CN109376518A - Privacy leakage method and relevant device are prevented based on recognition of face - Google Patents
Privacy leakage method and relevant device are prevented based on recognition of face Download PDFInfo
- Publication number
- CN109376518A CN109376518A CN201811214066.5A CN201811214066A CN109376518A CN 109376518 A CN109376518 A CN 109376518A CN 201811214066 A CN201811214066 A CN 201811214066A CN 109376518 A CN109376518 A CN 109376518A
- Authority
- CN
- China
- Prior art keywords
- face
- image
- facial image
- point
- eyes
- 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
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000001815 facial effect Effects 0.000 claims abstract description 103
- 239000000284 extract Substances 0.000 claims abstract description 12
- 239000011159 matrix material Substances 0.000 claims description 32
- 230000008569 process Effects 0.000 claims description 13
- 230000009466 transformation Effects 0.000 claims description 12
- 210000000887 face Anatomy 0.000 claims description 8
- 230000006872 improvement Effects 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000008439 repair process Effects 0.000 claims description 4
- 230000001419 dependent effect Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 239000007787 solid Substances 0.000 claims 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 2
- 230000004438 eyesight Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 208000008918 voyeurism Diseases 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000009545 invasion Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 210000004279 orbit Anatomy 0.000 description 1
- 244000144985 peep Species 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 210000000216 zygoma Anatomy 0.000 description 1
Classifications
-
- 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
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/165—Detection; Localisation; Normalisation using facial parts and geometric relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/40—Spoof detection, e.g. liveness detection
- G06V40/45—Detection of the body part being alive
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Security & Cryptography (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Collating Specific Patterns (AREA)
Abstract
This application involves field of information security technology more particularly to a kind of privacy leakage method and relevant device are prevented based on recognition of face.It is a kind of that privacy leakage method is prevented based on recognition of face, include the following steps: to obtain original image, extracts the facial image in the original image, the facial image is handled to enhance clarity；The face three-dimensional reference model for establishing the facial image, determines the position coordinates of eyes；Eyes position coordinate is compared with pre-set distance threshold, face characteristic acquisition is carried out to the face three-dimensional reference model if being less than the distance threshold, otherwise without acquisition；The face characteristic is parsed, the identification point of the facial image is obtained；The identification point is compared with inherent feature point, if the identification point compares inconsistent with the inherent feature point, carries out screen locking operation.The application realizes the purpose for preventing mobile phone private from revealing by carrying out automatic identification to the face around mobile phone.
Description
Technical field
This application involves field of information security technology more particularly to a kind of privacy leakage method is prevented based on recognition of face
And relevant device.
Background technique
Individual privacy, which refers to, is reluctant for other people secrets that are open or knowing in individual citizens life.The right of privacy refers to natural person
A kind of people that its people, unrelated with public relations interests personal information, personal activity and privately owned field are dominated enjoyed
Lattice power.In the epoch of this information prosperity, increasingly focus on the protection of individual privacy between people.But often some public
Occasion, people chat the modes such as wechat, QQ, short message by mobile phone, and the people that is often not intended to pass by one's side, which sees, to be sent
The information content even will lead to privacy leakage, and the thing of some invasions of privacy occurs.
Currently, secret protection is solved usually by way of inputting password, however in mobile phone use process, such as
Fruit encounters other people and peeps, then can only manually be identified by user, frequently can lead to the leakage of personal information, or in hand
Reflective mirror is arranged on machine the image of person of peeping is presented on Mobile phone screen.
But this method cannot in time identify person of peeping, and information cannot when using mobile phone by user
It is effectively protected.
Summary of the invention
In view of this, it is necessary to prevent privacy leakage to be unable to automatic identification for existing, the low problem of recognition efficiency mentions
Privacy leakage method and relevant device are prevented for one kind.
It is a kind of that privacy leakage method is prevented based on recognition of face, include the following steps:
Original image is obtained, the facial image in the original image is extracted, the facial image is handled to increase
Strong clarity；
The face three-dimensional reference model for establishing the facial image, determines the position coordinates of eyes；
Eyes position coordinate is compared with pre-set distance threshold, if being less than the distance threshold
Face characteristic acquisition then is carried out to the face three-dimensional reference model, otherwise without acquisition；
The face characteristic is parsed, the identification point of the facial image is obtained；
The identification point is compared with inherent feature point, if the identification point compared with the inherent feature point it is different
It causes, then carries out screen locking operation, the inherent feature point refers to the had biological characteristic of people's birth, and storage is in the database.
The acquisition original image in one of the embodiments, extracts the facial image in the original image, to institute
Facial image is stated to be handled to enhance clarity, comprising:
Acquire original image；
The original image is scanned, facial contour and nose position are obtained；
According to the facial contour and the nose shape, facial image interception is carried out from the original image；
The facial image being truncated to is carried out gray scale adjusting and sharpened to adjust, obtains the enhanced face figure of clarity
Picture.
The face three-dimensional reference model for establishing the facial image in one of the embodiments, determines eyes
Position coordinates, comprising:
Three dimensional face reference model, the history data store of the facial image are established according to the historical data of facial image
In the database；
Assessed by posture of the three dimensional face reference model to the facial image of extraction, obtain attitude parameter and
The virtual image of the relatively described facial image of three dimensional face reference model；
Using the virtual image as prior information, the facial depth image of the facial image is rebuild；
The position that eyes in the facial depth image are detected according to the attitude parameter, obtains the position pair of the eyes
The scale invariant feature SIFT descriptor answered；
According to location information corresponding to the SIFT descriptor, the coordinate of the eye position is determined, wherein described
Location information corresponding to SIFT descriptor stores in the database.
It is described in one of the embodiments, to carry out eyes position coordinate and pre-set distance threshold
Compare, face characteristic acquisition is carried out to the face three-dimensional reference model if being less than the distance threshold, otherwise without adopting
Collection, comprising:
According to the historical data set distance threshold value of face characteristic clarity, the historical data of the face characteristic clarity
Storage is in the database；
Emit light to the eyes, receives the reflection light through the eye reflections；
The walking path for obtaining the reflection light revises the distance threshold according to the walking path；
According to eyes position coordinate, the eyes are calculated to the distance L of screen, formula is L=(x2+y2+z2)1/2,
Wherein, L is distance of the eyes to screen, and x, y, z is the coordinate of eyes position；
L is compared with the distance threshold, to the face three-dimensional reference model if being less than the distance threshold
Face characteristic acquisition is carried out, otherwise without acquisition.
The blocking node for obtaining the data group in one of the embodiments, it is special according to the parsing face
Sign, obtains the identification point of the facial image, comprising:
By the face characteristic according to horizontally and vertically decomposing, each point in the face characteristic is obtained
Horizontal gradient and vertical gradient；
The set A of the horizontal gradient and the set B of the vertical gradient are established, using Pasteur's distance algorithm to similarity
It is calculated:
Dij=1- Σ kfeature_i (k) feature_j (k) (Σ s feature j (s)) (Σ s feature i
(s)),
In formula, DijJ-th point is indicated in the set A of the horizontal gradient in the set B of the and vertical gradient at i-th point
Similarity, variable k and s be two dependent variables, feature_i (k) indicate k-th of characteristic point histogram in, horizontal gradient i's
Pixel quantity, feature_j (k) indicate k-th of characteristic point histogram in, the pixel quantity of vertical gradient i, feature_i (s)
It indicates in s-th of characteristic point histogram, the pixel quantity of horizontal gradient i, feature_j (s) indicates s-th of characteristic point histogram
In, the pixel quantity of vertical gradient i；
According to the similarity calculation as a result, establishing similarity matrix；
The similarity matrix is optimized using Hungary Algorithm, extracts similitude, the similitude is described
The identification point of facial image.
It is described in one of the embodiments, that the identification point is compared with inherent feature point, if the identification point
It is compared with the inherent feature point inconsistent, then carries out screen locking operation, the inherent feature point refers to that people is born had life
Object feature, storage is in the database, comprising:
Establish the first picture element matrix of the identification point and the second picture element matrix of inherent feature point；
Numerical value in first picture element matrix and second picture element matrix is subjected to binary conversion treatment, is only wrapped
First containing 0 and 1 binary code, which improves picture element matrix and second, improves picture element matrix；
It compares in the first improvement picture element matrix and the second improvement picture element matrix in mutually going together or same column
Each numerical value carry out screen locking operation if there is any pair numerical value different.
The acquisition original image in one of the embodiments, extracts the facial image in the original image, to institute
It states facial image to be handled to enhance clarity, further includes judging whether facial image is living body faces, specifically include:
The facial image count the lighting process of mean value, obtains and repairs image；
Fourier transformation processing is carried out to the repairerment image, obtains the transformed value for repairing each pixel of image；
According to the transformed value for repairing each pixel of image, most high frequency component values and lowest frequency component value are obtained, it will
The most high frequency component values and the lowest frequency component value difference obtain transformed value；
The transformed value is put into SVM model and is trained, true transformation value is obtained；
The true transformation value is compared with default classification thresholds, if the true transformation value is greater than described default point
Class threshold value, then otherwise it is reproduction image that explanation, which is living body faces,.
It is a kind of that privacy leakage device, including following module are prevented based on recognition of face:
Image acquisition process module is set as obtaining original image, the facial image in the original image is extracted, to institute
Facial image is stated to be handled to enhance clarity；
File type identification module is arranged to set up the face three-dimensional reference model of the facial image, determines eyes
Position coordinates；
Reference model module is established, is set as according to the file type, application tree-model carries out the file
A data group is aggregated into after data pick-up；
Image capture module is set as comparing on eyes position coordinate and pre-set distance threshold
Compared with to face three-dimensional reference model progress face characteristic acquisition if being less than the distance threshold, otherwise without acquisition；
Identification point obtains module, is set as parsing the face characteristic, obtains the identification point of the facial image；
Identification module is compareed, is set as the identification point being compared with inherent feature point, if the identification point and institute
It is inconsistent to state the comparison of inherent feature point, then carries out screen locking operation, the inherent feature point refers to the biology spy that people is born had
Sign, storage is in the database.
A kind of computer equipment, including memory and processor are stored with computer-readable instruction in the memory, institute
When stating computer-readable instruction and being executed by the processor so that the processor execute it is above-mentioned based on recognition of face prevent it is hidden
The step of private leakage method.
A kind of storage medium being stored with computer-readable instruction, the computer-readable instruction are handled by one or more
Device execute when so that one or more processors execute it is above-mentioned based on recognition of face the step of preventing privacy leakage.
It is above-mentioned that privacy leakage method, apparatus, computer equipment and storage medium are prevented based on recognition of face, including obtain
Original image extracts the facial image in the original image, is handled the facial image to enhance clarity；It establishes
The face three-dimensional reference model of the facial image, determines the position coordinates of eyes；By eyes position coordinate and in advance
The distance threshold being first arranged is compared, and it is special to carry out face to the face three-dimensional reference model if being less than the distance threshold
Sign acquisition, otherwise without acquisition；The face characteristic is parsed, the identification point of the facial image is obtained；By the identification point
It is compared with inherent feature point, if the identification point compares inconsistent with the inherent feature point, carries out screen locking operation, institute
It states inherent feature point and refers to the had biological characteristic of people's birth, storage is in the database.The technical program is prevented for existing
Privacy leakage is unable to automatic identification, the low problem of recognition efficiency, by carrying out automatic identification realization to the face around mobile phone
The purpose for preventing mobile phone private from revealing.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the application
Limitation.
Fig. 1 is a kind of overall flow figure for preventing privacy leakage method based on recognition of face of the application；
Fig. 2 is a kind of determination eye position coordinate mistake prevented in privacy leakage method based on recognition of face of the application
Journey schematic diagram；
Fig. 3 is that a kind of face characteristic collection process prevented in privacy leakage method based on recognition of face of the application is shown
It is intended to；
Fig. 4 is a kind of structure chart for preventing privacy leakage device based on recognition of face of the application.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, and
It is not used in restriction the application.
Those skilled in the art of the present technique are appreciated that unless expressly stated, singular " one " used herein, " one
It is a ", " described " and "the" may also comprise plural form.It is to be further understood that being arranged used in the description of the present application
Diction " comprising " refer to that there are the feature, integer, step, operation, element and/or component, but it is not excluded that in the presence of or addition
Other one or more features, integer, step, operation, element, component and/or their group.
Fig. 1 is the flow chart for preventing privacy leakage method based on recognition of face in the application one embodiment, such as Fig. 1
Shown, one kind preventing privacy leakage method, comprising the following steps:
S1, obtain original image, extract the facial image in the original image, to the facial image handled with
Enhance clarity；
Specifically, collecting the representative people of different expressions and different angle that designated person is shot under different light environments
Face picture, and the face picture of a large amount of random non-designated personage, by the sample face of designated person and a large amount of non-designated at random
The face of personage carries out deformation fusion to obtain the face edge pattern of designated person.Face edge pattern is by designated person
Sample face and the face of a large amount of random non-designated personages carry out deformation and merge to obtain.Wherein, positive sample edge face is set
It is set to the face for just belonging to designated person, is obtained by the deformation of shallower degree；Negative sample edge face is set as not belonging to just
In the face of designated person, obtained by the deformation of deeper degree.
S2, the face three-dimensional reference model for establishing the facial image, determine the position coordinates of eyes；
Specifically, three-dimensional face reference model is the knowledge for increasing stereoscopic features point on the basis of two-dimension human face reference model
Not, it establishes three-dimensional system of coordinate to identify the position of human body face especially eyes, mainly passes through depth measurement and spy
Sign point identification is to achieve the purpose that determining eye position.
S3, eyes position coordinate is compared with pre-set distance threshold, if being less than the distance
Threshold value then carries out face characteristic acquisition to the face three-dimensional reference model, otherwise without acquisition；
Specifically, the distance in this step refers at a distance from eye distance screen with threshold value, be usually arranged as 2m, i.e., if
The eyesight of one people be 5.0 its can see that the 1cm figure outside distance 2m, this distance threshold can be fed back according to user
Situation is adjusted, and can also be counted and is brought into according to big data and is trained to obtain optimal distance threshold in deep learning model
Value.
S4, the parsing face characteristic, obtain the identification point of the facial image；
Specifically, the face characteristic of acquisition is decomposed, such as nose feature, nose depth of beam, the size of nose is double
These face characteristics are decomposed to obtain different characteristic points, these characteristic points are carried out by the spacing of eye, the size etc. of cheekbone
Sequence obtains feature identification point of the ranking in preceding 5 characteristic points as this face.
S5, the identification point is compared with inherent feature point, if the identification point is compared with the inherent feature point
It is inconsistent, then screen locking operation is carried out, the inherent feature point refers to the had biological characteristic of people's birth, is stored in database
In.
Specifically, inherent feature point is the biological characteristic just having after people is born, change will not be generated because of lift face etc.
Change, if the identification point obtained in the previous step all meets after being compared with inherent feature point, it is intrinsic special to transfer this
The corresponding people of sign point, that is to say, that prove that the face belongs to this person.
The application realizes the quick identification to facial image, and can effectively exclude not by above method step
Facial image in monitoring range.
The acquisition original image in one of the embodiments, extracts the facial image in the original image, to institute
Facial image is stated to be handled to enhance clarity, comprising:
Acquire original image；
The original image is scanned, facial contour and nose position are obtained；
Specifically, original image is the face images in monitoring range, since the position of nose in face is near preceding,
Therefore during carrying out Image Acquisition, first determine that whether again the position of nose can determine the face within monitoring range.
According to the facial contour and the nose shape, facial image interception is carried out from the original image；
The facial image being truncated to is carried out gray scale adjusting and sharpened to adjust, obtains the enhanced face figure of clarity
Picture.
Specifically, the brightness Y and tri- color components of R, G, B that are established according to the variation relation of RGB and YUV color space
Corresponding relationship expresses the gray value of image, used formula are as follows: Y=0.5R+0.8G+0.2B, and expressed according to brightness value Y
The gray value of image；Sharpen adjust refer to using sharpening tool can quickly focus blur edge, improve image in a certain position
Clarity or focal length degree, keep the color of image specific region distincter.In application sharpening tool, if choosing its choosing
" sampling to All Layers " check box in item column, then can be sharpened the image in all visible layers.
It in the present embodiment, is determined by the nose shape to face, can effectively judge position at face
It sets, so that those faces not in monitoring range be discharged, promotes the recognition efficiency of effective face.
Fig. 2 in one embodiment, determines eye position coordinate process schematic diagram, as shown, described build for the application
The face three-dimensional reference model for founding the facial image, determines the position coordinates of eyes, comprising:
S201, three dimensional face reference model, the history number of the facial image are established according to the historical data of facial image
In the database according to storage；
Image is trained to obtain multiple people specifically, building three dimensional face reference model and mainly passing through SVM model
Face feature identifier carries out decomposition-training to face according to face characteristic identifier and obtains the facial skeleton feature of face, eyes
Feature and nose feature are acquired historical data and obtain different three dimensional face reference models.
S202, it is assessed by posture of the three dimensional face reference model to the facial image of extraction, obtains posture
The virtual image of parameter and the relatively described facial image of three dimensional face reference model；
Specifically, so-called attitude parameter be by face in binocular vision image three-dimensional space rotation angle and face
It is mapped to the position in two-dimensional surface.Virtual image then contains the face with three dimensional face reference model, but the face then with
The consistent image of the human face posture and size of binocular vision image pair, obtains people according to virtual image and binocular vision image
The parameter of face posture.
S203, using the virtual image as prior information, rebuild the facial depth image of the facial image；
Specifically, virtual image is two dimensional image, the correspondence mappings relationship of it and 3-D image is established, according to mapping relations
The depth value of facial depth image is established, this depth value of root rebuilds facial image.
S204, the position that eyes in the facial depth image are detected according to the attitude parameter, obtain the eyes
The corresponding scale invariant feature SIFT descriptor in position；
Wherein, the purpose of feature description is that this key point is depicted come this with one group of vector after key point calculating
A description not only includes key point, also includes around key point to its contributive pixel.It is used as object matching
Foundation can also make key point have more invariant features, such as illumination variation, 3D viewpoint change.The thinking of feature description: right
Image-region piecemeal around key point, calculation block inside gradient histogram, generates unique vector, this vector is the area
One kind of area image information is abstract, has uniqueness.
S205, the location information according to corresponding to the SIFT descriptor, determine the coordinate of the eye position, wherein
Location information corresponding to the SIFT descriptor stores in the database.
Specifically, characteristic point can be the point on nose, the point on left eye and point on right eye etc., it will be by that can have
Effect reflects the Local grid scale invariant feature descriptor (SIFT feature) of three dimensional face shape mesoscale and rotational invariance
The accurate description of characteristic point is carried out, ensure that the consistency of characteristic point, and then improves the standard of three dimensional face shape description
True property and robustness.
In the present embodiment, by establishing three dimensional face reference model, to determine that the position of eyes can be to the position of person of peeping
It sets and is accurately determined, can preferably judge there are several individuals peeping after then positioning to eyes.
Fig. 3 be the application in one embodiment, face characteristic collection process schematic diagram, as shown, described will be described
Eyes position coordinate is compared with pre-set distance threshold, to the face three if being less than the distance threshold
It ties up reference model and carries out face characteristic acquisition, otherwise without acquisition, comprising:
S301, the historical data set distance threshold value according to face characteristic clarity, the face characteristic clarity are gone through
History data store in the database；
Specifically, the clarity of face characteristic is to carry out DPI judgement to the facial image of acquisition, but when DPI is less than 300,
Then think that this man face image acquiring is invalid, need again to be acquired facial image, if be still less than 300 if abandon this
Facial image, and adjust the requirement that distance threshold makes acquired image meet clarity.
S302, emit light to the eyes, receive the reflection light through the eye reflections；
Specifically, emitting light to eyes, the launch angle set is 30 °~150 °, the transmitting light beyond this range
Nonsensical, the transmitting light for receiving eye reflections is also defined within the scope of this, can determine this person in this way it can be seen that
Content on screen.
S303, the walking path for obtaining the reflection light, repair the distance threshold according to the walking path
It orders；
Specifically, if the walking path of reflection light concentrate with a point of screen, such as screen edge if think
This person can not obtain the main contents of screen, then exclude the facial image of this distance.
S304, according to eyes position coordinate, calculate the eyes to the distance L of screen, formula is L=(x2+
y2+z2)1/2, wherein L is distance of the eyes to screen, and x, y, z is the coordinate of eyes position；
Specifically, calculate eye distance screen apart from when, be based on three-dimensional cartesian coordinate system and calculated, can also
To use polar coordinate system to be calculated, meanwhile, when calculating eyes can be obtained in a word to screen using taking camber line to calculate
The shortest distance.
S305, L is compared with the distance threshold, the face three-dimensional is referred to if being less than the distance threshold
Model carries out face characteristic acquisition, otherwise without acquisition.
In the present embodiment, by being defined to face characteristic collection process, can effectively collect can be can be carried out steathily
The facial image peeped, to promote the efficiency of recognition of face.
In one embodiment, the parsing face characteristic, obtains the identification point of the facial image, comprising:
By the face characteristic according to horizontally and vertically decomposing, each point in the face characteristic is obtained
Horizontal gradient and vertical gradient；
Specifically, horizontal gradient be used to indicate face cheek to ear depth variation, vertical gradient then indicates nose
Depth or eye socket depth.
The set A of the horizontal gradient and the set B of the vertical gradient are established, using Pasteur's distance algorithm to similarity
It is calculated:
Dij=1- Σ kfeature_i (k) feature_j (k) (Σ s feature j (s)) (Σ s feature i
(s)),
In formula, DijJ-th point is indicated in the set A of the horizontal gradient in the set B of the and vertical gradient at i-th point
Similarity, variable k and s be two dependent variables, feature_i (k) indicate k-th of characteristic point histogram in, horizontal gradient i's
Pixel quantity, feature_j (k) indicate k-th of characteristic point histogram in, the pixel quantity of vertical gradient i, feature_i (s)
It indicates in s-th of characteristic point histogram, the pixel quantity of horizontal gradient i, feature_j (s) indicates s-th of characteristic point histogram
In, the pixel quantity of vertical gradient i；
According to the similarity calculation as a result, establishing similarity matrix；
The similarity matrix is optimized using Hungary Algorithm, extracts similitude, the similitude is described
The identification point of facial image.
Wherein, Hungary Algorithm is a kind of combinatorial optimization algorithm that Task Allocation Problem is solved in polynomial time, and
Later original Dual Method is pushed.U.S.'s mathematician Harold library grace proposes the algorithm in nineteen fifty-five.The institute of this algorithm
It is because algorithm is greatly based on former Hungary mathematician D é nes to be referred to as Hungary AlgorithmWithCreation is got up on the work of Egerv á ry, mainly finds the path of best match.
In the present embodiment, the combination of Hungary Algorithm and Pasteur's distance algorithm can preferably find identification point.
In one embodiment, described that the identification point is compared with inherent feature point, if the identification point and institute
It is inconsistent to state the comparison of inherent feature point, then carries out screen locking operation, the inherent feature point refers to the biology spy that people is born had
Sign, storage is in the database, comprising:
Establish the first picture element matrix of the identification point and the second picture element matrix of inherent feature point；
Numerical value in first picture element matrix and second picture element matrix is subjected to binary conversion treatment, is only wrapped
First containing 0 and 1 binary code, which improves picture element matrix and second, improves picture element matrix；
It compares in the first improvement picture element matrix and the second improvement picture element matrix in mutually going together or same column
Each numerical value carry out screen locking operation if there is any pair numerical value different.
In the present embodiment, become the rate of exchange by establishing the first picture element matrix and the second picture element matrix and binary conversion treatment
It is simple and fast.
In one embodiment, the acquisition original image, extracts the facial image in the original image, to the people
Face image is handled to enhance clarity, is further included judging whether facial image is living body faces, is specifically included:
The facial image count the lighting process of mean value, obtains and repairs image；
Specifically, carrying out PDAM (Point Divid Arithmetic to each pixel in the facial image
Mean, arithmetic equal value quotient) lighting process, that is, the pixel mean value in the facial image is calculated, by the original pixel value of each pixel
It is determined as the pixel value of each pixel in first image with the ratio of the pixel mean value.The repairerment image is carried out in Fu
Leaf transformation processing obtains the transformed value for repairing each pixel of image.
According to the transformed value for repairing each pixel of image, most high frequency component values and lowest frequency component value are obtained, it will
The most high frequency component values and the lowest frequency component value difference obtain transformed value；
The transformed value is put into SVM model and is trained, true transformation value is obtained；
Specifically, the transformed value of input first carries out supporting vector product conversion, extensive when progress SVM model is trained
The formula of error bounds are as follows: R (w)≤Remp (w)+Ф (n/h), R (w) is exactly real risk in formula, and Remp (w) is exactly experience wind
Danger, Ф (n/h) is exactly confidence risk.The target of statistical learning has become seeking empiric risk and confidence from empirical risk minimization
It is risk and minimum, i.e. Structural risk minization.
The true transformation value is compared with default classification thresholds, if the true transformation value is greater than described default point
Class threshold value, then otherwise it is reproduction image that explanation, which is living body faces,.
In the present embodiment, the interference that prevents inhuman face image from generating to identification process by the identification to living body faces.
In one embodiment, provide it is a kind of privacy leakage device is prevented based on recognition of face, as shown in figure 4, packet
It includes:
Image acquisition process module is set as obtaining original image, the facial image in the original image is extracted, to institute
Facial image is stated to be handled to enhance clarity；
File type identification module is arranged to set up the face three-dimensional reference model of the facial image, determines eyes
Position coordinates；
Reference model module is established, is set as according to the file type, application tree-model carries out the file
A data group is aggregated into after data pick-up；
Image capture module is set as comparing on eyes position coordinate and pre-set distance threshold
Compared with to face three-dimensional reference model progress face characteristic acquisition if being less than the distance threshold, otherwise without acquisition；
Identification point obtains module, is set as parsing the face characteristic, obtains the identification point of the facial image；
Identification module is compareed, is set as the identification point being compared with inherent feature point, if the identification point and institute
It is inconsistent to state the comparison of inherent feature point, then carries out screen locking operation, the inherent feature point refers to the biology spy that people is born had
Sign, storage is in the database.
In one embodiment it is proposed that a kind of computer equipment, including memory and processor, deposited in the memory
Computer-readable instruction is contained, when the computer-readable instruction is executed by the processor, so that the processor executes
State described in each embodiment based on recognition of face the step of preventing privacy leakage method.
In one embodiment it is proposed that a kind of storage medium for being stored with computer-readable instruction, computer-readable finger
When order is executed by one or more processors, so that one or more processors, which execute, is based on face described in the various embodiments described above
The step of preventing privacy leakage method of identification.Wherein, the storage medium can be non-volatile memory medium.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random
Access Memory), disk or CD etc..
Each technical characteristic of embodiment described above can be combined arbitrarily, for simplicity of description, not to above-mentioned reality
It applies all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, all should be considered as described in this specification.
The some exemplary embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but
It cannot be understood as the limitations to the application the scope of the patents.It should be pointed out that for the ordinary skill people of this field
For member, without departing from the concept of this application, various modifications and improvements can be made, these belong to the application's
Protection scope.Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of prevent privacy leakage method based on recognition of face characterized by comprising
Original image is obtained, the facial image in the original image is extracted, the facial image is handled clear to enhance
Clear degree；
The face three-dimensional reference model for establishing the facial image, determines the position coordinates of eyes；
Eyes position coordinate is compared with pre-set distance threshold, it is right if being less than the distance threshold
The face three-dimensional reference model carries out face characteristic acquisition, otherwise without acquisition；
The face characteristic is parsed, the identification point of the facial image is obtained；
The identification point is compared with inherent feature point, if the identification point compared with the inherent feature point it is inconsistent,
Screen locking operation is then carried out, the inherent feature point refers to the had biological characteristic of people's birth, and storage is in the database.
2. according to claim 1 prevent privacy leakage method based on recognition of face, which is characterized in that the acquisition is former
Beginning image extracts the facial image in the original image, is handled the facial image to enhance clarity, comprising:
Acquire original image；
The original image is scanned, facial contour and nose position are obtained；
According to the facial contour and the nose shape, facial image interception is carried out from the original image；
The facial image being truncated to is carried out gray scale adjusting and sharpened to adjust, obtains the enhanced facial image of clarity.
3. according to claim 1 prevent privacy leakage method based on recognition of face, which is characterized in that described to establish institute
The face three-dimensional reference model for stating facial image, determines the position coordinates of eyes, comprising:
Three dimensional face reference model is established according to the historical data of facial image, the history data store of the facial image is in number
According in library；
It is assessed by posture of the three dimensional face reference model to the facial image of extraction, obtains attitude parameter and three-dimensional
The virtual image of the relatively described facial image of facial reference model；
Using the virtual image as prior information, the facial depth image of the facial image is rebuild；
The position of eyes in the facial depth image is detected according to the attitude parameter, the position for obtaining the eyes is corresponding
Scale invariant feature SIFT descriptor；
According to location information corresponding to the SIFT descriptor, the coordinate of the eye position is determined, wherein the SIFT is retouched
State the corresponding location information storage of symbol in the database.
4. according to claim 1 prevent privacy leakage method based on recognition of face, which is characterized in that it is described will be described
Eyes position coordinate is compared with pre-set distance threshold, to the face three if being less than the distance threshold
It ties up reference model and carries out face characteristic acquisition, otherwise without acquisition, comprising:
According to the historical data set distance threshold value of face characteristic clarity, the history data store of the face characteristic clarity
In the database；
Emit light to the eyes, receives the reflection light through the eye reflections；
The walking path for obtaining the reflection light revises the distance threshold according to the walking path；
According to eyes position coordinate, the eyes are calculated to the distance L of screen, formula is L=(x2+y2+z2)1/2,
Wherein, L is distance of the eyes to screen, and x, y, z is the coordinate of eyes position；
L is compared with the distance threshold, the face three-dimensional reference model is carried out if being less than the distance threshold
Face characteristic acquisition, otherwise without acquisition.
5. according to claim 1 prevent privacy leakage method based on recognition of face, which is characterized in that the parsing institute
Face characteristic is stated, the identification point of the facial image is obtained, comprising:
By the face characteristic according to horizontally and vertically decomposing, the water of each point in the face characteristic is obtained
Flat ladder degree and vertical gradient；
The set A of the horizontal gradient and the set B of the vertical gradient are established, similarity is carried out using Pasteur's distance algorithm
It calculates:
Dij=1- Σ k feature_i (k) feature_j (k) (Σ s feature j (s)) (Σ s feature i
(s)),
In formula, DijIndicate in the set A of the horizontal gradient j-th point in the set B of the and vertical gradient of phase at i-th point
Like degree, variable k and s are two dependent variables, and feature_i (k) is indicated in k-th of characteristic point histogram, the pixel of horizontal gradient i
Quantity, feature_j (k) indicate in k-th of characteristic point histogram that the pixel quantity of vertical gradient i, feature_i (s) is indicated
In s-th of characteristic point histogram, the pixel quantity of horizontal gradient i, feature_j (s) is indicated in s-th of characteristic point histogram,
The pixel quantity of vertical gradient i；
According to the similarity calculation as a result, establishing similarity matrix；
The similarity matrix is optimized using Hungary Algorithm, extracts similitude, the similitude is the face
The identification point of image.
6. according to claim 1 prevent privacy leakage method based on recognition of face, which is characterized in that it is described will be described
Identification point is compared with inherent feature point, if the identification point compares inconsistent with the inherent feature point, carries out screen locking
Operation, the inherent feature point refer to the had biological characteristic of people's birth, and storage is in the database, comprising:
Establish the first picture element matrix of the identification point and the second picture element matrix of inherent feature point；
Numerical value in first picture element matrix and second picture element matrix is subjected to binary conversion treatment, is obtained only comprising 0
Picture element matrix and second, which is improved, with the first of 1 binary code improves picture element matrix；
Compare it is described first improvement picture element matrix and it is described second improvement picture element matrix in mutually go together or same column it is every
One numerical value carries out screen locking operation if there is any pair of numerical value different.
7. according to claim 1 prevent privacy leakage method based on recognition of face, which is characterized in that the acquisition is former
Beginning image extracts the facial image in the original image, is handled the facial image to enhance clarity, further includes
Judge whether facial image is living body faces, is specifically included:
The facial image count the lighting process of mean value, obtains and repairs image；
Fourier transformation processing is carried out to the repairerment image, obtains the transformed value for repairing each pixel of image；
According to the transformed value for repairing each pixel of image, most high frequency component values and lowest frequency component value are obtained, it will be described
Most high frequency component values and the lowest frequency component value difference obtain transformed value；
The transformed value is put into SVM model and is trained, true transformation value is obtained；
The true transformation value is compared with default classification thresholds, if the true transformation value is greater than the default classification threshold
Value, then otherwise it is reproduction image that explanation, which is living body faces,.
8. a kind of prevent privacy leakage device based on recognition of face characterized by comprising
Image acquisition process module is set as obtaining original image, the facial image in the original image is extracted, to the people
Face image is handled to enhance clarity；
Reference model module is established, the face three-dimensional reference model of the facial image is arranged to set up, determines the position of eyes
Coordinate；
Image capture module is set as eyes position coordinate being compared with pre-set distance threshold, if
Face characteristic acquisition then is carried out to the face three-dimensional reference model less than the distance threshold, otherwise without acquisition；
Identification point obtains module, is set as parsing the face characteristic, obtains the identification point of the facial image；
Compare identification module, be set as the identification point being compared with inherent feature point, if the identification point with it is described solid
There is characteristic point to compare inconsistent, then carry out screen locking operation, the inherent feature point refers to the had biological characteristic of people's birth, deposits
Storage is in the database.
9. a kind of computer equipment, including memory and processor, it is stored with computer-readable instruction in the memory, it is described
When computer-readable instruction is executed by the processor, so that the processor executes such as any one of claims 1 to 7 right
It is required that it is described based on recognition of face the step of preventing privacy leakage.
10. a kind of storage medium for being stored with computer-readable instruction, the computer-readable instruction is handled by one or more
Device execute when so that one or more processors execute as described in any one of claims 1 to 7 claim based on face
The step of preventing privacy leakage of identification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811214066.5A CN109376518A (en) | 2018-10-18 | 2018-10-18 | Privacy leakage method and relevant device are prevented based on recognition of face |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811214066.5A CN109376518A (en) | 2018-10-18 | 2018-10-18 | Privacy leakage method and relevant device are prevented based on recognition of face |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109376518A true CN109376518A (en) | 2019-02-22 |
Family
ID=65400235
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811214066.5A Pending CN109376518A (en) | 2018-10-18 | 2018-10-18 | Privacy leakage method and relevant device are prevented based on recognition of face |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109376518A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110047591A (en) * | 2019-04-23 | 2019-07-23 | 吉林大学 | One kind is for doctor's posture appraisal procedure in surgical procedures |
| CN110135259A (en) * | 2019-04-15 | 2019-08-16 | 深圳壹账通智能科技有限公司 | Silent formula living body image identification method, device, computer equipment and storage medium |
| CN110135262A (en) * | 2019-04-16 | 2019-08-16 | 深圳壹账通智能科技有限公司 | The anti-peeping processing method of sensitive data, device, equipment and storage medium |
| CN110163814A (en) * | 2019-04-16 | 2019-08-23 | 平安科技（深圳）有限公司 | The method, apparatus and computer equipment of modification picture based on recognition of face |
| CN110955912A (en) * | 2019-10-29 | 2020-04-03 | 平安科技（深圳）有限公司 | Privacy protection method, device and equipment based on image recognition and storage medium thereof |
| CN111310150A (en) * | 2020-01-19 | 2020-06-19 | 湖北工程学院新技术学院 | Security authentication system based on security computer |
| CN111523513A (en) * | 2020-05-09 | 2020-08-11 | 陈正刚 | Working method for personnel-to-home safety verification through big data screening |
| CN111967436A (en) * | 2020-09-02 | 2020-11-20 | 北京猿力未来科技有限公司 | Image processing method and device |
| CN111985448A (en) * | 2020-09-02 | 2020-11-24 | 深圳壹账通智能科技有限公司 | Vehicle image recognition method and device, computer equipment and readable storage medium |
| CN112784323A (en) * | 2020-12-31 | 2021-05-11 | 深圳阜时科技有限公司 | Information protection device and electronic equipment |
| CN113219870A (en) * | 2021-05-07 | 2021-08-06 | 禹焱科技河北有限公司 | Intelligent data acquisition and sharing device for industrial instrument |
| CN114494017A (en) * | 2022-01-25 | 2022-05-13 | 北京至简墨奇科技有限公司 | Method, device, equipment and medium for adjusting DPI (deep packet inspection) image according to scale |
| CN116049868A (en) * | 2022-07-26 | 2023-05-02 | 荣耀终端有限公司 | Privacy protection method, device and storage medium |
| CN116308762A (en) * | 2023-05-19 | 2023-06-23 | 杭州钱袋数字科技有限公司 | Credibility evaluation and trust processing method based on artificial intelligence |
| CN117633867A (en) * | 2023-10-26 | 2024-03-01 | 唐山启奥科技股份有限公司 | Medical image desensitizing method, device, electronic equipment and readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1786980A (en) * | 2005-12-08 | 2006-06-14 | 上海交通大学 | Melthod for realizing searching new position of person's face feature point by tow-dimensional profile |
| CN105260705A (en) * | 2015-09-15 | 2016-01-20 | 西安邦威电子科技有限公司 | Detection method suitable for call receiving and making behavior of driver under multiple postures |
| WO2017219391A1 (en) * | 2016-06-24 | 2017-12-28 | 深圳市唯特视科技有限公司 | Face recognition system based on three-dimensional data |
| WO2018040099A1 (en) * | 2016-08-31 | 2018-03-08 | 深圳市唯特视科技有限公司 | Three-dimensional face reconstruction method based on grayscale and depth information |
| CN108549873A (en) * | 2018-04-19 | 2018-09-18 | 北京华捷艾米科技有限公司 | Three-dimensional face identification method and three-dimensional face recognition system |
-
2018
- 2018-10-18 CN CN201811214066.5A patent/CN109376518A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1786980A (en) * | 2005-12-08 | 2006-06-14 | 上海交通大学 | Melthod for realizing searching new position of person's face feature point by tow-dimensional profile |
| CN105260705A (en) * | 2015-09-15 | 2016-01-20 | 西安邦威电子科技有限公司 | Detection method suitable for call receiving and making behavior of driver under multiple postures |
| WO2017219391A1 (en) * | 2016-06-24 | 2017-12-28 | 深圳市唯特视科技有限公司 | Face recognition system based on three-dimensional data |
| WO2018040099A1 (en) * | 2016-08-31 | 2018-03-08 | 深圳市唯特视科技有限公司 | Three-dimensional face reconstruction method based on grayscale and depth information |
| CN108549873A (en) * | 2018-04-19 | 2018-09-18 | 北京华捷艾米科技有限公司 | Three-dimensional face identification method and three-dimensional face recognition system |
Non-Patent Citations (1)
| Title |
|---|
| 宋顶利;杨炳儒;于复兴;: "关键点匹配三维人脸识别方法", 计算机应用研究, no. 11, 15 November 2010 (2010-11-15) * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110135259A (en) * | 2019-04-15 | 2019-08-16 | 深圳壹账通智能科技有限公司 | Silent formula living body image identification method, device, computer equipment and storage medium |
| CN110135262A (en) * | 2019-04-16 | 2019-08-16 | 深圳壹账通智能科技有限公司 | The anti-peeping processing method of sensitive data, device, equipment and storage medium |
| CN110163814A (en) * | 2019-04-16 | 2019-08-23 | 平安科技（深圳）有限公司 | The method, apparatus and computer equipment of modification picture based on recognition of face |
| CN110047591A (en) * | 2019-04-23 | 2019-07-23 | 吉林大学 | One kind is for doctor's posture appraisal procedure in surgical procedures |
| CN110955912B (en) * | 2019-10-29 | 2023-08-08 | 平安科技（深圳）有限公司 | Privacy protection method, device, equipment and storage medium based on image recognition |
| CN110955912A (en) * | 2019-10-29 | 2020-04-03 | 平安科技（深圳）有限公司 | Privacy protection method, device and equipment based on image recognition and storage medium thereof |
| CN111310150A (en) * | 2020-01-19 | 2020-06-19 | 湖北工程学院新技术学院 | Security authentication system based on security computer |
| CN111523513B (en) * | 2020-05-09 | 2023-08-18 | 深圳市华百安智能技术有限公司 | Working method for carrying out personnel home security verification through big data screening |
| CN111523513A (en) * | 2020-05-09 | 2020-08-11 | 陈正刚 | Working method for personnel-to-home safety verification through big data screening |
| CN111985448A (en) * | 2020-09-02 | 2020-11-24 | 深圳壹账通智能科技有限公司 | Vehicle image recognition method and device, computer equipment and readable storage medium |
| CN111967436A (en) * | 2020-09-02 | 2020-11-20 | 北京猿力未来科技有限公司 | Image processing method and device |
| CN111967436B (en) * | 2020-09-02 | 2024-03-19 | 北京猿力未来科技有限公司 | Image processing method and device |
| CN112784323A (en) * | 2020-12-31 | 2021-05-11 | 深圳阜时科技有限公司 | Information protection device and electronic equipment |
| CN112784323B (en) * | 2020-12-31 | 2024-05-03 | 深圳阜时科技有限公司 | Information protection device and electronic equipment |
| CN113219870A (en) * | 2021-05-07 | 2021-08-06 | 禹焱科技河北有限公司 | Intelligent data acquisition and sharing device for industrial instrument |
| CN113219870B (en) * | 2021-05-07 | 2022-03-08 | 禹焱科技河北有限公司 | Intelligent data acquisition and sharing device for industrial instrument |
| CN114494017A (en) * | 2022-01-25 | 2022-05-13 | 北京至简墨奇科技有限公司 | Method, device, equipment and medium for adjusting DPI (deep packet inspection) image according to scale |
| CN116049868A (en) * | 2022-07-26 | 2023-05-02 | 荣耀终端有限公司 | Privacy protection method, device and storage medium |
| CN116049868B (en) * | 2022-07-26 | 2023-12-08 | 荣耀终端有限公司 | Privacy protection method, device and storage medium |
| CN116308762B (en) * | 2023-05-19 | 2023-08-11 | 杭州钱袋数字科技有限公司 | Credibility evaluation and trust processing method based on artificial intelligence |
| CN116308762A (en) * | 2023-05-19 | 2023-06-23 | 杭州钱袋数字科技有限公司 | Credibility evaluation and trust processing method based on artificial intelligence |
| CN117633867A (en) * | 2023-10-26 | 2024-03-01 | 唐山启奥科技股份有限公司 | Medical image desensitizing method, device, electronic equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109376518A (en) | Privacy leakage method and relevant device are prevented based on recognition of face | |
| CN109558764B (en) | Face recognition method and device and computer equipment | |
| CN111340214B (en) | Method and device for training anti-attack model | |
| US20210012093A1 (en) | Method and apparatus for generating face rotation image | |
| CN107590430A (en) | Biopsy method, device, equipment and storage medium | |
| CN111783748B (en) | Face recognition method and device, electronic equipment and storage medium | |
| TWI738717B (en) | Verification processing method and device based on verification code | |
| US8755607B2 (en) | Method of normalizing a digital image of an iris of an eye | |
| CN105518710B (en) | Video detecting method, video detection system and computer program product | |
| CN108549836A (en) | Reproduction detection method, device, equipment and the readable storage medium storing program for executing of photo | |
| CN110728196A (en) | Face recognition method and device and terminal equipment | |
| EP4033458A2 (en) | Method and apparatus of face anti-spoofing, device, storage medium, and computer program product | |
| CN113449704B (en) | Face recognition model training method and device, electronic equipment and storage medium | |
| CN113128428B (en) | Depth map prediction-based in vivo detection method and related equipment | |
| US11620854B2 (en) | Evaluating the security of a facial recognition system using light projections | |
| CN112802081B (en) | Depth detection method and device, electronic equipment and storage medium | |
| CN111079470A (en) | Method and device for detecting living human face | |
| CN114926892A (en) | Fundus image matching method and system based on deep learning and readable medium | |
| CN107368817B (en) | Face recognition method and device | |
| CN114299569A (en) | Safe face authentication method based on eyeball motion | |
| CN116229528A (en) | Living body palm vein detection method, device, equipment and storage medium | |
| Gilani et al. | Towards large-scale 3D face recognition | |
| Bastias et al. | A method for 3D iris reconstruction from multiple 2D near-infrared images | |
| WO2024198475A1 (en) | Face anti-spoofing recognition method and apparatus, and electronic device and storage medium | |
| Phoka et al. | Image based phishing detection using transfer learning |
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 |