed withdrawn after publication |