CN109241868A - Face identification method, device, computer equipment and storage medium - Google Patents
Face identification method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109241868A CN109241868A CN201810934470.3A CN201810934470A CN109241868A CN 109241868 A CN109241868 A CN 109241868A CN 201810934470 A CN201810934470 A CN 201810934470A CN 109241868 A CN109241868 A CN 109241868A
- Authority
- CN
- China
- Prior art keywords
- feature vector
- user
- vector
- target
- similarity
- 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 54
- 238000003860 storage Methods 0.000 title claims abstract description 17
- 239000013598 vector Substances 0.000 claims abstract description 260
- 239000000284 extract Substances 0.000 claims abstract description 20
- 238000004590 computer program Methods 0.000 claims description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 4
- 230000008569 process Effects 0.000 abstract description 18
- 238000012795 verification Methods 0.000 abstract description 5
- 238000003062 neural network model Methods 0.000 abstract description 4
- 238000012549 training Methods 0.000 abstract description 4
- 230000001815 facial effect Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000000513 principal component analysis Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 239000003550 marker Substances 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000036651 mood Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000011840 criminal investigation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- 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
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a kind of face identification method, device, computer equipment and storage medium, which includes: acquisition images to be recognized, extracts feature vector to be identified according to images to be recognized；Obtain the characteristic similarity between feature vector to be identified and each reference characteristic vector；By the corresponding user identifier of the highest reference characteristic vector of characteristic similarity, identified as target user；The target area obtained in images to be recognized is identified according to target user, and target feature vector is extracted according to target area；It calculates target feature vector and target user identifies the vector similarity between corresponding user-defined feature vector, recognition result is obtained according to vector similarity.The face identification method of offer of the invention does not need multiple image and carries out comprehensive verification, does not need the neural network model that training is complicated in advance to realize yet, can also improve recognition efficiency while guaranteeing face recognition process safety.
Description
Technical field
The invention belongs to field of image processings, are to be related to a kind of face identification method, device, computer more specifically
Equipment and storage medium.
Background technique
Face recognition technology is a kind of biometrics identification technology, obtains the development advanced by leaps and bounds in recent years.Currently,
Face recognition technology is widely used to the fields such as criminal investigation and case detection, banking system, customs inspection, the civil affairs department, work and rest attendance.So
And general face recognition process requires to input several pictures progress comprehensive verification, or needs the nerve that training is complicated in advance
Network model realizes that the process of realization is relatively complicated, the mechanism of identification needs excessive computer resource, and the efficiency of identification is not
It is high.
Summary of the invention
The embodiment of the present invention provides a kind of face identification method, device, computer equipment and storage medium, to solve face
The cumbersome problem of the realization process of identification.
A kind of face identification method, comprising:
Images to be recognized is obtained, feature vector to be identified is extracted according to the images to be recognized；
Obtain the characteristic similarity between the feature vector to be identified and each reference characteristic vector；
By the corresponding user identifier of the highest reference characteristic vector of characteristic similarity, identified as target user；
The target area obtained in images to be recognized is identified according to the target user, and is extracted according to the target area
Target feature vector；
It calculates the target feature vector and the target user identifies vector between corresponding user-defined feature vector
Similarity obtains recognition result according to the vector similarity.
A kind of face identification device, comprising:
Characteristic vector pickup module extracts feature to be identified according to the images to be recognized for obtaining images to be recognized
Vector；
Characteristic similarity obtains module, for obtaining between the feature vector to be identified and each reference characteristic vector
Characteristic similarity；
User identifier obtains module, for making the corresponding user identifier of the highest reference characteristic vector of characteristic similarity
For target user's mark；
Target feature vector extraction module, for identifying the target area obtained in images to be recognized according to the target user
Domain, and target feature vector is extracted according to the target area；
Recognition result obtains module, identifies corresponding make by oneself for calculating the target feature vector and the target user
Vector similarity between adopted feature vector obtains recognition result according to the vector similarity.
A kind of computer equipment, including memory, processor and storage are in the memory and can be in the processing
The computer program run on device, the processor realize the step of above-mentioned face identification method when executing the computer program
Suddenly.
A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the meter
The step of calculation machine program realizes above-mentioned face identification method when being executed by processor.
Above-mentioned face identification method, device, computer equipment and storage medium, by obtain images to be recognized, according to
Identify image zooming-out feature vector to be identified；Then the feature between feature vector to be identified and each reference characteristic vector is obtained
Similarity is identified the corresponding user identifier of the highest reference characteristic vector of characteristic similarity as target user；Then according to
Target user identifies the target area obtained in images to be recognized, and extracts target feature vector according to target area；Finally count
It calculates target feature vector and target user identifies the vector similarity between corresponding user-defined feature vector, according to the vector phase
Recognition result is obtained like degree.It is the different identification information of each user configuration by user defined feature vector, knows in face
It can directly be matched during not, not need to carry out comprehensive verification by multiple image, it is complicated not need training in advance yet
Neural network model realize, while guaranteeing face recognition process safety also improve recognition efficiency.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is an application environment schematic diagram of face identification method in one embodiment of the invention；
Fig. 2 is a flow chart of face identification method in one embodiment of the invention；
Fig. 3 is another flow chart of face identification method in one embodiment of the invention；
Fig. 4 is another flow chart of face identification method in one embodiment of the invention；
Fig. 5 is a functional block diagram of face identification device in one embodiment of the invention；
Fig. 6 is a schematic diagram of computer equipment in one embodiment of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Face identification method provided by the present application can be applicable in the application environment such as Fig. 1, wherein client passes through net
Network is communicated with server-side, and server-side obtains images to be recognized by client, extracts spy to be identified according to images to be recognized
Levy vector；Then the characteristic similarity between feature vector to be identified and each reference characteristic vector is obtained, and feature is similar
The corresponding user identifier of highest reference characteristic vector is spent to identify as target user；Then according to target user identify obtain to
It identifies the target area of image, and target feature vector is extracted according to target area；Finally calculate target feature vector and target
Vector similarity between the corresponding user-defined feature vector of user identifier, obtains recognition result according to vector similarity and returns
Client.Wherein, client can be, but not limited to be various personal computers, laptop, smart phone, tablet computer and
Portable wearable device.Server-side can be with the server cluster of the either multiple server compositions of independent server come real
It is existing.
In one embodiment, as shown in Fig. 2, providing a kind of face identification method, the service in Fig. 1 is applied in this way
It is illustrated, includes the following steps: for end
S10: obtaining images to be recognized, extracts feature vector to be identified according to images to be recognized.
Wherein, images to be recognized refers to the detection image for carrying out recognition of face.Optionally, the figure that server-side passes through client
Images to be recognized is obtained as acquisition equipment, for example, subscriber station carries out Image Acquisition in the pickup area of image capture device, is used
Acquired image is sent to server-side by family end, to obtain images to be recognized.Feature vector refers to for characterizing image letter
The vector for ceasing feature, can avoid subsequent duplicate extraction operation with simple data characterization image information.Feature vector has more
Seed type, such as: HOG (Histogra mof Oriented Gradient, gradient orientation histogram) feature vector, LBP
(Local Binary Patterns, local binary patterns) feature vector or PCA (Principal Component
Analysis, principal component analysis) feature vector etc..
Preferably, the HOG feature vector of images to be recognized can be extracted in the present embodiment as feature vector to be identified.By
It in the HOG feature vector of images to be recognized is described by the gradient of the local message of images to be recognized, therefore, extracted wait know
The HOG feature vector of other image can be avoided the Factors on Human face such as geometric deformation and light variation as feature vector to be identified and know
Other influence.
Specifically, the extraction of HOG feature vector can be realized by following steps:
(1) collected images to be recognized is subjected to greyscale image transitions and gamma normalized.
(2) by the unit of step (1) treated images to be recognized is divided into several pixels, per 4 adjacent units
Constitute a section.
(3) histogram according to gradient direction of each pixel in each unit is counted.
(4) histogram for calculating each unit, obtains the feature of unit.
(5) finally the feature of each unit in all sections is together in series, obtains HOG feature, i.e., HOG feature to
Amount.
S20: the characteristic similarity between feature vector to be identified and each reference characteristic vector is obtained.
Wherein, reference characteristic vector is the corresponding feature vector of benchmark face image extracted in advance.And benchmark face figure
It seem the facial image of different user gathered in advance, it is different to represent by the reference characteristic vector for extracting each user
User.Be due to reference characteristic vector user in the database with server-side be it is one-to-one, by calculating wait know
Characteristic similarity between other feature vector and each reference characteristic vector is it may determine that images to be recognized is and which width out
Benchmark face image is closest.
Specifically, the process for extracting reference characteristic vector can be with are as follows: the facial image of the preset quantity of one user of acquisition,
Then the feature vector of each width facial image is extracted, and calculates the mean value of these feature vectors, using obtained mean value as this
The reference characteristic vector of user.It should be understood that feature vector to be identified be it is corresponding with reference characteristic vector, for example, if benchmark
Feature vector is HOG feature vector, then feature vector to be identified is also HOG feature vector.
Optionally, characteristic similarity can be obtained using similarity calculation algorithm, such as: Euclidean distance algorithm,
Manhatton distance algorithm, Minkowski distance algorithm or cosine similarity algorithm.
In one embodiment, it is calculated between feature vector to be identified and reference characteristic vector using Euclidean distance algorithm
Characteristic similarity:
Wherein, U is feature vector to be identified, and V is benchmark feature vector, sim (U, V)EDGFeature between U and V is similar
Degree.
The characteristic similarity between feature vector and reference characteristic vector to be identified is calculated by Euclidean algorithm, it can be with
Accurate data basis is improved for the judgement of subsequent similarity.
S30: the corresponding user identifier of the highest reference characteristic vector of characteristic similarity is identified as target user.
Specifically, the feature phase between feature vector to be identified and each reference characteristic vector is got by step S20
After degree, by the corresponding user identifier of the highest reference characteristic vector of characteristic similarity, identifies, that is, judge as target user
Target face in images to be recognized is corresponding with which user identifier.
It is appreciated that the user in images to be recognized may be not present in the database of server-side, for recognition of face
The optimization of process and user information safety, in one embodiment, before step S30, if the maximum value of characteristic similarity is less than
When default similarity threshold, then the recognition result of recognition failures is exported.
Wherein, preset similarity threshold server-side it is pre-set can with user identifier to it is corresponding when feature to be identified to
The characteristic similarity minimum value of amount and reference characteristic vector.
Specifically, the calculating that images to be recognized is carried out with each reference characteristic vector to characteristic similarity respectively, if wait know
When the maximum value of the characteristic similarity of the feature vector of other image and all reference characteristic vectors is less than preset similarity threshold,
Then illustrate that any one of the target face and data library in identification image user keeps off, illustrates that the target face can
It can not be in the database.In this way, the recognition result of output recognition failures.Optionally, when exporting recognition failures result, stop people
The process of face identification, refusal user log in, and guarantee user information safety.
Optionally, in order to avoid during obtaining images to be recognized due to faulty operation etc. caused by feature it is similar
It spends maximum value and is less than similarity threshold, can be set allows user's recognition failures to carry out the mistake for resurveying image and calculating again
Journey if exporting the recognition result of recognition failures again beyond preset times, and stops face recognition process, and refusal user steps on
Record, guarantees the information security of user.
In the present embodiment, if the maximum value of the characteristic similarity of the feature vector of images to be recognized and reference characteristic vector
Less than similarity threshold, then the recognition result of recognition failures is exported, prevent user from maliciously being logged in, guarantees the information peace of user
Entirely.
In one embodiment, can also by neural network (NN, Neural Networks) model or support to
Amount machine (SVM, Support Vector Machine) realizes step S10-S30.Specifically, after obtaining images to be recognized
Directly images to be recognized is input in preparatory trained neural network model or support vector machines, obtains recognition result.
Wherein, recognition result be images to be recognized has been carried out with user identifier it is corresponding, because with can be by the corresponding use of recognition result
Family mark is identified as target user.
S40: the target area obtained in images to be recognized is identified according to target user, and target is extracted according to target area
Feature vector.
Wherein, target area is a region of the customized setting of user, (establishes data carrying out user data acquisition
Library) during, can be customized by the user a target area, and the respective action of the target area is set.In step
Target user's mark is determined in S30 and then obtains the target area in images to be recognized according to target user's mark, most
Afterwards extract target area target feature vector with judge the target area movement whether the correspondence with the customized setting of user
Keep strokes.
Specifically, the point of the target critical in images to be recognized, then root can be oriented using facial feature points detection algorithm
Target area is obtained according to target critical point, target feature vector is finally extracted according to target area.
Such as: the target area of the customized setting of user is left eye, and the respective action of left eye is arranged to close one's eyes.Then service
End can obtain the left eye region in images to be recognized according to the corresponding target user's mark of the user, and extract the left eye region
Feature vector, with judge the target area movement it is whether consistent with the respective action of the customized setting of user.
S50: it is similar to calculate the vector that target feature vector and target user identify between corresponding user-defined feature vector
Degree, obtains recognition result according to vector similarity.
Wherein, user-defined feature vector refers to the feature vector that user obtains according to the image of customized setting, Ke Yizuo
For the foundation of user's recognition of face.Such as user is the image closed one's eyes by customized setting left eye, then obtains on this image
Feature vector be user-defined feature vector.After extracting target feature vector, target feature vector and customized is calculated
Vector similarity between feature vector, specific calculation method is similar with step S30, and details are not described herein.Obtaining vector
After similarity, different recognition results can be exported according to the numerical value of the vector similarity.
In one embodiment, can judge and export recognition result according to baseline threshold, i.e. in step S50, according to
Amount similarity obtains recognition result, can specifically include: if vector similarity is more than or equal to baseline threshold, exporting matching
Successful recognition result；If vector similarity is less than baseline threshold, the output recognition result that it fails to match.
Specifically, a baseline threshold can be preset, is then closed according to the size of the vector similarity and baseline threshold
System is to export different recognition results.If the vector similarity be greater than or equal to baseline threshold, then it is assumed that target feature vector and from
Defined feature vector be it is matched, export the recognition result of successful match；If the vector similarity is less than baseline threshold, then it is assumed that
Target feature vector and user-defined feature vector be it is unmatched, export the recognition result that it fails to match.
In the corresponding embodiment of Fig. 2, by obtaining images to be recognized, according to images to be recognized extract feature to be identified to
Amount；Then the characteristic similarity between feature vector to be identified and each reference characteristic vector is obtained, by characteristic similarity highest
The corresponding user identifier of reference characteristic vector as target user identify；It is identified then according to target user and obtains figure to be identified
Target area as in, and target feature vector is extracted according to target area；Finally calculate target feature vector and target user
The vector similarity between corresponding user-defined feature vector is identified, recognition result is obtained according to the vector similarity.Pass through use
Family user-defined feature vector is the different identification information of each user configuration, can directly be carried out in face recognition process
Match, do not need to carry out comprehensive verification by multiple image, does not need the neural network model that training is complicated in advance also to realize,
Guarantee to also improve recognition efficiency while face recognition process safety.
In one embodiment, as shown in figure 3, before step S40, i.e., acquisition figure to be identified is being identified according to target user
Target area as in, and before the step of extracting target feature vector according to target area, people provided in an embodiment of the present invention
Face recognition method further includes following steps:
S61: customized configuring request is obtained, wherein custom-configuring request includes log-on message.
Wherein, custom-configuring request is the initiation request that user configures own verification information.Log-on message
Refer to the information of the progress authentication of user's input, optionally, log-on message includes user account and user password.
Specifically, server-side obtains customized configuring request by client, when get custom-configure request when, lead to
It crosses client and sends login authentication to user, make user in client input log-on message to verify.
S62: if log-on message is verified, zone list to be selected is sent.
Specifically, server-side can inquire login password corresponding with user account in the database according to user account,
Then whether verifying user password is consistent with login password.If the two is consistent, it is verified, and sends to user terminal to constituency
Domain list is selected for user.Wherein, zone list to be selected refers to the table listed to favored area, and refers to favored area and work as
The region for user's selection of preceding support, for example, can be eye areas, mouth region or body other parts to favored area
Region etc..
It optionally, is entire facial area to favored area.It correspondingly, can be with during subsequent user is customized
Defined by different moods, for example, by it is happy, sad, sad, frightened, angry, surprised, detest it is peaceful wait quietly mood come
Definition；Then the corresponding region of images to be recognized is identified, using corresponding Emotion identification model again to obtain corresponding judgement
As a result.
S63: the configuration image that user selects information and acquires predetermined quantity is obtained based on zone list to be selected, according to user
Select the target area in each configuration image of acquisition of information.
Wherein, user selects information to refer to the selection information that user makes according to zone list to be selected, i.e. user selects
Target area.
Specifically, user is obtained to select information according to the user of zone list to be selected and acquire the configuration diagram of predetermined quantity
Picture, and the target area in each configuration image of acquisition of information is selected according to user.Wherein, predetermined quantity can be according to practical feelings
Condition is configured, and the embodiment of the present invention is not specifically limited.
For example, user selects eyes as target area by zone list to be selected, then client obtains the user
The facial image of predetermined quantity is as configuration image, then acquires the eye areas in each configuration image.
S64: extracting the feature vector of the target area in each configuration image and calculates the feature vector of the target area
Mean value, obtain user-defined feature vector.
Specifically, the feature vector of the target area in each configuration image obtained according to step S63 is extracted,
Then the mean value for calculating the feature vector of these target areas, using obtained mean value as user-defined feature vector.Wherein, it extracts
The method that vector sum calculates vector mean value is identical as previous embodiment, and which is not described herein again.By extracting in each configuration image
Target area feature vector, and calculate the mean value of the feature vector of these target areas, user can be custom-configured
Movement be converted to user-defined feature vector.
In the corresponding embodiment of Fig. 3, by obtaining customized configuring request, the login custom-configured in request is believed
Breath is verified, if log-on message is verified, sends zone list to be selected to user；Zone list to be selected is then based on to obtain
It takes family selection information and acquires the configuration image of predetermined quantity, the mesh in each configuration image of acquisition of information is selected according to user
Mark region；Finally extract the feature vector of the target area in each configuration image and the feature vector that calculates the target area
Mean value obtains user-defined feature vector.It, can be different for each user configuration by the way that the user-defined feature vector of user is arranged
Feature vector, so that the identification for face provides support.Also, by user-defined feature vector, it is subsequent according to target signature to
It measures with the similarity of user-defined feature vector and identifies face, the efficiency that user carries out recognition of face can be promoted.
In one embodiment, as shown in figure 4, after step S50, that is, target feature vector and target user's mark are being calculated
The vector similarity for knowing corresponding user-defined feature vector, after the step of obtaining recognition result according to vector similarity, this hair
The face identification method that bright embodiment provides further includes following steps:
S71: if recognition result is that it fails to match, counting in preset time target user's mark it fails to match number,
Judging in preset time it fails to match, whether number reaches frequency threshold value.
Specifically, it sets a frequency threshold value and counts target user's mark it fails to match number within a preset time,
When target user's mark occurs that it fails to match within a preset time, correspondingly by the target user number that identifies that it fails to match
Carry out plus 1 operation, and judge in preset time target user's mark it fails to match whether number reaches frequency threshold value.Its
In, preset time is a preset period, and optionally, which can be 12 hours, 24 hours or 48
Hour etc..
S72: if it fails to match in preset time, number reaches frequency threshold value, and matching terminates, and is marked according to target user
Know and sends prompting message.
Specifically, if it fails to match that number reaches frequency threshold value for target user's mark within a preset time, illustrate
It is more with the frequency of failure, then stop matched process, and identify to send to corresponding user according to target user and prompt to disappear
Breath, prompting the user, there may be the risks that account is maliciously verified, so that user is modified configuration or other operations.
S73: if it fails to match in preset time, number is not up to frequency threshold value, sends image capturing request, and obtain
New images to be recognized.
Specifically, if it fails to match that number is not up to frequency threshold value for target user's mark within a preset time, in order to
Reduce the error of Image Acquisition or matching process, server-side sends an image capturing request to user terminal, makes user terminal pair
Image is resurveyed, to obtain new images to be recognized.
S74: extracting the new feature vector to be identified of new images to be recognized, and calculate new feature vector to be identified with
Target user identifies the characteristic similarity between corresponding reference characteristic vector.
Specifically, after obtaining new images to be recognized, extract the new feature to be identified of new images to be recognized to
Amount, and calculate new feature vector to be identified and target user and identify the characteristic similarity between corresponding reference characteristic vector,
Wherein, new feature vector to be identified is specifically extracted identical with step S20 and S30 with calculating process, and which is not described herein again.
S75: it if characteristic similarity is greater than or equal to similarity threshold, returns to execution and is waited for according to target user's marker extraction
Identify the target area in image, and the step of target feature vector is extracted according to target area.
Specifically, if characteristic similarity is greater than or equal to similarity threshold, show that corresponding user is present in database,
But cause the target area of input imperfect or inaccurate due to Image Acquisition, cause that previous it fails to match；
Therefore S40 is returned to step, i.e., according to the target area in target user's marker extraction images to be recognized, and according to target area
Domain is extracted the step of target feature vector, and the step of continuing to execute below.
S76: it if characteristic similarity is less than similarity threshold, returns and executes target user's mark in statistics preset time
It fails to match number judges in preset time target user's mark it fails to match the step of whether number reaches frequency threshold value.
Specifically, if characteristic similarity is less than similarity threshold, show that user may be not present in database, but
The reason of being likely due to Image Acquisition causes new images to be recognized imperfect or inaccurate, cause new feature to be identified to
Amount and the similarity of reference characteristic vector are lower than similarity threshold, therefore return to step S71, i.e. mesh in statistics preset time
User identifier it fails to match number is marked, judges in preset time target user's mark it fails to match whether number reaches number
The step of threshold value, and the step of continuing to execute below.
In the corresponding embodiment of Fig. 4, pass through target user's mark it fails to match number, judgement in statistics preset time
It fails to match in preset time, and whether number reaches frequency threshold value, if reaching frequency threshold value, matching terminates, according to target user
Mark sends prompting message, and prompting user, there may be the risks that account is maliciously verified；If not up to frequency threshold value is sent
Image capturing request, acquires new images to be recognized, and extracts new feature vector to be identified, calculate new feature to be identified to
Amount identifies the characteristic similarity between corresponding reference characteristic vector with target user；If similarity is greater than or equal to similarity threshold
Value then returns to the step of execution extracts target area and extracts target feature vector, if similarity is less than similarity threshold, returns
The step of receipt row statistical match frequency of failure, subsequent operation is carried out as needed, improve the integrality of recognition of face, guarantee
User experience is promoted while user information safety.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
In one embodiment, a kind of face identification device is provided, face is known in the face identification device and above-described embodiment
Other method corresponds.As shown in figure 5, the face identification device includes characteristic vector pickup module 10, characteristic similarity acquisition
Module 20, user identifier obtain module 30, target feature vector extraction module 40 and recognition result and obtain module 50.Each function mould
Detailed description are as follows for block:
Characteristic vector pickup module 10, for obtaining images to be recognized, according to images to be recognized extract feature to be identified to
Amount.
Characteristic similarity obtains module 20, for obtaining the spy between feature vector to be identified and each reference characteristic vector
Levy similarity.
Further, it is to calculate feature vector to be identified using Euclidean distance algorithm that characteristic similarity, which obtains module 20,
Characteristic similarity between reference characteristic vector:
Wherein, U is feature vector to be identified, and V is benchmark feature vector, sim (U, V)EDGFeature between U and V is similar
Degree.
User identifier obtains module 30, is used for the corresponding user identifier of the highest reference characteristic vector of characteristic similarity,
It is identified as target user.
Optionally, if the maximum value of characteristic similarity is less than default similarity threshold, it is defeated that user identifier obtains module 30
The recognition result of recognition failures out.
Target feature vector extraction module 40, for identifying the target area obtained in images to be recognized according to target user
Domain, and target feature vector is extracted according to target area.
Recognition result obtains module 50, identifies corresponding user-defined feature for calculating target feature vector and target user
Vector similarity between vector obtains recognition result according to vector similarity.
Further, if recognition result acquisition module 50, which is also used to vector similarity, is more than or equal to baseline threshold,
Export the recognition result of successful match；If vector similarity is less than baseline threshold, the output recognition result that it fails to match.
Further, face identification device provided in an embodiment of the present invention further includes user defined logic interface, optionally, is used
Family custom block includes configuring request acquiring unit, to favored area transmission unit, target area acquiring unit and customized spy
Levy vector acquiring unit.
Configuring request acquiring unit, for obtaining customized configuring request, wherein custom-configuring request includes logging in letter
Breath.
To favored area transmission unit, if being verified for log-on message, zone list to be selected is sent.
Target area acquiring unit, for obtaining user's selection information based on zone list to be selected and acquiring predetermined quantity
Image is configured, the target area in each configuration image of acquisition of information is selected according to user.
User-defined feature vector acquiring unit, by extract it is each configuration image in target area feature vector and based on
The mean value for calculating the feature vector of target area, obtains user-defined feature vector.
Further, face identification device provided in an embodiment of the present invention further includes again authentication module, optionally, again
Authentication module includes: matching times statistic unit, stopping matching unit, image reacquires unit, similarity recalculates list
Member, first return to execution unit and second and return to execution unit.
Matching times statistic unit counts target user in preset time and marks if being that it fails to match for recognition result
Know it fails to match number, judging in preset time that it fails to match, whether number reaches frequency threshold value.
Stop matching unit, if number reaches frequency threshold value for it fails to match in preset time, matching terminates, and root
It is identified according to target user and sends prompting message.
Image reacquires unit and sends figure if number is not up to frequency threshold value for it fails to match in preset time
As acquisition request, and obtain new images to be recognized.
Similarity recalculates unit, for extracting the new feature vector to be identified of new images to be recognized, and calculates
New feature vector to be identified identifies the characteristic similarity between corresponding reference characteristic vector with target user.
First return execution unit, if for characteristic similarity be greater than or equal to similarity threshold, return execution according to
Target user identifies the target area obtained in images to be recognized, and extracts target feature vector according to target area.
Second returns to execution unit, if being less than similarity threshold for characteristic similarity, return execution statistics it is default when
Interior target user mark it fails to match number judges in preset time target user's mark it fails to match whether number reaches
To frequency threshold value.
Specific about face identification device limits the restriction that may refer to above for face identification method, herein not
It repeats again.Modules in above-mentioned face identification device can be realized fully or partially through software, hardware and combinations thereof.On
Stating each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also store in a software form
In memory in computer equipment, the corresponding operation of the above modules is executed in order to which processor calls.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 6.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is for storing characteristic similarity algorithm, reference characteristic vector, user identifier and user-defined feature vector etc..
The network interface of the computer equipment is used to communicate with external terminal by network connection.The computer program is held by processor
To realize a kind of face identification method when row.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor perform the steps of when executing computer program
Images to be recognized is obtained, feature vector to be identified is extracted according to images to be recognized；
Obtain the characteristic similarity between feature vector to be identified and each reference characteristic vector；
By the corresponding user identifier of the highest reference characteristic vector of characteristic similarity, identified as target user；
The target area obtained in images to be recognized is identified according to target user, and target signature is extracted according to target area
Vector；
It calculates target feature vector and target user identifies the vector similarity between corresponding user-defined feature vector, root
Recognition result is obtained according to vector similarity.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor
Images to be recognized is obtained, feature vector to be identified is extracted according to images to be recognized；
Obtain the characteristic similarity between feature vector to be identified and each reference characteristic vector；
By the corresponding user identifier of the highest reference characteristic vector of characteristic similarity, identified as target user；
The target area obtained in images to be recognized is identified according to target user, and target signature is extracted according to target area
Vector；
It calculates target feature vector and target user identifies the vector similarity between corresponding user-defined feature vector, root
Recognition result is obtained according to vector similarity.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of described device is divided into different functional unit or module, more than completing
The all or part of function of description.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of face identification method characterized by comprising
Images to be recognized is obtained, feature vector to be identified is extracted according to the images to be recognized；
Obtain the characteristic similarity between the feature vector to be identified and each reference characteristic vector；
By the corresponding user identifier of the highest reference characteristic vector of characteristic similarity, identified as target user；
The target area obtained in images to be recognized is identified according to the target user, and target is extracted according to the target area
Feature vector；
It calculates the target feature vector and vector that the target user identifies between corresponding user-defined feature vector is similar
Degree, obtains recognition result according to the vector similarity.
2. face identification method as described in claim 1, which is characterized in that obtained in described identified according to the target user
Target area in images to be recognized, and before the step of extracting target feature vector according to the target area, the face
Recognition methods further include:
Obtain customized configuring request, wherein described custom-configure requests to include log-on message；
If the log-on message is verified, zone list to be selected is sent；
The configuration image that user selects information and acquires predetermined quantity is obtained based on the zone list to be selected, according to the user
Select the target area in each configuration image of acquisition of information；
Extract the feature vector of the target area in each configuration image and the feature vector that calculates the target area
Mean value obtains user-defined feature vector.
3. face identification method as described in claim 1, which is characterized in that described to be identified according to the vector similarity
As a result, comprising:
If the vector similarity is more than or equal to baseline threshold, the recognition result of successful match is exported；
If the vector similarity is less than the baseline threshold, the output recognition result that it fails to match.
4. face identification method as claimed in claim 3, which is characterized in that calculate the target feature vector and institute described
It states target user and identifies vector similarity between corresponding user-defined feature vector, identified according to the vector similarity
As a result after, the face identification method further include:
If the recognition result is that it fails to match, target user mark in preset time is counted it fails to match number,
It fails to match described in judging in preset time, and whether number reaches frequency threshold value；
If it fails to match described in preset time, number reaches frequency threshold value, and matching terminates, and is marked according to the target user
Know and sends prompting message；
If it fails to match described in preset time, number is not up to frequency threshold value, sends image capturing request, and obtain newly
Images to be recognized；
Extract the new feature vector to be identified of the new images to be recognized, and calculate the new feature vector to be identified with
Target user identifies the characteristic similarity between corresponding reference characteristic vector；
If the characteristic similarity is greater than or equal to similarity threshold, returns to described identify according to the target user of execution and obtain
Take the target area in images to be recognized, and the step of target feature vector is extracted according to the target area；
If the characteristic similarity is less than the similarity threshold, returns and execute target user's mark in the statistics preset time
Know it fails to match number, judges target user's mark in preset time it fails to match whether number reaches the step of frequency threshold value
Suddenly.
5. face identification method as described in claim 1, which is characterized in that described to obtain feature vector to be identified and each base
Characteristic similarity between quasi- feature vector, comprising:
Feature phase between the feature vector to be identified and the reference characteristic vector is calculated using Euclidean distance algorithm
Like degree:
Wherein, U is feature vector to be identified, and V is benchmark feature vector, sim (U, V)EDGCharacteristic similarity between U and V.
6. face identification method as described in claim 1, which is characterized in that described that the highest benchmark of characteristic similarity is special
Before the step of levying the corresponding user identifier of vector, being identified as target user, the face identification method further include:
If the maximum value of the characteristic similarity is less than default similarity threshold, the recognition result of recognition failures is exported.
7. a kind of face identification device characterized by comprising
Characteristic vector pickup module extracts feature vector to be identified according to the images to be recognized for obtaining images to be recognized；
Characteristic similarity obtains module, for obtaining the feature between the feature vector to be identified and each reference characteristic vector
Similarity；
User identifier obtains module, is used for by the corresponding user identifier of the highest reference characteristic vector of characteristic similarity, as mesh
Mark user identifier；
Target feature vector extraction module, for identifying the target area obtained in images to be recognized according to the target user,
And target feature vector is extracted according to the target area；
Recognition result obtains module, identifies corresponding customized spy for calculating the target feature vector and the target user
The vector similarity between vector is levied, recognition result is obtained according to the vector similarity.
8. face identification device as claimed in claim 7, which is characterized in that the face identification device further includes that user makes by oneself
Adopted module, the user defined logic interface include configuring request acquiring unit, to favored area transmission unit, target area acquisition list
Member and user-defined feature vector acquiring unit；
The configuring request acquiring unit, it is described to custom-configure request including logging in letter for obtaining customized configuring request
Breath；
It is described to favored area transmission unit, if being verified for the log-on message, send zone list to be selected；
The target area acquiring unit, for obtaining user's selection information based on the zone list to be selected and acquiring predetermined number
The configuration image of amount selects the target area in each configuration image of acquisition of information according to the user；
The user-defined feature vector acquiring unit, for extracting the feature vector of the target area in each configuration image
And the mean value of the feature vector of the target area is calculated, obtain user-defined feature vector.
9. a kind of computer equipment, including memory, processor and storage are in the memory and can be in the processor
The computer program of upper operation, which is characterized in that the processor realized when executing the computer program as claim 1 to
The step of any one of 6 face identification method.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In the step of realization face identification method as described in any one of claim 1 to 6 when the computer program is executed by processor
Suddenly.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810934470.3A CN109241868A (en) | 2018-08-16 | 2018-08-16 | Face identification method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810934470.3A CN109241868A (en) | 2018-08-16 | 2018-08-16 | Face identification method, device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109241868A true CN109241868A (en) | 2019-01-18 |
Family
ID=65070629
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810934470.3A Pending CN109241868A (en) | 2018-08-16 | 2018-08-16 | Face identification method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109241868A (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109829424A (en) * | 2019-01-30 | 2019-05-31 | 华南理工大学 | A kind of intelligent three-dimensional garage Transport Vehicle method and system based on face recognition technology |
| CN110490142A (en) * | 2019-08-21 | 2019-11-22 | 北京百度网讯科技有限公司 | For generating the method and device of alarm signal |
| CN110909766A (en) * | 2019-10-29 | 2020-03-24 | 北京明略软件系统有限公司 | Similarity determination method and device, storage medium and electronic device |
| CN111144241A (en) * | 2019-12-13 | 2020-05-12 | 深圳奇迹智慧网络有限公司 | Target identification method and device based on image verification and computer equipment |
| CN111325156A (en) * | 2020-02-24 | 2020-06-23 | 北京沃东天骏信息技术有限公司 | Face recognition method, device, equipment and storage medium |
| CN111462059A (en) * | 2020-03-24 | 2020-07-28 | 湖南大学 | Parallel processing method and device for intelligent target detection of fetal ultrasound image |
| CN111488562A (en) * | 2020-04-07 | 2020-08-04 | 中国移动通信集团江苏有限公司 | Threshold determination method, device, equipment and medium based on face comparison |
| CN111639700A (en) * | 2020-05-28 | 2020-09-08 | 深圳壹账通智能科技有限公司 | Target similarity recognition method and device, computer equipment and readable storage medium |
| CN111738062A (en) * | 2020-05-09 | 2020-10-02 | 广智微芯(扬州)有限公司 | Automatic re-identification method and system based on embedded platform |
| CN111931718A (en) * | 2020-09-22 | 2020-11-13 | 深圳市视美泰技术股份有限公司 | Method and device for updating face features based on face recognition and computer equipment |
| CN111967448A (en) * | 2020-09-25 | 2020-11-20 | 北京百度网讯科技有限公司 | Sample image processing method, device, equipment and storage medium |
| CN112101215A (en) * | 2020-09-15 | 2020-12-18 | Oppo广东移动通信有限公司 | Face input method, terminal equipment and computer readable storage medium |
| CN112215183A (en) * | 2020-10-21 | 2021-01-12 | 中国银行股份有限公司 | Bank customer identification method and device |
| CN112232103A (en) * | 2019-07-15 | 2021-01-15 | 北京蜂盒科技有限公司 | Face recognition method and system, storage medium, terminal and server |
| CN112633297A (en) * | 2020-12-28 | 2021-04-09 | 浙江大华技术股份有限公司 | Target object identification method and device, storage medium and electronic device |
| CN112906571A (en) * | 2021-02-20 | 2021-06-04 | 成都新希望金融信息有限公司 | Living body identification method and device and electronic equipment |
| CN113343004A (en) * | 2021-06-10 | 2021-09-03 | 浙江大华技术股份有限公司 | Object recognition method and device, storage medium and electronic device |
| CN113378837A (en) * | 2021-06-28 | 2021-09-10 | 北京百度网讯科技有限公司 | License plate shielding identification method and device, electronic equipment and storage medium |
| CN113762106A (en) * | 2021-08-23 | 2021-12-07 | 深圳云天励飞技术股份有限公司 | Face recognition method and device, electronic equipment and storage medium |
| CN113837270A (en) * | 2021-09-18 | 2021-12-24 | 广东人工智能与先进计算研究院 | Target identification method, device, equipment and storage medium |
| CN113901648A (en) * | 2021-09-24 | 2022-01-07 | 成都飞机工业（集团）有限责任公司 | Method and device for compiling part process rules |
| CN114783085A (en) * | 2022-03-21 | 2022-07-22 | 南京信息工程大学 | Novel sharing bicycle based on face recognition |
| CN115171312A (en) * | 2022-06-28 | 2022-10-11 | 重庆京东方智慧科技有限公司 | Image processing method, device, equipment, monitoring system and storage medium |
| CN115909434A (en) * | 2022-09-07 | 2023-04-04 | 以萨技术股份有限公司 | Data processing system for acquiring human face image characteristics |
| CN112215183B (en) * | 2020-10-21 | 2024-06-21 | 中国银行股份有限公司 | Bank customer identification method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101070442B1 (en) * | 2010-05-03 | 2011-10-05 | 주식회사 크라스아이디 | Face recognition system and method using multi-level face recognition |
| CN105139470A (en) * | 2015-09-30 | 2015-12-09 | 杭州海康威视数字技术股份有限公司 | Checking-in method, device and system based on face recognition |
| CN105590089A (en) * | 2015-10-22 | 2016-05-18 | 广州视源电子科技股份有限公司 | Face identification method and device |
| CN106844492A (en) * | 2016-12-24 | 2017-06-13 | 深圳云天励飞技术有限公司 | A kind of method of recognition of face, client, server and system |
| CN108062673A (en) * | 2017-11-15 | 2018-05-22 | 平安科技（深圳）有限公司 | Method of payment, terminal device and computer readable storage medium |
| CN108346191A (en) * | 2018-02-06 | 2018-07-31 | 中国平安人寿保险股份有限公司 | Work attendance method, device, computer equipment and storage medium |
-
2018
- 2018-08-16 CN CN201810934470.3A patent/CN109241868A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101070442B1 (en) * | 2010-05-03 | 2011-10-05 | 주식회사 크라스아이디 | Face recognition system and method using multi-level face recognition |
| CN105139470A (en) * | 2015-09-30 | 2015-12-09 | 杭州海康威视数字技术股份有限公司 | Checking-in method, device and system based on face recognition |
| CN105590089A (en) * | 2015-10-22 | 2016-05-18 | 广州视源电子科技股份有限公司 | Face identification method and device |
| CN106844492A (en) * | 2016-12-24 | 2017-06-13 | 深圳云天励飞技术有限公司 | A kind of method of recognition of face, client, server and system |
| CN108062673A (en) * | 2017-11-15 | 2018-05-22 | 平安科技（深圳）有限公司 | Method of payment, terminal device and computer readable storage medium |
| CN108346191A (en) * | 2018-02-06 | 2018-07-31 | 中国平安人寿保险股份有限公司 | Work attendance method, device, computer equipment and storage medium |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109829424A (en) * | 2019-01-30 | 2019-05-31 | 华南理工大学 | A kind of intelligent three-dimensional garage Transport Vehicle method and system based on face recognition technology |
| CN112232103A (en) * | 2019-07-15 | 2021-01-15 | 北京蜂盒科技有限公司 | Face recognition method and system, storage medium, terminal and server |
| CN110490142B (en) * | 2019-08-21 | 2022-08-23 | 北京百度网讯科技有限公司 | Method and device for generating alarm signal |
| CN110490142A (en) * | 2019-08-21 | 2019-11-22 | 北京百度网讯科技有限公司 | For generating the method and device of alarm signal |
| CN110909766A (en) * | 2019-10-29 | 2020-03-24 | 北京明略软件系统有限公司 | Similarity determination method and device, storage medium and electronic device |
| CN111144241A (en) * | 2019-12-13 | 2020-05-12 | 深圳奇迹智慧网络有限公司 | Target identification method and device based on image verification and computer equipment |
| CN111325156A (en) * | 2020-02-24 | 2020-06-23 | 北京沃东天骏信息技术有限公司 | Face recognition method, device, equipment and storage medium |
| CN111325156B (en) * | 2020-02-24 | 2023-08-11 | 北京沃东天骏信息技术有限公司 | Face recognition method, device, equipment and storage medium |
| CN111462059A (en) * | 2020-03-24 | 2020-07-28 | 湖南大学 | Parallel processing method and device for intelligent target detection of fetal ultrasound image |
| CN111462059B (en) * | 2020-03-24 | 2023-09-29 | 湖南大学 | Parallel processing method and device for intelligent target detection of fetal ultrasonic image |
| CN111488562A (en) * | 2020-04-07 | 2020-08-04 | 中国移动通信集团江苏有限公司 | Threshold determination method, device, equipment and medium based on face comparison |
| CN111738062A (en) * | 2020-05-09 | 2020-10-02 | 广智微芯(扬州)有限公司 | Automatic re-identification method and system based on embedded platform |
| CN111738062B (en) * | 2020-05-09 | 2024-05-17 | 广智微芯(扬州)有限公司 | Automatic re-identification system based on embedded platform |
| CN111639700A (en) * | 2020-05-28 | 2020-09-08 | 深圳壹账通智能科技有限公司 | Target similarity recognition method and device, computer equipment and readable storage medium |
| CN112101215A (en) * | 2020-09-15 | 2020-12-18 | Oppo广东移动通信有限公司 | Face input method, terminal equipment and computer readable storage medium |
| CN111931718A (en) * | 2020-09-22 | 2020-11-13 | 深圳市视美泰技术股份有限公司 | Method and device for updating face features based on face recognition and computer equipment |
| CN111931718B (en) * | 2020-09-22 | 2024-05-14 | 深圳市视美泰技术股份有限公司 | Method, device and computer equipment for updating face characteristics based on face recognition |
| CN111967448B (en) * | 2020-09-25 | 2024-06-14 | 北京百度网讯科技有限公司 | Sample image processing method, device, equipment and storage medium |
| CN111967448A (en) * | 2020-09-25 | 2020-11-20 | 北京百度网讯科技有限公司 | Sample image processing method, device, equipment and storage medium |
| CN112215183B (en) * | 2020-10-21 | 2024-06-21 | 中国银行股份有限公司 | Bank customer identification method and device |
| CN112215183A (en) * | 2020-10-21 | 2021-01-12 | 中国银行股份有限公司 | Bank customer identification method and device |
| CN112633297B (en) * | 2020-12-28 | 2023-04-07 | 浙江大华技术股份有限公司 | Target object identification method and device, storage medium and electronic device |
| CN112633297A (en) * | 2020-12-28 | 2021-04-09 | 浙江大华技术股份有限公司 | Target object identification method and device, storage medium and electronic device |
| CN112906571A (en) * | 2021-02-20 | 2021-06-04 | 成都新希望金融信息有限公司 | Living body identification method and device and electronic equipment |
| CN112906571B (en) * | 2021-02-20 | 2023-09-05 | 成都新希望金融信息有限公司 | Living body identification method and device and electronic equipment |
| CN113343004A (en) * | 2021-06-10 | 2021-09-03 | 浙江大华技术股份有限公司 | Object recognition method and device, storage medium and electronic device |
| CN113378837A (en) * | 2021-06-28 | 2021-09-10 | 北京百度网讯科技有限公司 | License plate shielding identification method and device, electronic equipment and storage medium |
| CN113762106A (en) * | 2021-08-23 | 2021-12-07 | 深圳云天励飞技术股份有限公司 | Face recognition method and device, electronic equipment and storage medium |
| CN113837270B (en) * | 2021-09-18 | 2022-08-30 | 广东人工智能与先进计算研究院 | Target identification method, device, equipment and storage medium |
| CN113837270A (en) * | 2021-09-18 | 2021-12-24 | 广东人工智能与先进计算研究院 | Target identification method, device, equipment and storage medium |
| CN113901648A (en) * | 2021-09-24 | 2022-01-07 | 成都飞机工业（集团）有限责任公司 | Method and device for compiling part process rules |
| CN114783085A (en) * | 2022-03-21 | 2022-07-22 | 南京信息工程大学 | Novel sharing bicycle based on face recognition |
| CN115171312A (en) * | 2022-06-28 | 2022-10-11 | 重庆京东方智慧科技有限公司 | Image processing method, device, equipment, monitoring system and storage medium |
| CN115909434A (en) * | 2022-09-07 | 2023-04-04 | 以萨技术股份有限公司 | Data processing system for acquiring human face image characteristics |
| CN115909434B (en) * | 2022-09-07 | 2023-07-04 | 以萨技术股份有限公司 | Data processing system for acquiring facial image characteristics |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109241868A (en) | Face identification method, device, computer equipment and storage medium | |
| KR102038851B1 (en) | Method and system for verifying identities | |
| CN109472269A (en) | Characteristics of image configuration and method of calibration, device, computer equipment and medium | |
| CN108197557A (en) | Testimony of a witness consistency check method, terminal device and computer readable storage medium | |
| CN106710017B (en) | Identity verification method, device and system for logistics signing | |
| CN109472208A (en) | Certificates handling method, apparatus, computer equipment and storage medium based on recognition of face | |
| CN110705451A (en) | Face recognition method, face recognition device, terminal and server | |
| WO2019062080A1 (en) | Identity recognition method, electronic device, and computer readable storage medium | |
| CN109410026A (en) | Identity identifying method, device, equipment and storage medium based on recognition of face | |
| CN109756458B (en) | Identity authentication method and system | |
| CN109063628B (en) | Face recognition method, device, computer equipment and storage medium | |
| HUE031428T2 (en) | Method, system and computer program for comparing images | |
| CN110348193A (en) | Verification method, device, equipment and storage medium | |
| CN110955874A (en) | Identity authentication method, identity authentication device, computer equipment and storage medium | |
| US11126827B2 (en) | Method and system for image identification | |
| CN106557678A (en) | A kind of intelligent terminal's mode switching method and its device | |
| CN109711357A (en) | A kind of face identification method and device | |
| CN109886110A (en) | Micro- expression methods of marking, device, computer equipment and storage medium | |
| CN108429745A (en) | The authentication method and Verification System of login, webpage login method and system | |
| CN110163094A (en) | Biopsy method, device, equipment and storage medium based on gesture motion | |
| CN110197107A (en) | Micro- expression recognition method, device, computer equipment and storage medium | |
| WO2019178753A1 (en) | Payment method, device and system | |
| CN110298246A (en) | Unlocking verification method, device, computer equipment and storage medium | |
| CN111191506A (en) | Personnel flow statistical method and device, computer equipment and storage medium | |
| CN108875336A (en) | The method of face authentication and typing face, authenticating device and system |
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 |