CN106250858A - A kind of recognition methods merging multiple face recognition algorithms and system - Google Patents
A kind of recognition methods merging multiple face recognition algorithms and system Download PDFInfo
- Publication number
- CN106250858A CN106250858A CN201610634586.6A CN201610634586A CN106250858A CN 106250858 A CN106250858 A CN 106250858A CN 201610634586 A CN201610634586 A CN 201610634586A CN 106250858 A CN106250858 A CN 106250858A
- Authority
- CN
- China
- Prior art keywords
- facial image
- face recognition
- recognition algorithms
- face
- recognition 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.)
- Granted
Links
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
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
The present invention provides a kind of recognition methods merging multiple face recognition algorithms and system, including: preset at least two face recognition algorithms;By every kind of face recognition algorithms, facial image to be identified and sample facial image are carried out feature extraction and identification, respectively obtain the characteristic vector of facial image to be identified of correspondence, the characteristic vector of sample facial image, contrast Similarity value and recognition result;The characteristic vector of the facial image to be identified extracted is carried out fusion treatment and the characteristic vector of the sample facial image extracted is carried out fusion treatment, determines final recognition result according to the fusion feature vector of facial image to be identified and the fusion feature vector of sample facial image;Or, the contrast Similarity value obtained is carried out fusion treatment, the contrast Similarity value according to merging determines final recognition result;Or, recognition result is carried out decision-making treatment, determines final recognition result according to result.So, the embodiment of the present invention can improve the precision of recognition of face.
Description
Technical field
The present invention relates to technical field of face recognition, particularly relate to a kind of identification side merging multiple face recognition algorithms
Method and system.
Background technology
Face recognition technology is recognition method based on biological characteristic, utilizes that mankind itself has and can be unique
Indicate the physiological feature of its identity or behavior characteristics carries out the technology of authentication.Along with artificial intelligence technology and computer patterns
The development identified, face recognition technology moves towards actual application from laboratory.
At present, existing face identification system is realized through optimizing by a kind of face recognition algorithms, recognition of face
Therefore precision and accuracy are subject to certain restrictions.As can be seen here, for solving to use single face recognition algorithms in recognition of face
Precision and the problem that is restricted of accuracy aspect, need badly at present and a kind of merge multiple face recognition algorithms and carry out recognition of face
Method, to improve the precision of recognition of face.
Summary of the invention
The shortcoming of prior art in view of the above, it is an object of the invention to provide the multiple recognition of face of a kind of fusion and calculates
The recognition methods of method and system, merge multiple face recognition algorithms and be identified certification, it is possible to increase the precision of recognition of face.
For achieving the above object and other relevant purposes, the embodiment of the present invention provides one to merge multiple face recognition algorithms
Recognition methods, preset at least two face recognition algorithms, the method also includes:
By every kind of face recognition algorithms, facial image to be identified and sample facial image are carried out feature extraction, obtain respectively
Characteristic vector and the characteristic vector of sample facial image to facial image to be identified corresponding to every kind of face recognition algorithms;
The characteristic vector of the facial image to be identified according to every kind of face recognition algorithms extraction and the spy of sample facial image
Levy vector, be calculated the contrast Similarity value of every kind of face recognition algorithms respectively;
Contrast Similarity value according to every kind of face recognition algorithms determines the recognition result of every kind of face recognition algorithms respectively;
The characteristic vector of the facial image to be identified that every kind of face recognition algorithms is extracted is carried out fusion treatment and to every kind
The characteristic vector of the sample facial image that face recognition algorithms is extracted carries out fusion treatment, according to the fusion of facial image to be identified
The fusion feature vector of characteristic vector and sample facial image determines final recognition result;Or,
The contrast Similarity value of every kind of face recognition algorithms is carried out fusion treatment, true according to the contrast Similarity value merged
Fixed final recognition result;Or,
The recognition result of every kind of face recognition algorithms is carried out decision-making treatment, determines final identification knot according to result
Really.
Preferably, the described characteristic vector to the facial image to be identified that every kind of face recognition algorithms is extracted is carried out at fusion
Reason also carries out fusion treatment, according to face to be identified to the characteristic vector of the sample facial image that every kind of face recognition algorithms is extracted
The fusion feature vector of image and the fusion feature vector of sample facial image determine final recognition result, including:
Characteristic vector for the facial image to be identified of every kind of face recognition algorithms extraction carries out fusion treatment, is treated
Identify the fusion feature vector f of facial imageA, and by Principal Component Analysis Algorithm PCA to fusion feature vector fACarry out at dimensionality reduction
Reason, obtains new fusion feature vector FA:
Characteristic vector for the sample facial image of every kind of face recognition algorithms extraction carries out fusion treatment, obtains sample
The fusion feature vector f of facial imageB, and by PCA to fusion feature vector fBCarry out dimension-reduction treatment, obtain new fusion special
Levy vector FB:
New fusion feature vector F is calculated according to probability linear discriminant analysis algorithm PLDAAWith new fusion feature vector FB
Contrast similarity, obtain merge contrast Similarity value;
According to the contrast Similarity value merged and default contrast similarity threshold, determine final recognition result.
Preferably, the described contrast Similarity value to every kind of face recognition algorithms carries out fusion treatment, right according to merge
Final recognition result is determined than Similarity value, including:
Preset the weight coefficient of every kind of face recognition algorithms;
Contrast Similarity value according to every kind of face recognition algorithms and weight coefficient, obtain merge contrast Similarity value:
According to the contrast Similarity value merged and default contrast similarity threshold, determine final recognition result.
Preferably, the described recognition result to every kind of face recognition algorithms carries out decision-making treatment, determines according to result
Final recognition result, including:
Determine quantity and the quantity of all recognition results identifying successful recognition result;
Judge the quantity of all recognition results identifying whether the quantity of successful recognition result is more than 1/2nd;
If identifying, the quantity of successful recognition result is more than the quantity of all recognition results of 1/2nd, it is determined that final
Recognition result for identify successfully;
If identifying, the quantity of successful recognition result is less than or equal to the quantity of all recognition results of 1/2nd, the most really
Fixed final recognition result is recognition failures.
Preferably, the described recognition result to every kind of face recognition algorithms carries out decision-making treatment, determines according to result
Final recognition result, including:
Previously according to every kind of face recognition algorithms, every class facial image is identified, obtains every kind of face recognition algorithms pair
The discrimination of every class facial image;
Determine the object type belonging to described facial image to be identified and sample facial image;
The discrimination to every class facial image according to the object type determined and every kind of face recognition algorithms being previously obtained,
Choose the face recognition algorithms that discrimination corresponding to this object type is the highest;
The recognition result that the face recognition algorithms chosen obtains is defined as final recognition result.
According to said method, embodiments provide a kind of identification system merging multiple face recognition algorithms, should
System includes: presetting module, characteristic extracting module, computing module, first determines module, second determine module;Wherein,
Described presetting module, is used for presetting at least two face recognition algorithms;
Described characteristic extracting module, is used for by every kind of face recognition algorithms facial image to be identified and sample face figure
As carrying out feature extraction, respectively obtain characteristic vector and the sample people of every kind of facial image to be identified corresponding to face recognition algorithms
The characteristic vector of face image;
Described computing module, for the characteristic vector of facial image to be identified extracted according to every kind of face recognition algorithms and
The characteristic vector of sample facial image, is calculated the contrast Similarity value of every kind of face recognition algorithms respectively;
Described first determines module, for determining every kind of people respectively according to the contrast Similarity value of every kind of face recognition algorithms
The recognition result of face recognizer;
Described second determines module, for the characteristic vector to the facial image to be identified that every kind of face recognition algorithms is extracted
Carry out fusion treatment and the characteristic vector of the sample facial image that every kind of face recognition algorithms is extracted is carried out fusion treatment, according to
The fusion feature vector of facial image to be identified and the fusion feature vector of sample facial image determine final recognition result;Or
Person, carries out fusion treatment to the contrast Similarity value of every kind of face recognition algorithms, and the contrast Similarity value according to merging determines
Whole recognition result;Or, the recognition result of every kind of face recognition algorithms is carried out decision-making treatment, determines according to result
Whole recognition result.
Preferably, described second determine module specifically for:
Characteristic vector for the facial image to be identified of every kind of face recognition algorithms extraction carries out fusion treatment, is treated
Identify the fusion feature vector f of facial imageA, and by PCA to fusion feature vector fACarry out dimension-reduction treatment, obtain new melting
Close characteristic vector FA:
Characteristic vector for the sample facial image of every kind of face recognition algorithms extraction carries out fusion treatment, obtains sample
The fusion feature vector f of facial imageB, and by PCA to fusion feature vector fBCarry out dimension-reduction treatment, obtain new fusion special
Levy vector FB:
New fusion feature vector F is calculated according to PLDAAWith new fusion feature vector FBContrast similarity, melted
The contrast Similarity value closed;
According to the contrast Similarity value merged and default contrast similarity threshold, determine final recognition result.
Preferably, described second determine module specifically for:
Preset the weight coefficient of every kind of face recognition algorithms;
Contrast Similarity value according to every kind of face recognition algorithms and weight coefficient, obtain merge contrast Similarity value:
According to the contrast Similarity value merged and default contrast similarity threshold, determine final recognition result.
Preferably, described second determine module specifically for:
Determine quantity and the quantity of all recognition results identifying successful recognition result;
Judge the quantity of all recognition results identifying whether the quantity of successful recognition result is more than 1/2nd;
If identifying, the quantity of successful recognition result is more than the quantity of all recognition results of 1/2nd, it is determined that final
Recognition result for identify successfully;
If identifying, the quantity of successful recognition result is less than or equal to the quantity of all recognition results of 1/2nd, the most really
Fixed final recognition result is recognition failures.
Preferably, described second determine module specifically for:
Previously according to every kind of face recognition algorithms, every class facial image is identified, obtains every kind of face recognition algorithms pair
The discrimination of every class facial image;
Determine the object type belonging to described facial image to be identified and sample facial image;
The discrimination to every class facial image according to the object type determined and every kind of face recognition algorithms being previously obtained,
Choose the face recognition algorithms that discrimination corresponding to this object type is the highest;
The recognition result that the face recognition algorithms chosen obtains is defined as final recognition result.
The recognition methods of the multiple face recognition algorithms of fusion that the present invention provides and system, including: preset at least two people
Face recognizer;By every kind of face recognition algorithms, facial image to be identified and sample facial image are carried out feature extraction, point
Do not obtain characteristic vector and the characteristic vector of sample facial image of every kind of facial image to be identified corresponding to face recognition algorithms;
The characteristic vector of the facial image to be identified according to every kind of face recognition algorithms extraction and the characteristic vector of sample facial image, point
It is not calculated the contrast Similarity value of every kind of face recognition algorithms;Contrast Similarity value according to every kind of face recognition algorithms divides
Do not determine the recognition result of every kind of face recognition algorithms;Feature to the facial image to be identified that every kind of face recognition algorithms is extracted
Vector carries out fusion treatment and the characteristic vector of the sample facial image that every kind of face recognition algorithms is extracted is carried out fusion treatment,
Fusion feature vector and the fusion feature vector of sample facial image according to facial image to be identified determine final identification knot
Really;Or, the contrast Similarity value of every kind of face recognition algorithms is carried out decision-making treatment, true according to the contrast Similarity value merged
Fixed final recognition result;Or, the recognition result of every kind of face recognition algorithms is carried out decision-making treatment, true according to result
Fixed final recognition result.So, the embodiment of the present invention merges multiple face recognition algorithms and carries out recognition of face, from feature to
Amount end, contrast Similarity value end or recognition result end carry out Decision fusion, determine final identification according to the result of Decision fusion
As a result, solve the problem using single face recognition algorithms to be restricted in terms of the precision and accuracy of recognition of face, thus
Improve the precision of recognition of face.
Accompanying drawing explanation
Fig. 1 is shown as the schematic flow sheet of the recognition methods merging multiple face recognition algorithms of the present invention;
Fig. 2 is shown as the composition structural representation of the identification system merging multiple face recognition algorithms of the present invention.
Detailed description of the invention
In the embodiment of the present invention, for solving to use single face recognition algorithms to be subject in terms of the precision and accuracy of recognition of face
To the problem limited, multiple face recognition algorithms is used to carry out recognition of face, and in different aspects to multiple recognition of face
Algorithm carries out Decision fusion, specifically can carry out Decision fusion from characteristic vector end, contrast Similarity value end or recognition result end,
Result according to Decision fusion determines final recognition result, thus improves the precision of recognition of face.
Below in conjunction with the accompanying drawings and specific embodiment the present invention will be further described in detail.
The embodiment of the present invention proposes a kind of recognition methods merging multiple face recognition algorithms, as it is shown in figure 1, the method
Including:
Step S100: preset at least two face recognition algorithms.
In this step, arbitrary at least two face recognition algorithms can be set, here to default face recognition algorithms
It is not especially limited.
Step S101: by every kind of face recognition algorithms, facial image to be identified and sample facial image are carried out feature and carry
Take, respectively obtain characteristic vector and the feature of sample facial image of every kind of facial image to be identified corresponding to face recognition algorithms
Vector.
In this step, can be random length according to the length of the characteristic vector of different face recognition algorithms extractions, but
The characteristic vector extracted must is fulfilled for following condition:
Norm (F)=1
Wherein, F is characterized vector, and the mould that Norm (F) is characterized vector F is long.
Step S102: the characteristic vector of the facial image to be identified extracted according to every kind of face recognition algorithms and sample face
The characteristic vector of image, is calculated the contrast Similarity value of every kind of face recognition algorithms respectively.
In this step, the contrast Similarity value of every kind of face recognition algorithms can be normalized to [0,1] interval, be united
The contrast Similarity value of one, in order to follow-up carry out Classification and Identification.
Step S103: determine every kind of face recognition algorithms respectively according to the contrast Similarity value of every kind of face recognition algorithms
Recognition result.
In this step, the contrast Similarity value respectively every kind of face recognition algorithms obtained and the contrast similarity threshold preset
Value contrasts;
If the contrast Similarity value of this face recognition algorithms is more than the contrast similarity threshold preset, it is determined that this face is known
The recognition result of other algorithm is for identify successfully;
If the contrast Similarity value of this face recognition algorithms is less than or equal to the contrast similarity threshold preset, it is determined that should
The recognition result of face recognition algorithms is recognition failures.
Here it is possible to preset contrast similarity threshold according to practical situation and demand, here to contrast similarity threshold not
Make concrete restriction.
Step S104: the characteristic vector of the facial image to be identified that every kind of face recognition algorithms is extracted is carried out fusion treatment
And the characteristic vector of the sample facial image that every kind of face recognition algorithms is extracted is carried out fusion treatment, according to face figure to be identified
The fusion feature vector of picture and the fusion feature vector of sample facial image determine final recognition result;Or, to every kind of people
The contrast Similarity value of face recognizer carries out fusion treatment, determines final identification knot according to the contrast Similarity value merged
Really;Or, the recognition result of every kind of face recognition algorithms is carried out decision-making treatment, determines final identification knot according to result
Really.
In this step, to specifically the most how the characteristic vector of the facial image to be identified that every kind of face recognition algorithms is extracted being entered
Row fusion treatment also carries out fusion treatment, according to treating to the characteristic vector of the sample facial image that every kind of face recognition algorithms is extracted
Identify that the fusion feature vector of facial image and the fusion feature vector of sample facial image determine final recognition result, carry out
Describe in detail:
First, the characteristic vector for the facial image to be identified of every kind of face recognition algorithms extraction carries out fusion treatment,
Obtain the fusion feature vector f of facial image to be identifiedA:
fA=[f1;f2;…;fn], wherein, fAFor fusion feature vector, f1~fnExtract for every kind of face recognition algorithms
The characteristic vector of facial image to be identified;
By Principal Component Analysis Algorithm (Principal Components Analysis, PCA) to fusion feature vector fA
Carry out dimension-reduction treatment, obtain new fusion feature vector FA:
Wherein, Z is the eigenmatrix of PCA;
In like manner, the characteristic vector for the sample facial image of every kind of face recognition algorithms extraction carries out fusion treatment,
Fusion feature vector f to sample facial imageB:
fB=[f1;f2;…;fn], wherein, fBFor fusion feature vector, f1~fnExtract for every kind of face recognition algorithms
The characteristic vector of sample facial image;
By PCA to fusion feature vector fBCarry out dimension-reduction treatment, obtain new fusion feature vector FB:
Wherein, Z is the eigenmatrix of PCA;
Then, according to probability linear discriminant analysis algorithm (Probabilistic Linear Discriminant
Analysis, PLDA) calculate new fusion feature vector FAWith new fusion feature vector FBContrast similarity, merged
Contrast Similarity value S (AB):
Wherein, P is the variance within clusters of PLDA, and Q is the inter-class variance of PLDA;
Finally, according to the contrast Similarity value merged and default contrast similarity threshold, final recognition result is determined;
Concrete, the contrast Similarity value merged is contrasted with the contrast similarity threshold preset;
If the contrast Similarity value merged is more than the contrast similarity threshold preset, it is determined that final recognition result is for knowing
The most successful;
If the contrast Similarity value merged is less than or equal to the contrast similarity threshold preset, it is determined that final identification knot
Fruit is recognition failures.
Here it is possible to preset contrast similarity threshold according to practical situation and demand, here to contrast similarity threshold not
Make concrete restriction.
In this step, to specifically the most how the contrast Similarity value of every kind of face recognition algorithms being carried out fusion treatment, according to
The contrast Similarity value merged determines final recognition result, is described in detail:
First, the weight coefficient of every kind of face recognition algorithms is preset;
Then, according to contrast Similarity value and the weight coefficient of every kind of face recognition algorithms, the contrast obtaining merging is similar
Angle value:
Wherein, S (A, B) is the contrast Similarity value merged, ωiIt is the weight coefficient of i-th kind of face recognition algorithms, FiFor
The contrast Similarity value of i-th kind of face recognition algorithms, i=1,2 ..., n, n are positive integer;
Finally, according to the contrast Similarity value merged and default contrast similarity threshold, final recognition result is determined;
Concrete, the contrast Similarity value merged is contrasted with the contrast similarity threshold preset;
If the contrast Similarity value merged is more than the contrast similarity threshold preset, it is determined that final recognition result is for knowing
The most successful;
If the contrast Similarity value merged is less than or equal to the contrast similarity threshold preset, it is determined that final identification knot
Fruit is recognition failures.
Here it is possible to preset contrast similarity threshold according to practical situation and demand, here to contrast similarity threshold not
Make concrete restriction.
In this step, to specifically the most how the recognition result of every kind of face recognition algorithms being carried out decision-making treatment, according to process
Result determines final recognition result, is described in detail:
Mode one, decision strategy based on ballot
First, determine and identify amount R and quantity G of all recognition results of successful recognition result;
Then, it is judged that identify whether the amount R of successful recognition result is more than the number of all recognition results of 1/2nd
Amount G;
If identifying, the amount R of successful recognition result is more than quantity G of all recognition results of 1/2nd, it is determined that
Whole recognition result is for identify successfully;
If identifying, the amount R of successful recognition result is less than or equal to quantity G of all recognition results of 1/2nd, then
Determine that final recognition result is recognition failures.
Wherein, every kind of face recognition algorithms obtains a recognition result, and quantity G is equal to the quantity of face recognition algorithms, number
Amount R is less than or equal to quantity G.
Mode two, decision strategy based on prior information
Having, in view of every kind of face recognition algorithms, the identification object type that itself is good at, such as, some face is known
Other algorithm is preferable to the discrimination of child's facial image, and undesirable to the discrimination of certificate photo.Therefore, it can previously according to
Every class facial image is identified by every kind of face recognition algorithms, can obtain every kind of face recognition algorithms to every class facial image
Discrimination table;, when there being one group of image to need to be identified, first obtain the feature of this group facial image, such as, according to face
The feature of image determines it is the facial image of child, or determines it is the facial image of ethnic groups according to the feature of facial image,
Then using the object type determined and the discrimination that is previously obtained as prior information, select discrimination according to this prior information
Good face recognition algorithms.
Concrete, previously according to every kind of face recognition algorithms, every class facial image is identified, obtains every kind of face and know
The other algorithm discrimination to every class facial image, as shown in table 1:
Face recognition algorithms 1 | Face recognition algorithms | …… | Face recognition algorithms N | |
A class | 0.980 | 0.965 | …… | 0.910 |
B class | 0.923 | 0.995 | …… | 0.945 |
C class | 0.921 | 0.936 | …… | 0.988 |
…… | …… | …… | …… | …… |
Table 1
Determine the object type belonging to described facial image to be identified and sample facial image;
According to the every kind of face recognition algorithms the being previously obtained discrimination to every class facial image, choose this object type pair
The face recognition algorithms that the discrimination answered is the highest;
The recognition result that the face recognition algorithms chosen obtains is defined as final recognition result.
Concrete, first facial image is divided into multiple object type, such as, A class: collection in worksite facial image and certificate
According to certification;B class: child's certification;C class: ethnic groups certification;Then according to described facial image to be identified and sample facial image
Feature determine affiliated object type;According to the every kind of face recognition algorithms the being previously obtained identification to every class facial image
Rate, chooses the face recognition algorithms that discrimination corresponding to this object type is the highest;If the face knowledge that only a discrimination is the highest
Other algorithm, then be defined as final recognition result by this face recognition algorithms, if there being the recognition of face that multiple discrimination is the highest to calculate
Method, then be defined as final recognition result by the recognition result that face recognition algorithms the highest for any one discrimination obtains.
For realizing said method, the embodiment of the present invention additionally provides a kind of identification system merging multiple face recognition algorithms
System, owing to the principle of system solution problem is similar to method, therefore, before the implementation process of system and enforcement principle all may refer to
State the implementation process of method and implement principles illustrated, repeating no more in place of repetition.
Embodiments provide a kind of identification system merging multiple face recognition algorithms, as in figure 2 it is shown, this system
Including: presetting module 200, characteristic extracting module 201, computing module 202, first determine module 203, second determine module 204;
Wherein,
Described presetting module 200, is used for presetting at least two face recognition algorithms;
Described characteristic extracting module 201, is used for by every kind of face recognition algorithms facial image to be identified and sample people
Face image carries out feature extraction, respectively obtains characteristic vector and the sample of every kind of facial image to be identified corresponding to face recognition algorithms
The characteristic vector of this facial image;
Described computing module 202, for the feature of facial image to be identified extracted according to every kind of face recognition algorithms to
Amount and the characteristic vector of sample facial image, be calculated the contrast Similarity value of every kind of face recognition algorithms respectively;
Described first determines module 203, for determining respectively often according to the contrast Similarity value of every kind of face recognition algorithms
Plant the recognition result of face recognition algorithms;
Described second determines module 204, for the feature to the facial image to be identified that every kind of face recognition algorithms is extracted
Vector carries out fusion treatment and the characteristic vector of the sample facial image that every kind of face recognition algorithms is extracted is carried out fusion treatment,
Fusion feature vector and the fusion feature vector of sample facial image according to facial image to be identified determine final identification knot
Really;Or, the contrast Similarity value of every kind of face recognition algorithms is carried out fusion treatment, true according to the contrast Similarity value merged
Fixed final recognition result;Or, the recognition result of every kind of face recognition algorithms is carried out fusion treatment, according to the identification merged
Result determines final recognition result.
In being embodied as, described second determine module 204 specifically for:
Characteristic vector for the facial image to be identified of every kind of face recognition algorithms extraction carries out fusion treatment, is treated
Identify the fusion feature vector f of facial imageA, and by PCA to fusion feature vector fACarry out dimension-reduction treatment, obtain new melting
Close characteristic vector FA:
Characteristic vector for the sample facial image of every kind of face recognition algorithms extraction carries out fusion treatment, obtains sample
The fusion feature vector f of facial imageB, and by PCA to fusion feature vector fBCarry out dimension-reduction treatment, obtain new fusion special
Levy vector FB:
New fusion feature vector F is calculated according to PLDAAWith new fusion feature vector FBContrast similarity, melted
The contrast Similarity value closed;
According to the contrast Similarity value merged and default contrast similarity threshold, determine final recognition result.
In being embodied as, described second determine module 204 specifically for:
Preset the weight coefficient of every kind of face recognition algorithms;
Contrast Similarity value according to every kind of face recognition algorithms and weight coefficient, obtain merge contrast Similarity value:
According to the contrast Similarity value merged and default contrast similarity threshold, determine final recognition result.
In being embodied as, described second determine module 204 specifically for:
Determine quantity and the quantity of all recognition results identifying successful recognition result;
Judge the quantity of all recognition results identifying whether the quantity of successful recognition result is more than 1/2nd;
If identifying, the quantity of successful recognition result is more than the quantity of all recognition results of 1/2nd, it is determined that final
Recognition result for identify successfully;
If identifying, the quantity of successful recognition result is less than or equal to the quantity of all recognition results of 1/2nd, the most really
Fixed final recognition result is recognition failures.
In being embodied as, described second determine module 204 specifically for:
Previously according to every kind of face recognition algorithms, every class facial image is identified, obtains every kind of face recognition algorithms pair
The discrimination of every class facial image;
Determine the object type belonging to described facial image to be identified and sample facial image;
The discrimination to every class facial image according to the object type determined and every kind of face recognition algorithms being previously obtained,
Choose the face recognition algorithms that discrimination corresponding to this object type is the highest;
The recognition result that the face recognition algorithms chosen obtains is defined as final recognition result.
A kind of preferred implementation that the dividing mode of the function above module only embodiment of the present invention is given, functional module
Dividing mode be not construed as limiting the invention.For convenience of description, each several part of system above is divided into function
Various modules or unit are respectively described.Certainly, when implementing the present invention can the function of each module or unit same or
Multiple softwares or hardware realize.
In sum, first the embodiment of the present invention is preset at least two face recognition algorithms;By every kind of recognition of face
Algorithm carries out feature extraction to facial image to be identified and sample facial image, respectively obtains every kind of face recognition algorithms corresponding
The characteristic vector of facial image to be identified and the characteristic vector of sample facial image;According to treating that every kind of face recognition algorithms is extracted
Identify characteristic vector and the characteristic vector of sample facial image of facial image, be calculated every kind of face recognition algorithms respectively
Contrast Similarity value;Contrast Similarity value according to every kind of face recognition algorithms determines the identification of every kind of face recognition algorithms respectively
Result;The characteristic vector of the facial image to be identified that every kind of face recognition algorithms is extracted is carried out fusion treatment and to every kind of face
The characteristic vector of the sample facial image that recognizer is extracted carries out fusion treatment, according to the fusion feature of facial image to be identified
The fusion feature vector of vector and sample facial image determines final recognition result;Or, to every kind of face recognition algorithms
Contrast Similarity value carries out fusion treatment, and the contrast Similarity value according to merging determines final recognition result;Or, to every kind
The recognition result of face recognition algorithms carries out fusion treatment, and the recognition result according to merging determines final recognition result.So,
The embodiment of the present invention merges multiple face recognition algorithms and carries out recognition of face, respectively to characteristic vector, contrast Similarity value or knowledge
Other result carries out fusion treatment, determines final recognition result according to the result of fusion treatment, solves the single face of use
The problem that recognizer is restricted in terms of the precision and accuracy of recognition of face, thus improve the precision of recognition of face.
The principle of above-described embodiment only illustrative present invention and effect thereof, not for limiting the present invention.Any ripe
Above-described embodiment all can be modified under the spirit and the scope of the present invention or change by the personage knowing this technology.Cause
This, have usually intellectual such as complete with institute under technological thought without departing from disclosed spirit in art
All equivalences become are modified or change, and must be contained by the claim of the present invention.
Claims (10)
1. the recognition methods merging multiple face recognition algorithms and system, it is characterised in that preset at least two face and know
Other algorithm, described method also includes:
By every kind of face recognition algorithms, facial image to be identified and sample facial image are carried out feature extraction, respectively obtain every
Plant characteristic vector and the characteristic vector of sample facial image of facial image to be identified corresponding to face recognition algorithms;
The characteristic vector of facial image to be identified extracted according to every kind of face recognition algorithms and the feature of sample facial image to
Amount, is calculated the contrast Similarity value of every kind of face recognition algorithms respectively;
Contrast Similarity value according to every kind of face recognition algorithms determines the recognition result of every kind of face recognition algorithms respectively;
The characteristic vector of the facial image to be identified that every kind of face recognition algorithms is extracted is carried out fusion treatment and to every kind of face
The characteristic vector of the sample facial image that recognizer is extracted carries out fusion treatment, according to the fusion feature of facial image to be identified
The fusion feature vector of vector and sample facial image determines final recognition result;Or,
The contrast Similarity value of every kind of face recognition algorithms is carried out fusion treatment, and the contrast Similarity value according to merging determines
Whole recognition result;Or,
The recognition result of every kind of face recognition algorithms is carried out decision-making treatment, determines final recognition result according to result.
Method the most according to claim 1, it is characterised in that the described people to be identified that every kind of face recognition algorithms is extracted
The characteristic vector of face image carries out fusion treatment the characteristic vector to the sample facial image that every kind of face recognition algorithms is extracted
Carry out fusion treatment, determine according to the fusion feature vector of facial image to be identified and the fusion feature vector of sample facial image
Final recognition result, including:
Characteristic vector for the facial image to be identified of every kind of face recognition algorithms extraction carries out fusion treatment, obtains to be identified
The fusion feature vector f of facial imageA, and by Principal Component Analysis Algorithm PCA to fusion feature vector fACarry out dimension-reduction treatment,
Obtain new fusion feature vector FA:
Characteristic vector for the sample facial image of every kind of face recognition algorithms extraction carries out fusion treatment, obtains sample face
The fusion feature vector f of imageB, and by PCA to fusion feature vector fBCarry out dimension-reduction treatment, obtain new fusion feature to
Amount FB:
New fusion feature vector F is calculated according to probability linear discriminant analysis algorithm PLDAAWith new fusion feature vector FBRight
Ratio similarity, obtains the contrast Similarity value merged;
According to the contrast Similarity value merged and default contrast similarity threshold, determine final recognition result.
Method the most according to claim 1, it is characterised in that the described contrast Similarity value to every kind of face recognition algorithms
Carrying out fusion treatment, the contrast Similarity value according to merging determines final recognition result, including:
Preset the weight coefficient of every kind of face recognition algorithms;
Contrast Similarity value according to every kind of face recognition algorithms and weight coefficient, obtain merge contrast Similarity value:
According to the contrast Similarity value merged and default contrast similarity threshold, determine final recognition result.
Method the most according to claim 1, it is characterised in that the described recognition result to every kind of face recognition algorithms is carried out
Decision-making treatment, determines final recognition result according to result, including:
Determine quantity and the quantity of all recognition results identifying successful recognition result;
Judge the quantity of all recognition results identifying whether the quantity of successful recognition result is more than 1/2nd;
If identifying, the quantity of successful recognition result is more than the quantity of all recognition results of 1/2nd, it is determined that final knowledge
Other result is for identify successfully;
If identifying, the quantity of successful recognition result is less than or equal to the quantity of all recognition results of 1/2nd, it is determined that
Whole recognition result is recognition failures.
5. according to the method described in any one of Claims 1-4, it is characterised in that the described knowledge to every kind of face recognition algorithms
Other result carries out decision-making treatment, determines final recognition result according to result, including:
Previously according to every kind of face recognition algorithms, every class facial image is identified, obtains every kind of face recognition algorithms to every class
The discrimination of facial image;
Determine the object type belonging to described facial image to be identified and sample facial image;
The discrimination to every class facial image according to the object type determined and every kind of face recognition algorithms being previously obtained, chooses
The face recognition algorithms that discrimination corresponding to this object type is the highest;
The recognition result that the face recognition algorithms chosen obtains is defined as final recognition result.
6. the identification system merging multiple face recognition algorithms, it is characterised in that described system includes: presetting module, spy
Levy extraction module, computing module, first determine module, second determine module;Wherein,
Described presetting module, is used for presetting at least two face recognition algorithms;
Described characteristic extracting module, for entering facial image to be identified and sample facial image by every kind of face recognition algorithms
Row feature extraction, respectively obtains characteristic vector and the sample face figure of every kind of facial image to be identified corresponding to face recognition algorithms
The characteristic vector of picture;
Described computing module, for characteristic vector and the sample of the facial image to be identified according to every kind of face recognition algorithms extraction
The characteristic vector of facial image, is calculated the contrast Similarity value of every kind of face recognition algorithms respectively;
Described first determines module, for determining that every kind of face is known respectively according to the contrast Similarity value of every kind of face recognition algorithms
The recognition result of other algorithm;
Described second determines module, for carrying out the characteristic vector of the facial image to be identified that every kind of face recognition algorithms is extracted
Fusion treatment also carries out fusion treatment to the characteristic vector of the sample facial image that every kind of face recognition algorithms is extracted, and knows according to waiting
The fusion feature vector of other facial image and the fusion feature vector of sample facial image determine final recognition result;Or,
The contrast Similarity value of every kind of face recognition algorithms is carried out fusion treatment, and the contrast Similarity value according to merging determines final
Recognition result;Or, the recognition result of every kind of face recognition algorithms is carried out decision-making treatment, determines according to result final
Recognition result.
System the most according to claim 6, it is characterised in that described second determine module specifically for:
Characteristic vector for the facial image to be identified of every kind of face recognition algorithms extraction carries out fusion treatment, obtains to be identified
The fusion feature vector f of facial imageA, and by PCA to fusion feature vector fACarry out dimension-reduction treatment, obtain new fusion special
Levy vector FA:
Characteristic vector for the sample facial image of every kind of face recognition algorithms extraction carries out fusion treatment, obtains sample face
The fusion feature vector f of imageB, and by PCA to fusion feature vector fBCarry out dimension-reduction treatment, obtain new fusion feature to
Amount FB:
New fusion feature vector F is calculated according to PLDAAWith new fusion feature vector FBContrast similarity, obtain merge
Contrast Similarity value;
According to the contrast Similarity value merged and default contrast similarity threshold, determine final recognition result.
System the most according to claim 6, it is characterised in that described second determine module specifically for:
Preset the weight coefficient of every kind of face recognition algorithms;
Contrast Similarity value according to every kind of face recognition algorithms and weight coefficient, obtain merge contrast Similarity value:
According to the contrast Similarity value merged and default contrast similarity threshold, determine final recognition result.
System the most according to claim 6, it is characterised in that described second determine module specifically for:
Determine quantity and the quantity of all recognition results identifying successful recognition result;
Judge the quantity of all recognition results identifying whether the quantity of successful recognition result is more than 1/2nd;
If identifying, the quantity of successful recognition result is more than the quantity of all recognition results of 1/2nd, it is determined that final knowledge
Other result is for identify successfully;
If identifying, the quantity of successful recognition result is less than or equal to the quantity of all recognition results of 1/2nd, it is determined that
Whole recognition result is recognition failures.
10. according to the system described in any one of claim 6 to 9, it is characterised in that described second determine module specifically for:
Previously according to every kind of face recognition algorithms, every class facial image is identified, obtains every kind of face recognition algorithms to every class
The discrimination of facial image;
Determine the object type belonging to described facial image to be identified and sample facial image;
The discrimination to every class facial image according to the object type determined and every kind of face recognition algorithms being previously obtained, chooses
The face recognition algorithms that discrimination corresponding to this object type is the highest;
The recognition result that the face recognition algorithms chosen obtains is defined as final recognition result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610634586.6A CN106250858B (en) | 2016-08-05 | 2016-08-05 | Recognition method and system fusing multiple face recognition algorithms |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610634586.6A CN106250858B (en) | 2016-08-05 | 2016-08-05 | Recognition method and system fusing multiple face recognition algorithms |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106250858A true CN106250858A (en) | 2016-12-21 |
CN106250858B CN106250858B (en) | 2021-08-13 |
Family
ID=58077677
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610634586.6A Active CN106250858B (en) | 2016-08-05 | 2016-08-05 | Recognition method and system fusing multiple face recognition algorithms |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106250858B (en) |
Cited By (26)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106650693A (en) * | 2016-12-30 | 2017-05-10 | 河北三川科技有限公司 | Multi-feature fusion identification algorithm used for human face comparison |
CN107563280A (en) * | 2017-07-24 | 2018-01-09 | 南京道熵信息技术有限公司 | Face identification method and device based on multi-model |
CN107729300A (en) * | 2017-09-18 | 2018-02-23 | 百度在线网络技术(北京)有限公司 | Processing method, device, equipment and the computer-readable storage medium of text similarity |
CN108197532A (en) * | 2017-12-18 | 2018-06-22 | 深圳云天励飞技术有限公司 | The method, apparatus and computer installation of recognition of face |
CN108491794A (en) * | 2018-03-22 | 2018-09-04 | 腾讯科技(深圳)有限公司 | The method and apparatus of face recognition |
CN108764350A (en) * | 2018-05-30 | 2018-11-06 | 苏州科达科技股份有限公司 | Target identification method, device and electronic equipment |
CN108776768A (en) * | 2018-04-19 | 2018-11-09 | 广州视源电子科技股份有限公司 | Image-recognizing method and device |
CN108921140A (en) * | 2018-08-07 | 2018-11-30 | 安徽云森物联网科技有限公司 | Pedestrian's recognition methods again |
CN109145758A (en) * | 2018-07-25 | 2019-01-04 | 武汉恩智电子科技有限公司 | A kind of recognizer of the face based on video monitoring |
CN109377624A (en) * | 2018-11-23 | 2019-02-22 | 卢伟涛 | A kind of door intelligent opening system based on facial image identification |
CN109624844A (en) * | 2018-12-05 | 2019-04-16 | 电子科技大学成都学院 | A kind of bus driving protection system based on image recognition and voice transmission control |
CN109918976A (en) * | 2017-12-13 | 2019-06-21 | 航天信息股份有限公司 | A kind of portrait alignment algorithm fusion method and its device |
CN110245573A (en) * | 2019-05-21 | 2019-09-17 | 平安科技(深圳)有限公司 | A kind of register method, apparatus and terminal device based on recognition of face |
CN110309744A (en) * | 2019-06-21 | 2019-10-08 | 武汉市公安局视频侦查支队 | A kind of suspect's recognition methods and device |
CN110334695A (en) * | 2019-07-18 | 2019-10-15 | 国网电子商务有限公司 | A kind of identity identifying method and device based on face characteristic |
CN110532856A (en) * | 2019-07-16 | 2019-12-03 | 公安部第一研究所 | A kind of face identification method of more algorithm fusions |
CN110688892A (en) * | 2019-08-20 | 2020-01-14 | 武汉烽火众智数字技术有限责任公司 | Portrait identification alarm method and system based on data fusion technology |
CN110728522A (en) * | 2019-10-11 | 2020-01-24 | 上海可充电子商务有限公司 | Face brushing public traffic system |
CN110908289A (en) * | 2018-09-17 | 2020-03-24 | 珠海格力电器股份有限公司 | Smart home control method and device |
CN111178259A (en) * | 2019-12-30 | 2020-05-19 | 八维通科技有限公司 | Recognition method and system supporting multi-algorithm fusion |
CN111177436A (en) * | 2018-11-09 | 2020-05-19 | 浙江宇视科技有限公司 | Face feature retrieval method, device and equipment |
CN111209937A (en) * | 2019-12-27 | 2020-05-29 | 深圳智链物联科技有限公司 | Charging curve model classification method and device and server |
CN112308070A (en) * | 2020-10-30 | 2021-02-02 | 深圳前海微众银行股份有限公司 | Certificate information identification method and device, equipment and computer readable storage medium |
CN112597823A (en) * | 2020-12-07 | 2021-04-02 | 深延科技(北京)有限公司 | Attention recognition method and device, electronic equipment and storage medium |
CN116434313A (en) * | 2023-04-28 | 2023-07-14 | 北京声迅电子股份有限公司 | Face recognition method based on multiple face recognition modules |
CN112308070B (en) * | 2020-10-30 | 2024-04-26 | 深圳前海微众银行股份有限公司 | Identification method and device for certificate information, equipment and computer readable storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102411708A (en) * | 2011-12-02 | 2012-04-11 | 湖南大学 | Face recognition method combining dual-tree complex wavelet transform and discrete wavelet transform |
CN102521561A (en) * | 2011-11-16 | 2012-06-27 | 湖南大学 | Face identification method on basis of multi-scale weber local features and hierarchical decision fusion |
CN102831413A (en) * | 2012-09-11 | 2012-12-19 | 上海中原电子技术工程有限公司 | Face identification method and face identification system based on fusion of multiple classifiers |
CN103136504A (en) * | 2011-11-28 | 2013-06-05 | 汉王科技股份有限公司 | Face recognition method and device |
CN104899579A (en) * | 2015-06-29 | 2015-09-09 | 小米科技有限责任公司 | Face recognition method and face recognition device |
-
2016
- 2016-08-05 CN CN201610634586.6A patent/CN106250858B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102521561A (en) * | 2011-11-16 | 2012-06-27 | 湖南大学 | Face identification method on basis of multi-scale weber local features and hierarchical decision fusion |
CN103136504A (en) * | 2011-11-28 | 2013-06-05 | 汉王科技股份有限公司 | Face recognition method and device |
CN102411708A (en) * | 2011-12-02 | 2012-04-11 | 湖南大学 | Face recognition method combining dual-tree complex wavelet transform and discrete wavelet transform |
CN102831413A (en) * | 2012-09-11 | 2012-12-19 | 上海中原电子技术工程有限公司 | Face identification method and face identification system based on fusion of multiple classifiers |
CN104899579A (en) * | 2015-06-29 | 2015-09-09 | 小米科技有限责任公司 | Face recognition method and face recognition device |
Cited By (35)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106650693A (en) * | 2016-12-30 | 2017-05-10 | 河北三川科技有限公司 | Multi-feature fusion identification algorithm used for human face comparison |
CN107563280A (en) * | 2017-07-24 | 2018-01-09 | 南京道熵信息技术有限公司 | Face identification method and device based on multi-model |
CN107729300A (en) * | 2017-09-18 | 2018-02-23 | 百度在线网络技术(北京)有限公司 | Processing method, device, equipment and the computer-readable storage medium of text similarity |
CN109918976A (en) * | 2017-12-13 | 2019-06-21 | 航天信息股份有限公司 | A kind of portrait alignment algorithm fusion method and its device |
CN109918976B (en) * | 2017-12-13 | 2021-04-02 | 航天信息股份有限公司 | Portrait comparison algorithm fusion method and device thereof |
CN108197532A (en) * | 2017-12-18 | 2018-06-22 | 深圳云天励飞技术有限公司 | The method, apparatus and computer installation of recognition of face |
WO2019120115A1 (en) * | 2017-12-18 | 2019-06-27 | 深圳励飞科技有限公司 | Facial recognition method, apparatus, and computer apparatus |
WO2019119505A1 (en) * | 2017-12-18 | 2019-06-27 | 深圳云天励飞技术有限公司 | Face recognition method and device, computer device and storage medium |
CN108491794A (en) * | 2018-03-22 | 2018-09-04 | 腾讯科技(深圳)有限公司 | The method and apparatus of face recognition |
CN108776768A (en) * | 2018-04-19 | 2018-11-09 | 广州视源电子科技股份有限公司 | Image-recognizing method and device |
WO2019200902A1 (en) * | 2018-04-19 | 2019-10-24 | 广州视源电子科技股份有限公司 | Image recognition method and device |
CN108764350A (en) * | 2018-05-30 | 2018-11-06 | 苏州科达科技股份有限公司 | Target identification method, device and electronic equipment |
CN109145758A (en) * | 2018-07-25 | 2019-01-04 | 武汉恩智电子科技有限公司 | A kind of recognizer of the face based on video monitoring |
CN108921140A (en) * | 2018-08-07 | 2018-11-30 | 安徽云森物联网科技有限公司 | Pedestrian's recognition methods again |
CN110908289A (en) * | 2018-09-17 | 2020-03-24 | 珠海格力电器股份有限公司 | Smart home control method and device |
CN111177436A (en) * | 2018-11-09 | 2020-05-19 | 浙江宇视科技有限公司 | Face feature retrieval method, device and equipment |
CN111177436B (en) * | 2018-11-09 | 2023-08-22 | 浙江宇视科技有限公司 | Face feature retrieval method, device and equipment |
CN109377624A (en) * | 2018-11-23 | 2019-02-22 | 卢伟涛 | A kind of door intelligent opening system based on facial image identification |
CN109624844A (en) * | 2018-12-05 | 2019-04-16 | 电子科技大学成都学院 | A kind of bus driving protection system based on image recognition and voice transmission control |
CN110245573B (en) * | 2019-05-21 | 2023-05-26 | 平安科技(深圳)有限公司 | Sign-in method and device based on face recognition and terminal equipment |
CN110245573A (en) * | 2019-05-21 | 2019-09-17 | 平安科技(深圳)有限公司 | A kind of register method, apparatus and terminal device based on recognition of face |
CN110309744A (en) * | 2019-06-21 | 2019-10-08 | 武汉市公安局视频侦查支队 | A kind of suspect's recognition methods and device |
CN110532856A (en) * | 2019-07-16 | 2019-12-03 | 公安部第一研究所 | A kind of face identification method of more algorithm fusions |
CN110334695A (en) * | 2019-07-18 | 2019-10-15 | 国网电子商务有限公司 | A kind of identity identifying method and device based on face characteristic |
CN110688892B (en) * | 2019-08-20 | 2023-01-03 | 武汉烽火众智数字技术有限责任公司 | Portrait identification alarm method and system based on data fusion technology |
CN110688892A (en) * | 2019-08-20 | 2020-01-14 | 武汉烽火众智数字技术有限责任公司 | Portrait identification alarm method and system based on data fusion technology |
CN110728522A (en) * | 2019-10-11 | 2020-01-24 | 上海可充电子商务有限公司 | Face brushing public traffic system |
CN111209937A (en) * | 2019-12-27 | 2020-05-29 | 深圳智链物联科技有限公司 | Charging curve model classification method and device and server |
CN111209937B (en) * | 2019-12-27 | 2024-03-29 | 深圳智链物联科技有限公司 | Classification method and device for charging curve model and server |
CN111178259A (en) * | 2019-12-30 | 2020-05-19 | 八维通科技有限公司 | Recognition method and system supporting multi-algorithm fusion |
CN112308070A (en) * | 2020-10-30 | 2021-02-02 | 深圳前海微众银行股份有限公司 | Certificate information identification method and device, equipment and computer readable storage medium |
CN112308070B (en) * | 2020-10-30 | 2024-04-26 | 深圳前海微众银行股份有限公司 | Identification method and device for certificate information, equipment and computer readable storage medium |
CN112597823A (en) * | 2020-12-07 | 2021-04-02 | 深延科技(北京)有限公司 | Attention recognition method and device, electronic equipment and storage medium |
CN116434313A (en) * | 2023-04-28 | 2023-07-14 | 北京声迅电子股份有限公司 | Face recognition method based on multiple face recognition modules |
CN116434313B (en) * | 2023-04-28 | 2023-11-14 | 北京声迅电子股份有限公司 | Face recognition method based on multiple face recognition modules |
Also Published As
Publication number | Publication date |
---|---|
CN106250858B (en) | 2021-08-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106250858A (en) | A kind of recognition methods merging multiple face recognition algorithms and system | |
CN100356388C (en) | Biocharacteristics fusioned identity distinguishing and identification method | |
CN106529468B (en) | A kind of finger vein identification method and system based on convolutional neural networks | |
He et al. | Performance evaluation of score level fusion in multimodal biometric systems | |
CN101246543B (en) | Examiner identity identification method based on bionic and biological characteristic recognition | |
CN101661557B (en) | Face recognition system and face recognition method based on intelligent card | |
CN107103281A (en) | Face identification method based on aggregation Damage degree metric learning | |
CN106446754A (en) | Image identification method, metric learning method, image source identification method and devices | |
Qin et al. | A fuzzy authentication system based on neural network learning and extreme value statistics | |
KR20030046007A (en) | Iris image processing and recognizing method for personal identification | |
CN108564040B (en) | Fingerprint activity detection method based on deep convolution characteristics | |
CN106203284B (en) | Method for detecting human face based on convolutional neural networks and condition random field | |
US11062120B2 (en) | High speed reference point independent database filtering for fingerprint identification | |
CN106228045A (en) | A kind of identification system | |
CN101187977A (en) | A face authentication method and device | |
CN107153780A (en) | The writing behavioural characteristic authentication method of electronic equipment is dressed based on wrist | |
CN106529377A (en) | Age estimating method, age estimating device and age estimating system based on image | |
CN109993042A (en) | A kind of face identification method and its device | |
CN105956570A (en) | Lip characteristic and deep learning based smiling face recognition method | |
CN104408468A (en) | Face recognition method based on rough set and integrated learning | |
Saxena et al. | Palm vein recognition using local tetra patterns | |
R Al-nima | Design a biometric identification system based on the fusion of hand geometry and backhand patterns | |
CN102819754A (en) | Fingerprint score fusion system and method based on Sigmoid expansion | |
Foroozandeh et al. | Persian signature verification based on fractal dimension using testing hypothesis | |
Latha et al. | A novel method for person authentication using retinal images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information |
Address after: 401120 No. 1 and 2, floors 3-8, building 11, No. 2, Huizhu Road, Yubei District, Chongqing Applicant after: CHONGQING ZHONGKE YUNCONG TECHNOLOGY Co.,Ltd. Address before: 401122 Central Sixth Floor of Mercury Science and Technology Building B, Central Section of Huangshan Avenue, Northern New District of Chongqing Applicant before: CHONGQING ZHONGKE YUNCONG TECHNOLOGY Co.,Ltd. |
|
CB02 | Change of applicant information | ||
GR01 | Patent grant | ||
GR01 | Patent grant |