CN108447168A - Recognition of face intelligent unlocking method, system, intelligent door lock and storage medium - Google Patents
Recognition of face intelligent unlocking method, system, intelligent door lock and storage medium Download PDFInfo
- Publication number
- CN108447168A CN108447168A CN201810554708.XA CN201810554708A CN108447168A CN 108447168 A CN108447168 A CN 108447168A CN 201810554708 A CN201810554708 A CN 201810554708A CN 108447168 A CN108447168 A CN 108447168A
- Authority
- CN
- China
- Prior art keywords
- image
- face
- matching
- gender
- result
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/00174—Electronically operated locks; Circuits therefor; Nonmechanical keys therefor, e.g. passive or active electrical keys or other data carriers without mechanical keys
- G07C9/00563—Electronically operated locks; Circuits therefor; Nonmechanical keys therefor, e.g. passive or active electrical keys or other data carriers without mechanical keys using personal physical data of the operator, e.g. finger prints, retinal images, voicepatterns
-
- 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
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Collating Specific Patterns (AREA)
- Lock And Its Accessories (AREA)
Abstract
The invention discloses a kind of recognition of face intelligent unlocking method, system, intelligent door lock and storage mediums, the facial image to be identified of active user and figure's image to be identified are obtained by the intelligent door lock in public place, the facial image to be identified and figure's image to be identified are subjected to convolution algorithm, obtain operation result;The sample convolution characteristic value of operation result gender sample image corresponding with the intelligent door lock is matched, and generates matching result;When matching result is to match consistent, the gender matching corresponding with public place of the gender of active user is judged, and current door lock state is adjusted to open state;The problem of capable of effectively prompting user to enter in correct place, and preventing privacy violation, user itself privacy equity is protected, has ensured public safety, the user experience is improved.
Description
Technical field
The present invention relates to public safety field more particularly to a kind of recognition of face intelligent unlocking method, system, intelligent door locks
And storage medium.
Background technology
The safety problem of public place is increasingly worth noting and payes attention to, and the door lock in existing public place is only capable of providing
Poster guides or prompt, and being unable to effectively prevent dissimilarity, others goes to the room or place of mistake, often will appear into
Wrong room or place cause unnecessary awkward and dispute, and carry out malice there are the personnel of bad attempt and peep, even
It takes on the sly and waits molestations, can not ensure itself privacy and equity of user.
Invention content
The main purpose of the present invention is to provide a kind of recognition of face intelligent unlocking method, system, intelligent door lock and storages
Medium, it is intended to correct prompting can not be provided by solving the door lock in public place in the prior art, to cause user itself privacy or
The technical issues of equity is invaded.
To achieve the above object, the present invention provides a kind of recognition of face intelligent unlocking method, and the recognition of face is intelligently opened
Locking method includes the following steps:
The intelligent door lock in public place obtains the facial image to be identified of active user and figure's image to be identified, will be described
Facial image to be identified and figure's image to be identified carry out convolution algorithm, obtain operation result;
By the sample convolution characteristic value progress of operation result gender sample image corresponding with the intelligent door lock
Match, and generates matching result;
When matching result is to match consistent, the gender matching corresponding with public place of the gender of active user is judged, and
Current door lock state is adjusted to open state.
Preferably, the facial image to be identified for obtaining active user and figure's image to be identified, will be described to be identified
Facial image and figure's image to be identified carry out convolution algorithm, obtain operation result, specifically include:
The facial image to be identified and figure's image to be identified are split as to the face characteristic image of preset quantity respectively
With aspectual character image, and obtain split after face characteristic image and the corresponding location information of aspectual character image;
The face characteristic image and aspectual character image are subjected to bilinear interpolation processing, obtaining that treated, face is special
Levy image and aspectual character image;
Treated face characteristic image and aspectual character image are subjected to cross-layer convolution algorithm, obtain the convolution fortune of each layer
Calculate result and cross-layer feature difference;
The convolution algorithm result of each layer is compensated according to the cross-layer feature difference, by the convolution of each layer after compensation
Operation result is spliced according to the location information, using the convolution algorithm result of spliced each layer as operation result.
Preferably, described that treated face characteristic image and aspectual character image are subjected to cross-layer convolution algorithm, it obtains
The convolution algorithm result and cross-layer feature difference of each layer, specifically include:
Treated face characteristic image and aspectual character image are subjected to cross-layer convolution algorithm, the feature for obtaining each layer is reflected
Penetrate the cross-layer feature difference of figure and each layer;
The Feature Mapping figure of each layer is overlapped, the corresponding each feature vector of the Feature Mapping figure after being superimposed;
The corresponding covariance matrix of each feature vector is calculated, principal component analysis is carried out to the covariance matrix, is dropped
Feature vector after dimension, using the feature vector after dimensionality reduction as the convolution algorithm result of each layer.
Preferably, the sample convolution by operation result gender sample image corresponding with the intelligent door lock is special
Value indicative is matched, and generates matching result, is specifically included:
Obtain the corresponding sample convolution of corresponding gender sample image that the intelligent door lock stores in the preset database
Characteristic value;
Characteristic value in the operation result is subjected to Dynamic Matching with the sample characteristics, and generates matching result.
Preferably, the characteristic value by the operation result carries out Dynamic Matching with the sample characteristics, and raw
At matching result, specifically include:
Obtain the corresponding sample characteristics of gender sample image of different predetermined angles;
Successively respectively by the operation on multiple default human face characteristic point positions and multiple default physical trait point positions
As a result the characteristic value in is compared with each sample characteristics, and generates multiple characteristic value comparison results;
Matching result is generated according to multiple characteristic value comparison results.
Preferably, described to generate matching result according to multiple characteristic value comparison results, it specifically includes:
Multiple characteristic value comparison results are weighted average computation according to default weight proportion, obtain images match degree;
Matching result is generated according to described image matching degree.
Preferably, described when matching result is to match consistent, judge that the gender of active user is corresponding with public place
Gender matches, and current door lock state is adjusted to open state, specifically includes:
When described image matching degree is more than pre-set image matching threshold, it is consistent to match to judge the matching result, sentences
Determine the gender gender matching corresponding with public place of active user, and current door lock state is adjusted to open state;
Correspondingly, when described image matching degree is less than or equal to pre-set image matching threshold, judge that the matching result is
It mismatches, judges that the gender of active user gender corresponding with public place mismatches, and current door lock state is adjusted to close
Closed state, and generate prompt message and prompted.
In addition, to achieve the above object, the present invention also proposes that a kind of intelligent door lock, the intelligent door lock include:Memory,
Processor and the recognition of face intelligent unlocking program that is stored on the memory and can run on the processor, the people
Face identification intelligent unlocking program is arranged for carrying out the step of recognition of face intelligent unlocking method as described above.
In addition, to achieve the above object, the present invention also proposes a kind of storage medium, and face is stored on the storage medium
Identification intelligent unlocking program, the recognition of face intelligent unlocking program realize that face as described above is known when being executed by processor
The step of other intelligent unlocking method.
In addition, to achieve the above object, the present invention also provides a kind of recognition of face intelligent unlocking system, the recognitions of face
Intelligent unlocking system includes:
Computing module, the facial image to be identified for obtaining active user and figure's image to be identified, wait knowing by described
Others' face image and figure's image to be identified carry out convolution algorithm, obtain operation result;
Matching module is used for the sample convolution of operation result gender sample image corresponding with the intelligent door lock
Characteristic value is matched, and generates matching result;
Unlocking module, for when matching result is to match consistent, judging that the gender of active user is corresponding with public place
Gender matching, and current door lock state is adjusted to open state.
Recognition of face intelligent unlocking method proposed by the present invention obtains active user's by the intelligent door lock in public place
The facial image to be identified and figure's image to be identified are carried out convolution fortune by facial image to be identified and figure's image to be identified
It calculates, obtains operation result;By the sample convolution feature of operation result gender sample image corresponding with the intelligent door lock
Value is matched, and generates matching result;When matching result is to match consistent, the gender of active user and public place are judged
Corresponding gender matching, and current door lock state is adjusted to open state;User can be effectively prompted to enter correct field
In institute, and the problem of prevent privacy violation, user itself privacy equity is protected, has ensured public safety, improves user's body
It tests.
Description of the drawings
Fig. 1 is the intelligent door lock structural schematic diagram for the hardware running environment that the embodiment of the present invention is related to;
Fig. 2 is the flow diagram of face identification intelligent method for unlocking first embodiment of the present invention;
Fig. 3 is the flow diagram of face identification intelligent method for unlocking second embodiment of the present invention;
Fig. 4 is the flow diagram of face identification intelligent method for unlocking 3rd embodiment of the present invention;
Fig. 5 is the functional block diagram of face identification intelligent unlocking system first embodiment of the present invention;
Fig. 6 is the functional block diagram of face identification intelligent unlocking system second embodiment of the present invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific implementation mode
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
The solution of the embodiment of the present invention is mainly:The present invention obtains active user by the intelligent door lock in public place
Facial image to be identified and figure's image to be identified, the facial image to be identified and figure's image to be identified are subjected to convolution
Operation obtains operation result;The sample convolution of operation result gender sample image corresponding with the intelligent door lock is special
Value indicative is matched, and generates matching result;When matching result is to match consistent, the gender of active user and public field are judged
Corresponding gender matching, and current door lock state is adjusted to open state;User can be effectively prompted to enter correct
In place, and the problem of prevent privacy violation, user itself privacy equity is protected, public safety has been ensured, has improved user
Experience, the door lock for solving public place in the prior art can not provide correct prompting, to cause user itself privacy or power
The technical issues of benefit is invaded.
Referring to Fig.1, Fig. 1 is the intelligent door lock structural schematic diagram for the hardware running environment that the embodiment of the present invention is related to.
As shown in Figure 1, the intelligent door lock may include:Processor 1001, such as CPU, communication bus 1002, user's termination
Mouth 1003, network interface 1004, memory 1005.Wherein, communication bus 1002 is logical for realizing the connection between these components
Letter.User's end interface 1003 may include display screen (Display), input unit such as keyboard (Keyboard), optional user
End interface 1003 can also include standard wireline interface and wireless interface.Network interface 1004 may include optionally standard
Wireline interface, wireless interface (such as WI-FI interfaces).Memory 1005 can be high-speed RAM memory, can also be stable deposit
Reservoir (non-volatile memory), such as magnetic disk storage.Memory 1005 optionally can also be independently of aforementioned place
Manage the storage device of device 1001.
It will be understood by those skilled in the art that intelligent door lock structure shown in Fig. 1 is not constituted to the intelligent door lock
It limits, may include either combining certain components or different components arrangement than illustrating more or fewer components.
As shown in Figure 1, as may include operating system, network communication mould in a kind of memory 1005 of storage medium
Block, user terminal interface module and recognition of face intelligent unlocking program.
Intelligent door lock of the present invention calls the recognition of face intelligent unlocking journey stored in memory 1005 by processor 1001
Sequence, and execute following operation:
The facial image to be identified of active user and figure's image to be identified are obtained, by the facial image to be identified and is waited for
It identifies that figure's image carries out convolution algorithm, obtains operation result;
By the sample convolution characteristic value progress of operation result gender sample image corresponding with the intelligent door lock
Match, and generates matching result;
When matching result is to match consistent, the gender matching corresponding with public place of the gender of active user is judged, and
Current door lock state is adjusted to open state.
Further, processor 1001 can call the recognition of face intelligent unlocking program stored in memory 1005, also
Execute following operation:
The facial image to be identified and figure's image to be identified are split as to the face characteristic image of preset quantity respectively
With aspectual character image, and obtain split after face characteristic image and the corresponding location information of aspectual character image;
The face characteristic image and aspectual character image are subjected to bilinear interpolation processing, obtaining that treated, face is special
Levy image and aspectual character image;
Treated face characteristic image and aspectual character image are subjected to cross-layer convolution algorithm, obtain the convolution fortune of each layer
Calculate result and cross-layer feature difference;
The convolution algorithm result of each layer is compensated according to the cross-layer feature difference, by the convolution of each layer after compensation
Operation result is spliced according to the location information, using the convolution algorithm result of spliced each layer as operation result.
Further, processor 1001 can call the recognition of face intelligent unlocking program stored in memory 1005, also
Execute following operation:
Treated face characteristic image and aspectual character image are subjected to cross-layer convolution algorithm, the feature for obtaining each layer is reflected
Penetrate the cross-layer feature difference of figure and each layer;
The Feature Mapping figure of each layer is overlapped, the corresponding each feature vector of the Feature Mapping figure after being superimposed;
The corresponding covariance matrix of each feature vector is calculated, principal component analysis is carried out to the covariance matrix, is dropped
Feature vector after dimension, using the feature vector after dimensionality reduction as the convolution algorithm result of each layer.
Further, processor 1001 can call the recognition of face intelligent unlocking program stored in memory 1005, also
Execute following operation:
Obtain the corresponding sample convolution of corresponding gender sample image that the intelligent door lock stores in the preset database
Characteristic value;
Characteristic value in the operation result is subjected to Dynamic Matching with the sample characteristics, and generates matching result.
Further, processor 1001 can call the recognition of face intelligent unlocking program stored in memory 1005, also
Execute following operation:
Obtain the corresponding sample characteristics of gender sample image of different predetermined angles;
Successively respectively by the operation on multiple default human face characteristic point positions and multiple default physical trait point positions
As a result the characteristic value in is compared with each sample characteristics, and generates multiple characteristic value comparison results;
Matching result is generated according to multiple characteristic value comparison results.
Further, processor 1001 can call the recognition of face intelligent unlocking program stored in memory 1005, also
Execute following operation:
Multiple characteristic value comparison results are weighted average computation according to default weight proportion, obtain images match degree;
Matching result is generated according to described image matching degree.
Further, processor 1001 can call the recognition of face intelligent unlocking program stored in memory 1005, also
Execute following operation:
When described image matching degree is more than pre-set image matching threshold, it is consistent to match to judge the matching result, sentences
Determine the gender gender matching corresponding with public place of active user, and current door lock state is adjusted to open state;
Correspondingly, when described image matching degree is less than or equal to pre-set image matching threshold, judge that the matching result is
It mismatches, judges that the gender of active user gender corresponding with public place mismatches, and current door lock state is adjusted to close
Closed state, and generate prompt message and prompted.
The present embodiment through the above scheme, the face figure to be identified of active user is obtained by the intelligent door lock in public place
The facial image to be identified and figure's image to be identified are carried out convolution algorithm, obtain operation by picture and figure's image to be identified
As a result;The sample convolution characteristic value of operation result gender sample image corresponding with the intelligent door lock is matched,
And generate matching result;When matching result is to match consistent, the gender of active user gender corresponding with public place is judged
Matching, and current door lock state is adjusted to open state;It can effectively prompt user to enter in correct place, and prevent
The problem of privacy violation, protects user itself privacy equity, has ensured public safety, the user experience is improved.
Based on above-mentioned hardware configuration, face identification intelligent method for unlocking embodiment of the present invention is proposed.
It is the flow diagram of face identification intelligent method for unlocking first embodiment of the present invention with reference to Fig. 2, Fig. 2.
In the first embodiment, the recognition of face intelligent unlocking method includes the following steps:
The intelligent door lock in step S10, public place obtains the facial image to be identified and figure to be identified figure of active user
The facial image to be identified and figure's image to be identified are carried out convolution algorithm, obtain operation result by picture.
It should be noted that the public place is public place, it is different from the public place of home address, the public affairs
Can be the fitting room of each market or shops with place, or each market or the communal toilet of shops can also be certainly
The public place of other activity venues, the present embodiment do not limit this.
It is understood that obtaining the facial image to be identified of active user by the intelligent door lock in public place and waiting knowing
Complicated variant state image, and then the facial image to be identified and figure's image to be identified are subjected to convolution algorithm, operation result is obtained,
Effectively the face of active user and figure can accurately be identified, improve the accuracy of recognition of face.
Step S20, by the sample convolution feature of operation result gender sample image corresponding with the intelligent door lock
Value is matched, and generates matching result.
It should be understood that after obtaining the operation result by convolution algorithm, by the operation result and the intelligence
The sample convolution characteristic value of the corresponding gender sample image of door lock is matched, and matching result, the gender sample can be generated
Image can be the sample convolution characteristic value for being stored in advance in the local sample image for verifying user's gender of intelligent door lock,
It can also be the sample convolution characteristic value for the sample image for verifying user's gender for being stored in cloud server, may be used also certainly
To be the sample convolution characteristic value of the gender samples pictures stored otherwise, such as it is stored in the gender of cloud database
The sample convolution characteristic value of samples pictures, the present embodiment do not limit this;By the way that result and the gender sample will be transported
The sample convolution characteristic value of picture is matched, and can quickly determine whether are the operation result and the sample convolution characteristic value
Matching, and then quickly determine the gender of the active user.
Step S30, when matching result is to match consistent, judge the gender of active user gender corresponding with public place
Matching, and current door lock state is adjusted to open state.
It should be noted that when the matching result is consistent, judge that the gender of active user is corresponding with public place
Gender matches, i.e. the active user public place to be entered is correct, and current door lock state is adjusted to open state at this time,
It is accordingly prompted if incorrect, and forbids unlocking.
In the concrete realization, when user is prepared to enter into lavatory or fitting room, the people to be identified of quick obtaining active user
Face image and figure's image to be identified simultaneously carry out gender matching, and when successful match, opening door lock allows active user to enter, when
When matching unsuccessful, door lock is forbidden to open, and prompt the correct lavatory position of user or fitting room position, quickly and effectively guided
User goes to correct place, is gone to the wrong way so as to avoid lavatory or fitting room goes to the wrong way caused awkward and unnecessary dispute, and
The generation for preventing the molestations such as the personnel of bad motivation peep or take on the sly, has ensured privacy of user and public peace
Entirely.
The present embodiment through the above scheme, the face figure to be identified of active user is obtained by the intelligent door lock in public place
The facial image to be identified and figure's image to be identified are carried out convolution algorithm, obtain operation by picture and figure's image to be identified
As a result;The sample convolution characteristic value of operation result gender sample image corresponding with the intelligent door lock is matched,
And generate matching result;When matching result is to match consistent, the gender of active user gender corresponding with public place is judged
Matching, and current door lock state is adjusted to open state;It can effectively prompt user to enter in correct place, and prevent
The problem of privacy violation, protects user itself privacy equity, has ensured public safety, the user experience is improved.
Further, Fig. 3 is the flow diagram of face identification intelligent method for unlocking second embodiment of the present invention, such as Fig. 3
It is shown, face identification intelligent method for unlocking second embodiment of the present invention is proposed based on first embodiment, it is in the present embodiment, described
Step S10, specifically includes following steps:
Step S11, the facial image to be identified and figure's image to be identified are split as to the face of preset quantity respectively
Characteristic image and aspectual character image, and obtain the face characteristic image after splitting and the corresponding position letter of aspectual character image
Breath.
It should be noted that the preset quantity is preset value, can be the preset quantity can pass through
The fixed value that mass data training obtains can also be to be corresponded to according to the facial image to be identified and figure's image to be identified
Value or the value that is adaptively adjusted, can also be that quantity determining by other means, the present embodiment do not limit this certainly
System;By the facial image to be identified and figure's image to be identified are split as respectively preset quantity face characteristic image and
Aspectual character image can facilitate to the facial image to be identified and figure's image to be identified progress convolution algorithm, and and
When obtain split after face characteristic image and the corresponding location information of aspectual character image, can be subsequent image processing do standard
It is standby.
Step S12, the face characteristic image and aspectual character image are subjected to bilinear interpolation processing, after being handled
Face characteristic image and aspectual character image.
It is understood that the face characteristic image and aspectual character image are subjected to bilinear interpolation processing, it can
The face characteristic image and aspectual character image are effectively amplified and ensure the standard of each tile location after image magnification
True property, to be conducive to the image procossing of next step.
Step S13, treated face characteristic image and aspectual character image are subjected to cross-layer convolution algorithm, obtain each layer
Convolution algorithm result and cross-layer feature difference.
It should be understood that by bilinear interpolation treated the face characteristic image and aspectual character image carry out across
Layer convolution algorithm, can obtain the operation result of each layer, to improve the accuracy of picture recognition, further increase matching respectively
Accuracy.
Further, the step S13 specifically includes following steps:
Treated face characteristic image and aspectual character image are subjected to cross-layer convolution algorithm, the feature for obtaining each layer is reflected
Penetrate the cross-layer feature difference of figure and each layer;
The Feature Mapping figure of each layer is overlapped, the corresponding each feature vector of the Feature Mapping figure after being superimposed;
The corresponding covariance matrix of each feature vector is calculated, principal component analysis is carried out to the covariance matrix, is dropped
Feature vector after dimension, using the feature vector after dimensionality reduction as the convolution algorithm result of each layer.
It is understood that by the way that treated face characteristic image and aspectual character image are carried out cross-layer convolution fortune
Calculate, can obtain the set and the difference that is generated in cross-layer operation of each layer of the Feature Mapping figure after this word, i.e., each layer across
Layer feature difference;By the superposition of the Feature Mapping figure to each layer, each pixel pair in the Feature Mapping figure of each layer can be obtained
The feature vector answered, to obtain the eigenmatrix being made of all feature vectors according to each feature vector, to the feature square
The average value of feature vector in battle array is filtered out, and the corresponding covariance matrix of each feature vector is obtained, to the covariance square
Battle array carry out principal component analysis, obtain dimensionality reduction after feature vector, i.e., with the face characteristic image and aspectual character image size
Identical feature vector, i.e. each feature vector in eigenmatrix after dimensionality reduction, using the feature vector after dimensionality reduction as each layer
Convolution algorithm is as a result, ensure that the accuracy of characteristics of image identification.
Step S14, the convolution algorithm result of each layer is compensated according to the cross-layer feature difference, it will be each after compensation
The convolution algorithm result of layer is spliced according to the location information, using the convolution algorithm result of spliced each layer as operation
As a result.
It should be understood that being compensated to the convolution algorithm result of each layer according to the cross-layer feature difference, will compensate
The convolution algorithm result of each layer afterwards is spliced according to the location information, and the convolution algorithm result of spliced each layer is made
For operation result, the accuracy of characteristics of image identification can be improved, and then improves the accuracy of images match.
The present embodiment through the above scheme, by the way that the facial image to be identified and figure's image to be identified are split respectively
For the face characteristic image and aspectual character image of preset quantity, and obtain face characteristic image and aspectual character figure after splitting
As corresponding location information;The face characteristic image and aspectual character image are subjected to bilinear interpolation processing, handled
Face characteristic image afterwards and aspectual character image;Treated face characteristic image and aspectual character image are subjected to cross-layer volume
Product operation, obtains the convolution algorithm result and cross-layer feature difference of each layer;According to the cross-layer feature difference to the convolution of each layer
Operation result compensates, and the convolution algorithm result of each layer after compensation is spliced according to the location information, will be spliced
The convolution algorithm result of each layer afterwards can further promote the accuracy of image recognition, improve image as operation result
Matched accuracy, and then user can effectively be prompted to enter in correct place, and the problem of prevent privacy violation, protection
User itself privacy equity, has ensured public safety, the user experience is improved.
Further, Fig. 4 is the flow diagram of face identification intelligent method for unlocking 3rd embodiment of the present invention, such as Fig. 4
It is shown, face identification intelligent method for unlocking 3rd embodiment of the present invention is proposed based on second embodiment, it is in the present embodiment, described
Step S20 specifically includes following steps:
Step S21, it is corresponding to obtain the corresponding gender sample image that the intelligent door lock stores in the preset database
Sample convolution characteristic value.
It should be noted that the corresponding sample convolution characteristic value of the gender sample image is to pre-set by big
The sample convolution characteristic value that data training obtains is measured, being locally stored in the intelligent door lock can be arranged in the presetting database
In space, it can also be to be arranged in server beyond the clouds, can also be provided in the memory space of other forms, the present embodiment
This is not limited.
Step S22, the characteristic value in the operation result is subjected to Dynamic Matching, and generation with the sample characteristics
With result.
It is understood that the characteristic value in the operation result is subjected to Dynamic Matching with the sample characteristics, it can
To generate matching result, i.e., the matching degree of characteristic value and the sample characteristics in the more described operation result.
Further, the step S22 specifically includes following steps:
Obtain the corresponding sample characteristics of gender sample image of different predetermined angles;
Successively respectively by the operation on multiple default human face characteristic point positions and multiple default physical trait point positions
As a result the characteristic value in is compared with each sample characteristics, and generates multiple characteristic value comparison results;
Matching result is generated according to multiple characteristic value comparison results.
It should be understood that the face of different angle and figure can have differences, therefore by obtaining different predetermined angles
Face and figure can the corresponding sample characteristics of acquired other sample image, successively in multiple default human face characteristic point positions
The characteristic value in the operation result is compared with each sample characteristics respectively on multiple default physical trait point positions,
And generate multiple characteristic value comparison results, by multiple predetermined angles obtain characteristic value and multiple default human face characteristic point positions and
The operation result acquired on multiple default physical trait point positions is compared, and matched accuracy can be further promoted, from
And improve the success rate of gender certification.
Further, described to generate matching result according to multiple characteristic value comparison results, it specifically includes:
Multiple characteristic value comparison results are weighted average computation according to default weight proportion, obtain images match degree;
Matching result is generated according to described image matching degree.
It is understood that the default weight proportion is pre-set weight proportion corresponding with multiple characteristic values,
By being weighted average computation to multiple characteristic values, accurate characteristic value can be obtained, to obtain images match degree,
Matching result is generated according to described image matching degree.
Correspondingly, the step S30 specifically includes following steps:
When described image matching degree is more than pre-set image matching threshold, it is consistent to match to judge the matching result, sentences
Determine the gender gender matching corresponding with public place of active user, and current door lock state is adjusted to open state.
It is understood that the pre-set image matching angle value is pre-set value, can be instructed by mass data
Practice the value obtained, can also be the value that technical staff is voluntarily arranged according to regular job experience, can also be other modes certainly
Determining value, the present embodiment do not limit this;Described image matching degree is compared with pre-set image matching angle value, when
When described image matching degree is more than pre-set image matching threshold, it is consistent to match to judge the matching result, i.e., the described current use
The gender matching corresponding with public place of the gender at family, open state is adjusted to by current door lock state.
Correspondingly, when described image matching degree is less than or equal to pre-set image matching threshold, judge that the matching result is
It mismatches, judges that the gender of active user gender corresponding with public place mismatches, and current door lock state is adjusted to close
Closed state, and generate prompt message and prompted.
It should be understood that described image matching degree is compared with pre-set image matching angle value, when described image
When being less than or equal to pre-set image matching threshold with degree, judge the matching result for mismatch, i.e., the gender of the described active user
Gender corresponding with public place mismatches, and will adjust current door lock state and be in off state, and generates prompt message progress
Prompt, the mode of prompt can be prompted by screen display subtitle, can also be that can also be certainly logical by auditory tone cues
Other modes prompt is crossed, the present embodiment does not limit this.
The corresponding property that the present embodiment stores in the preset database through the above scheme, by obtaining the intelligent door lock
The corresponding sample convolution characteristic value of other sample image, obtains the corresponding sample characteristics of gender sample image of different predetermined angles
Value;It successively respectively will be in the operation result on multiple default human face characteristic point positions and multiple default physical trait point positions
Characteristic value be compared with each sample characteristics, and generate multiple characteristic value comparison results;By multiple characteristic value comparison results
It is weighted average computation according to default weight proportion, obtains images match degree;Matching knot is generated according to described image matching degree
Fruit further improves matched accuracy, saves match time, and then user can effectively be prompted to enter correct field
In institute, and the problem of prevent privacy violation, user itself privacy equity is protected, has ensured public safety, improves user's body
It tests.
Based on above-described embodiment, the present invention further provides a kind of recognition of face intelligent unlocking systems.
It is the functional block diagram of face identification intelligent unlocking system first embodiment of the present invention with reference to Fig. 5, Fig. 5.
In face identification intelligent unlocking system first embodiment of the present invention, which includes:
Computing module 10, the intelligent door lock for public place obtain the facial image to be identified of active user and to be identified
The facial image to be identified and figure's image to be identified are carried out convolution algorithm, obtain operation result by figure's image.
It should be noted that the public place is public place, it is different from the public place of home address, the public affairs
Can be the fitting room of each market or shops with place, or each market or the communal toilet of shops can also be certainly
The public place of other activity venues, the present embodiment do not limit this.
It is understood that obtaining the facial image to be identified of active user by the intelligent door lock in public place and waiting knowing
Complicated variant state image, and then the facial image to be identified and figure's image to be identified are subjected to convolution algorithm, operation result is obtained,
Effectively the face of active user and figure can accurately be identified, improve the accuracy of recognition of face.
Matching module 20, for rolling up the sample of operation result gender sample image corresponding with the intelligent door lock
Product characteristic value is matched, and generates matching result.
It should be understood that after obtaining the operation result by convolution algorithm, by the operation result and the intelligence
The sample convolution characteristic value of the corresponding gender sample image of door lock is matched, and matching result, the gender sample can be generated
Image can be the sample convolution characteristic value for being stored in advance in the local sample image for verifying user's gender of intelligent door lock,
It can also be the sample convolution characteristic value for the sample image for verifying user's gender for being stored in cloud server, may be used also certainly
To be the sample convolution characteristic value of the gender samples pictures stored otherwise, such as it is stored in the gender of cloud database
The sample convolution characteristic value of samples pictures, the present embodiment do not limit this;By the way that result and the gender sample will be transported
The sample convolution characteristic value of picture is matched, and can quickly determine whether are the operation result and the sample convolution characteristic value
Matching, and then quickly determine the gender of the active user.
Unlocking module 30, for when matching result is to match consistent, judging the gender of active user and public place pair
The gender matching answered, and current door lock state is adjusted to open state.
It should be noted that when the matching result is consistent, judge that the gender of active user is corresponding with public place
Gender matches, i.e. the active user public place to be entered is correct, and current door lock state is adjusted to open state at this time,
It is accordingly prompted if incorrect, and forbids unlocking.
In the concrete realization, when user is prepared to enter into lavatory or fitting room, the people to be identified of quick obtaining active user
Face image and figure's image to be identified simultaneously carry out gender matching, and when successful match, opening door lock allows active user to enter, when
When matching unsuccessful, door lock is forbidden to open, and prompt the correct lavatory position of user or fitting room position, quickly and effectively guided
User goes to correct place, is gone to the wrong way so as to avoid lavatory or fitting room goes to the wrong way caused awkward and unnecessary dispute, and
The generation for preventing the molestations such as the personnel of bad motivation peep or take on the sly, has ensured privacy of user and public peace
Entirely.
The present embodiment through the above scheme, the face figure to be identified of active user is obtained by the intelligent door lock in public place
The facial image to be identified and figure's image to be identified are carried out convolution algorithm, obtain operation by picture and figure's image to be identified
As a result;The sample convolution characteristic value of operation result gender sample image corresponding with the intelligent door lock is matched,
And generate matching result;When matching result is to match consistent, the gender of active user gender corresponding with public place is judged
Matching, and current door lock state is adjusted to open state;It can effectively prompt user to enter in correct place, and prevent
The problem of privacy violation, protects user itself privacy equity, has ensured public safety, the user experience is improved.
It is the functional block diagram of face identification intelligent unlocking system second embodiment of the present invention with reference to Fig. 6, Fig. 6.
Iing is proposed that face identification intelligent of the present invention is unlocked based on face identification intelligent unlocking system first embodiment of the present invention is
System second embodiment, in face identification intelligent unlocking system second embodiment of the present invention, the face identification intelligent unlocking system
Including:
Computing module 10 includes:
Module 11 is split, for the facial image to be identified and figure's image to be identified to be split as preset quantity respectively
Face characteristic image and aspectual character image, and obtain split after face characteristic image and the corresponding position of aspectual character image
Confidence ceases.
It should be noted that the preset quantity is preset value, can be the preset quantity can pass through
The fixed value that mass data training obtains can also be to be corresponded to according to the facial image to be identified and figure's image to be identified
Value or the value that is adaptively adjusted, can also be that quantity determining by other means, the present embodiment do not limit this certainly
System;By the facial image to be identified and figure's image to be identified are split as respectively preset quantity face characteristic image and
Aspectual character image can facilitate to the facial image to be identified and figure's image to be identified progress convolution algorithm, and and
When obtain split after face characteristic image and the corresponding location information of aspectual character image, can be subsequent image processing do standard
It is standby.
Interpolation processing module 12, for carrying out the face characteristic image and aspectual character image at bilinear interpolation
Reason obtains treated face characteristic image and aspectual character image.
It is understood that the face characteristic image and aspectual character image are subjected to bilinear interpolation processing, it can
The face characteristic image and aspectual character image are effectively amplified and ensure the standard of each tile location after image magnification
True property, to be conducive to the image procossing of next step.
Cross-layer computing module 13, for treated face characteristic image and aspectual character image to be carried out cross-layer convolution fortune
It calculates, obtains the convolution algorithm result and cross-layer feature difference of each layer.
It should be understood that by bilinear interpolation treated the face characteristic image and aspectual character image carry out across
Layer convolution algorithm, can obtain the operation result of each layer, to improve the accuracy of picture recognition, further increase matching respectively
Accuracy.
Concatenation module 14 is compensated, for being compensated to the convolution algorithm result of each layer according to the cross-layer feature difference,
The convolution algorithm result of each layer after compensation is spliced according to the location information, by the convolution algorithm of spliced each layer
As a result it is used as operation result.
It should be understood that being compensated to the convolution algorithm result of each layer according to the cross-layer feature difference, will compensate
The convolution algorithm result of each layer afterwards is spliced according to the location information, and the convolution algorithm result of spliced each layer is made
For operation result, the accuracy of characteristics of image identification can be improved, and then improves the accuracy of images match.
The present embodiment through the above scheme, by the way that the facial image to be identified and figure's image to be identified are split respectively
For the face characteristic image and aspectual character image of preset quantity, and obtain face characteristic image and aspectual character figure after splitting
As corresponding location information;The face characteristic image and aspectual character image are subjected to bilinear interpolation processing, handled
Face characteristic image afterwards and aspectual character image;Treated face characteristic image and aspectual character image are subjected to cross-layer volume
Product operation, obtains the convolution algorithm result and cross-layer feature difference of each layer;According to the cross-layer feature difference to the convolution of each layer
Operation result compensates, and the convolution algorithm result of each layer after compensation is spliced according to the location information, will be spliced
The convolution algorithm result of each layer afterwards can further promote the accuracy of image recognition, improve image as operation result
Matched accuracy, and then user can effectively be prompted to enter in correct place, and the problem of prevent privacy violation, protection
User itself privacy equity, has ensured public safety, the user experience is improved.
In addition, the embodiment of the present invention also proposes a kind of storage medium, recognition of face intelligence is stored on the storage medium
Unlocking program realizes following operation when the recognition of face intelligent unlocking program is executed by processor:
The intelligent door lock in public place obtains the facial image to be identified of active user and figure's image to be identified, will be described
Facial image to be identified and figure's image to be identified carry out convolution algorithm, obtain operation result;
By the sample convolution characteristic value progress of operation result gender sample image corresponding with the intelligent door lock
Match, and generates matching result;
When matching result is to match consistent, the gender matching corresponding with public place of the gender of active user is judged, and
Current door lock state is adjusted to open state.
Further, following operation is also realized when the recognition of face intelligent unlocking program is executed by processor:
The facial image to be identified and figure's image to be identified are split as to the face characteristic image of preset quantity respectively
With aspectual character image, and obtain split after face characteristic image and the corresponding location information of aspectual character image;
The face characteristic image and aspectual character image are subjected to bilinear interpolation processing, obtaining that treated, face is special
Levy image and aspectual character image;
Treated face characteristic image and aspectual character image are subjected to cross-layer convolution algorithm, obtain the convolution fortune of each layer
Calculate result and cross-layer feature difference;
The convolution algorithm result of each layer is compensated according to the cross-layer feature difference, by the convolution of each layer after compensation
Operation result is spliced according to the location information, using the convolution algorithm result of spliced each layer as operation result.
Further, following operation is also realized when the recognition of face intelligent unlocking program is executed by processor:
Treated face characteristic image and aspectual character image are subjected to cross-layer convolution algorithm, the feature for obtaining each layer is reflected
Penetrate the cross-layer feature difference of figure and each layer;
The Feature Mapping figure of each layer is overlapped, the corresponding each feature vector of the Feature Mapping figure after being superimposed;
The corresponding covariance matrix of each feature vector is calculated, principal component analysis is carried out to the covariance matrix, is dropped
Feature vector after dimension, using the feature vector after dimensionality reduction as the convolution algorithm result of each layer.
Further, following operation is also realized when the recognition of face intelligent unlocking program is executed by processor:
Obtain the corresponding sample convolution of corresponding gender sample image that the intelligent door lock stores in the preset database
Characteristic value;
Characteristic value in the operation result is subjected to Dynamic Matching with the sample characteristics, and generates matching result.
Further, following operation is also realized when the recognition of face intelligent unlocking program is executed by processor:
Obtain the corresponding sample characteristics of gender sample image of different predetermined angles;
Successively respectively by the operation on multiple default human face characteristic point positions and multiple default physical trait point positions
As a result the characteristic value in is compared with each sample characteristics, and generates multiple characteristic value comparison results;
Matching result is generated according to multiple characteristic value comparison results.
Further, following operation is also realized when the recognition of face intelligent unlocking program is executed by processor:
Multiple characteristic value comparison results are weighted average computation according to default weight proportion, obtain images match degree;
Matching result is generated according to described image matching degree.
Further, following operation is also realized when the recognition of face intelligent unlocking program is executed by processor:
When described image matching degree is more than pre-set image matching threshold, it is consistent to match to judge the matching result, sentences
Determine the gender gender matching corresponding with public place of active user, and current door lock state is adjusted to open state;
Correspondingly, when described image matching degree is less than or equal to pre-set image matching threshold, judge that the matching result is
It mismatches, judges that the gender of active user gender corresponding with public place mismatches, and current door lock state is adjusted to close
Closed state, and generate prompt message and prompted.
The present embodiment through the above scheme, the face figure to be identified of active user is obtained by the intelligent door lock in public place
The facial image to be identified and figure's image to be identified are carried out convolution algorithm, obtain operation by picture and figure's image to be identified
As a result;The sample convolution characteristic value of operation result gender sample image corresponding with the intelligent door lock is matched,
And generate matching result;When matching result is to match consistent, the gender of active user gender corresponding with public place is judged
Matching, and current door lock state is adjusted to open state;It can effectively prompt user to enter in correct place, and prevent
The problem of privacy violation, protects user itself privacy equity, has ensured public safety, the user experience is improved.
It should be noted that herein, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that process, method, article or system including a series of elements include not only those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or system institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including this
There is also other identical elements in the process of element, method, article or system.
The embodiments of the present invention are for illustration only, can not represent the quality of embodiment.
It these are only the preferred embodiment of the present invention, be not intended to limit the scope of the invention, it is every to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of recognition of face intelligent unlocking method, which is characterized in that the recognition of face intelligent unlocking method includes:
The intelligent door lock in public place obtains the facial image to be identified of active user and figure's image to be identified, waits knowing by described
Others' face image and figure's image to be identified carry out convolution algorithm, obtain operation result;
The sample convolution characteristic value of operation result gender sample image corresponding with the intelligent door lock is matched, and
Generate matching result;
When matching result is to match consistent, the gender matching corresponding with public place of the gender of active user is judged, and will work as
Front door lock status is adjusted to open state.
2. recognition of face intelligent unlocking method as described in claim 1, which is characterized in that the acquisition active user's waits knowing
The facial image to be identified and figure's image to be identified are carried out convolution algorithm by others' face image and figure's image to be identified,
Operation result is obtained, is specifically included:
The facial image to be identified and figure's image to be identified are split as to the face characteristic image and body of preset quantity respectively
State characteristic image, and obtain face characteristic image and the corresponding location information of aspectual character image after splitting;
The face characteristic image and aspectual character image are subjected to bilinear interpolation processing, the face characteristic figure that obtains that treated
Picture and aspectual character image;
Treated face characteristic image and aspectual character image are subjected to cross-layer convolution algorithm, obtain the convolution algorithm knot of each layer
Fruit and cross-layer feature difference;
The convolution algorithm result of each layer is compensated according to the cross-layer feature difference, by the convolution algorithm of each layer after compensation
As a result spliced according to the location information, using the convolution algorithm result of spliced each layer as operation result.
3. recognition of face intelligent unlocking method as claimed in claim 2, which is characterized in that it is described will treated face characteristic
Image and aspectual character image carry out cross-layer convolution algorithm, obtain the convolution algorithm result and cross-layer feature difference of each layer, specifically
Including:
Treated face characteristic image and aspectual character image are subjected to cross-layer convolution algorithm, obtain the Feature Mapping figure of each layer
With the cross-layer feature difference of each layer;
The Feature Mapping figure of each layer is overlapped, the corresponding each feature vector of the Feature Mapping figure after being superimposed;
The corresponding covariance matrix of each feature vector is calculated, principal component analysis is carried out to the covariance matrix, after obtaining dimensionality reduction
Feature vector, using the feature vector after dimensionality reduction as the convolution algorithm result of each layer.
4. recognition of face intelligent unlocking method as claimed in claim 3, which is characterized in that described by the operation result and institute
The sample convolution characteristic value for stating the corresponding gender sample image of intelligent door lock is matched, and generates matching result, is specifically included:
Obtain the corresponding sample convolution feature of corresponding gender sample image that the intelligent door lock stores in the preset database
Value;
Characteristic value in the operation result is subjected to Dynamic Matching with the sample characteristics, and generates matching result.
5. recognition of face intelligent unlocking method as claimed in claim 4, which is characterized in that it is described will be in the operation result
Characteristic value carries out Dynamic Matching with the sample characteristics, and generates matching result, specifically includes:
Obtain the corresponding sample characteristics of gender sample image of different predetermined angles;
Successively respectively by the operation result on multiple default human face characteristic point positions and multiple default physical trait point positions
In characteristic value be compared with each sample characteristics, and generate multiple characteristic value comparison results;
Matching result is generated according to multiple characteristic value comparison results.
6. recognition of face intelligent unlocking method as claimed in claim 5, which is characterized in that described to be compared according to multiple characteristic values
As a result matching result is generated, is specifically included:
Multiple characteristic value comparison results are weighted average computation according to default weight proportion, obtain images match degree;
Matching result is generated according to described image matching degree.
7. recognition of face intelligent unlocking method as claimed in claim 6, which is characterized in that described when matching result is matching one
When cause, the gender matching corresponding with public place of the gender of active user is judged, and current door lock state is adjusted to opening state
State specifically includes:
When described image matching degree is more than pre-set image matching threshold, it is consistent to match to judge the matching result, and judgement is worked as
The gender matching corresponding with public place of the gender of preceding user, and current door lock state is adjusted to open state;
Correspondingly, when described image matching degree is less than or equal to pre-set image matching threshold, judge the matching result for not
Match, judges that the gender of active user gender corresponding with public place mismatches, and current door lock state is adjusted to closing shape
State, and generate prompt message and prompted.
8. a kind of recognition of face intelligent unlocking system, which is characterized in that the recognition of face intelligent unlocking system includes:
Computing module, the facial image to be identified for obtaining active user and figure's image to be identified, by the people to be identified
Face image and figure's image to be identified carry out convolution algorithm, obtain operation result;
Matching module is used for the sample convolution feature of operation result gender sample image corresponding with the intelligent door lock
Value is matched, and generates matching result;
Unlocking module, for when matching result is to match consistent, judging the gender of active user property corresponding with public place
It does not match, and current door lock state is adjusted to open state.
9. a kind of intelligent door lock, which is characterized in that the intelligent door lock includes:Memory, processor and it is stored in the storage
On device and the recognition of face intelligent unlocking program that can run on the processor, the recognition of face intelligent unlocking program configuration
For the step of realizing the recognition of face intelligent unlocking method as described in any one of claim 1 to 7.
10. a kind of storage medium, which is characterized in that recognition of face intelligent unlocking program is stored on the storage medium, it is described
The recognition of face intelligence as described in any one of claim 1 to 7 is realized when recognition of face intelligent unlocking program is executed by processor
The step of energy method for unlocking.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810554708.XA CN108447168A (en) | 2018-05-31 | 2018-05-31 | Recognition of face intelligent unlocking method, system, intelligent door lock and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810554708.XA CN108447168A (en) | 2018-05-31 | 2018-05-31 | Recognition of face intelligent unlocking method, system, intelligent door lock and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108447168A true CN108447168A (en) | 2018-08-24 |
Family
ID=63206385
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810554708.XA Withdrawn CN108447168A (en) | 2018-05-31 | 2018-05-31 | Recognition of face intelligent unlocking method, system, intelligent door lock and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108447168A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109448204A (en) * | 2018-12-27 | 2019-03-08 | 深圳市宇昊电子科技有限公司 | Unattended bathing Saunas hydrotherapy managed operation method |
CN110374446A (en) * | 2019-04-03 | 2019-10-25 | 泰州市康平医疗科技有限公司 | Intelligent clothes cabinet door body control device |
CN110599639A (en) * | 2019-08-13 | 2019-12-20 | 深圳市天彦通信股份有限公司 | Identity verification method and related product |
CN111227522A (en) * | 2019-04-03 | 2020-06-05 | 泰州市康平医疗科技有限公司 | Intelligent wardrobe door control method |
CN117058787A (en) * | 2023-08-16 | 2023-11-14 | 鹿客科技(北京)股份有限公司 | Door lock control method, device, electronic equipment and computer readable medium |
-
2018
- 2018-05-31 CN CN201810554708.XA patent/CN108447168A/en not_active Withdrawn
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109448204A (en) * | 2018-12-27 | 2019-03-08 | 深圳市宇昊电子科技有限公司 | Unattended bathing Saunas hydrotherapy managed operation method |
CN110374446A (en) * | 2019-04-03 | 2019-10-25 | 泰州市康平医疗科技有限公司 | Intelligent clothes cabinet door body control device |
CN111227522A (en) * | 2019-04-03 | 2020-06-05 | 泰州市康平医疗科技有限公司 | Intelligent wardrobe door control method |
CN110599639A (en) * | 2019-08-13 | 2019-12-20 | 深圳市天彦通信股份有限公司 | Identity verification method and related product |
CN117058787A (en) * | 2023-08-16 | 2023-11-14 | 鹿客科技(北京)股份有限公司 | Door lock control method, device, electronic equipment and computer readable medium |
CN117058787B (en) * | 2023-08-16 | 2024-04-12 | 鹿客科技(北京)股份有限公司 | Door lock control method, device, electronic equipment and computer readable medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108447168A (en) | Recognition of face intelligent unlocking method, system, intelligent door lock and storage medium | |
EP3493088B1 (en) | Security gesture authentication | |
US10268910B1 (en) | Authentication based on heartbeat detection and facial recognition in video data | |
Kar et al. | Study of implementing automated attendance system using face recognition technique | |
CN108399665A (en) | Method for safety monitoring, device based on recognition of face and storage medium | |
CN105320921B (en) | Eyes localization method and eyes positioning device | |
CN110991249A (en) | Face detection method, face detection device, electronic equipment and medium | |
US10679082B2 (en) | Self-Service Terminal (SST) facial authentication processing | |
CN109389028A (en) | Face identification method, device, equipment and storage medium based on motion analysis | |
CN111104917A (en) | Face-based living body detection method and device, electronic equipment and medium | |
CN111241873A (en) | Image reproduction detection method, training method of model thereof, payment method and payment device | |
Jaikumar et al. | Fingerprint based student attendance system with SMS alert to parents | |
CN108230525A (en) | Access control management method and gate inhibition management apparatus | |
Carbone | Investigation of stochastic preference theory using experimental data | |
US20230063229A1 (en) | Spoof detection based on challenge response analysis | |
CN109670420A (en) | Store the control method and device, storage terminal, electronic equipment, medium of terminal | |
CN109409071A (en) | Unlocking method, device and the electronic equipment of electronic equipment | |
CN110516426A (en) | Identity identifying method, certification terminal, device and readable storage medium storing program for executing | |
CN111258414B (en) | Method and device for adjusting screen | |
CN107203647B (en) | Student information acquisition method and device based on biological characteristics | |
CN109665387A (en) | Elevator intelligent takes method, apparatus, computer equipment and storage medium | |
US11935009B2 (en) | Integrating healthcare screening with other identity-based functions | |
US11803898B2 (en) | Account establishment and transaction management using biometrics and intelligent recommendation engine | |
US11962602B2 (en) | Physical environment based account authentication | |
EP4210000B1 (en) | Methods and systems for determining authenticity of a document |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180824 |