CN107316023A - A kind of face identification system for being used to share equipment - Google Patents
A kind of face identification system for being used to share equipment Download PDFInfo
- Publication number
- CN107316023A CN107316023A CN201710501944.0A CN201710501944A CN107316023A CN 107316023 A CN107316023 A CN 107316023A CN 201710501944 A CN201710501944 A CN 201710501944A CN 107316023 A CN107316023 A CN 107316023A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- user
- sample
- mfrac
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/08—Payment architectures
- G06Q20/14—Payment architectures specially adapted for billing systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F17/00—Coin-freed apparatus for hiring articles; Coin-freed facilities or services
- G07F17/0042—Coin-freed apparatus for hiring articles; Coin-freed facilities or services for hiring of objects
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F17/00—Coin-freed apparatus for hiring articles; Coin-freed facilities or services
- G07F17/10—Coin-freed apparatus for hiring articles; Coin-freed facilities or services for means for safe-keeping of property, left temporarily, e.g. by fastening the property
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Accounting & Taxation (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Multimedia (AREA)
- Finance (AREA)
- Human Computer Interaction (AREA)
- General Health & Medical Sciences (AREA)
- Computer Security & Cryptography (AREA)
- Development Economics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of face identification system for being used to share equipment, belong to area of pattern recognition.The system obtains user images, constitutes positive and negative sample training user identity detection module, user identity is identified based on user identity detection module, determines that the related of the user preserved in vending system is extracted after user identity to be set and perform.Mode of the invention based on user biological feature recognition carries out authenticating user identification, when shared equipment is done shopping, it directly can on a sharing device be done shopping without using mobile phone terminal and pay expense, accurate authentication can be carried out, cost is reduced, and can further react the usage experience of user.
Description
【Technical field】
The present invention relates to area of pattern recognition, and in particular to a kind of face identification system for being used to share equipment.
【Background technology】
In recent years, with the fast development of computer industry, computer technology has goed deep into the life of people, has begun to
Gradually the living environment with us combines, and occurs in that the concept of shared equipment.So-called shared equipment, is exactly profit
With technologies such as computer, communication, sensor, household electrical appliances, the various shared equipment in family are all connected to together, by a terminal
It is controlled, so as to give people to provide an extremely easily living environment.
In shared device systems, in order to which preferably user oriented provides service, the generally usage behavior to user
It is collected, analyzes there is provided personalized service, to strengthen Consumer's Experience.Therefore, how to recognize that user becomes particularly important.
In the prior art, authenticating user identification can be carried out by way of user carries RFID card piece or other electronic installations, still
If the hardware unit can equally carry out authentication, and need to carry with by other people acquisitions of non-user
Corresponding hardware unit, reduces user experience, while improving the cost of vending system.In the prior art, it can also pass through
The system of pattern-recognition, for example, detect user biological feature, carries out user identity identification, and such as user's face recognition, fingerprint are known
Not, but system above needs user to be fixed, or perform identifying system required by authenticating step, such as user
Front is needed to stand on before face authentification device, fingerprint recognition needs user that finger is positioned on harvester, these systems
The operation of user is limited to a certain extent, reduces Consumer's Experience.
【The content of the invention】
In order to solve existing shared equipment identities authentication question, the present invention proposes a kind of for sharing equipment
Face identification system.
Goods equipment body, cargo scanner, identification authentication system, gateway, goods cabinet and Charging Detail Record unit;
The cargo scanner and Charging Detail Record unit are all connected with a control unit;Goods is placed on the goods cabinet, described
Cargo scanner scans goods, obtains goods information and transmits to described control unit;User is filled by the authentication
Put and the shared equipment is opened after confirmation identity, user takes goods and closed after the cabinet door of shared equipment, the cargo scanner
Goods is scanned, goods information is obtained and transmits to described control unit;The Charging Detail Record unit is carried out by described control unit
Charging, and deducted fees from user's registration account.
It is preferred that the identification authentication system include:Cloud server, image acquisition units, detection unit and result output
Unit, the cloud server includes database, and for preserving user's training sample, described image collecting unit is arranged on described
On shared device end, detection unit and the result output unit is set in a gateway, and spy is provided with the detection unit
Grader is levied, the feature classifiers are trained based on the training sample, the detection unit gathers single to described image
The user images of member collection are identified, and obtain identity authentication result, and the result output unit is by authenticating user identification result
Gateway is transferred to, gateway extracts the use habit of the shared device end of the user, sends to each shared device end, described
Shared device end is performed automatically to be met the operation of user's use habit or provides respective selection so that user is selected.
Preferably, the feature classifiers are trained based on improved Boost systems, feature point in the detection unit
The training process of class device comprises the following steps:
A1. user's training sample is useful in the whole body images that the user that gathers in advance continuously moves, training sample
What family was present includes N number of, i.e., N number of positive sample, and user is non-existent including L, i.e., L negative sample;
A2. the feature vector, X of training sample, i.e. X=(f are obtained1(x),f2(x),...fk(x))T, f (x) expression image samples
Eigen;Sample label is expressed as y, and y=1 represents positive sample label, and y=0 represents negative sample label, and X is the posteriority of positive sample
Probability can be expressed as
Wherein function δ (z) is defined as
So as to set up sorter model
Convolution (1) and formula (2), posterior probability can be expressed as
P (y=1 | X)=δ (Hk(X)) (4)
For the grader H of feature vector, Xk(X) it is represented by
Wherein,hk(fk(x) Weak Classifier) is represented, by K weak typing
Device can constitute strong classifier Hk(X)。
Positive negative sample is respectively put into two set:Positive sample set { X1j, j=0 ... N-1 } and negative sample set
{X0j, j=N ... N+L-1 }, constantly Weak Classifier is selected using multigroup sample from positive and negative sample set, and then structure
Produce discrimination highest assembled classifier, it is known that the posterior probability of single sample is expressed as
Pij=δ (Hk(Xij)) (6)
Wherein, the numbering of i value representative sample set, i=1 represents positive sample set, and i=0 represents negative sample set, j
It is sample number, setting grader hk(fk(xij)) in conditional probability be Gaussian Profile, i.e. conditional probability is
p(fk(xij) | y=1)~N (μ1,σ1)
p(fk(xij) | y=0)~N (μ0,σ0) (7)
Wherein, μ1,σ1,μ0,σ0Incremental update can be carried out
μ0,σ0Renewal it is identical with above formula;Can be in the hope of P by formula (7) and formula (8)ij, so, sample set i posteriority
Probability can be expressed as
A3. handmarking's human face region is carried out to first frame positive sample image, then obtained using Kalman filter per frame
Human face region in image, and handmarking's amendment is carried out every 10 two field pictures, missed with the accumulation for reducing Kalman filter
Difference;
A4. the human face region exported for Kalman filter, extracts the human face region feature, generates characteristic vector,
Characteristic vector set is formed, and weight is set, the posterior probability of this feature vector set is improved:
Wherein, wj0It is weighting function, monotone decreasing is expressed asWherein, | d (X1j)-d
(X10) | represent sample x1jThe Euclidean distance in region is manually demarcated to first frame, c is a constant;
A5. grader is selected after the posterior probability for determining sample set, the system representation of selection is
Wherein,It is the strong classifier for including k-1 Weak Classifier;L is the log-likelihood function of set, fixed
Justice is
L=∑si(yilogPi+(1-yi)log(1-Pi)) (11)
Preferably, the user images of the extraction are characterized as Haar-Like features.
Preferably, the user images that the detection unit is gathered to described image collecting unit are identified specifically, obtaining
User images are taken, the feature classifiers generated in characteristic vector, input detection unit are identified;
The identification process also includes, the face that the characteristic vector for being identified as user is exported with the Kalman filter
The characteristic vector in region carries out arest neighbors matching, and the arest neighbors matching is based on Euclidean distance, distance value is more than into predetermined threshold
Characteristic vector be added in the characteristic vector set of Kalman filter output.
The purpose so done is, is identified as the characteristic vector of user by the certification of feature classifiers, that is, is recognized
It is set to user, then is matched with the characteristic vector of human face region, it is assumed that the Euclidean distance between two samples is less than pre-
If threshold value, then the two samples just belong to same class, i.e., belong to same category with the sample in characteristic vector set;And if
Euclidean distance between two samples is more than predetermined threshold value, then it can be assumed that currently there is new sample, can be added
Enter into characteristic vector set, feature classifiers are trained again.
The beneficial effect realized of the present invention is:The system of authenticating user identification in the vending system that the present invention is used, no
User is needed additionally to carry hardware device, and user need to be only appeared in vending system, it is not necessary to specific identification step is performed,
Identification can be carried out to user, it is user-friendly, improve the usage experience of user.Also, use Kalman filtering
Output result weight is increased to the selection of Weak Classifier, improve the accuracy of classification..
【Brief description of the drawings】
Accompanying drawing described herein be for providing a further understanding of the present invention, constituting the part of the application, but
Inappropriate limitation of the present invention is not constituted, in the accompanying drawings:
Fig. 1 is the overall construction drawing that this hair shares equipment face identification system.
Fig. 2 is present system flow chart.
【Embodiment】
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning to end
Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached
The embodiment of figure description is exemplary, is only used for explaining the present invention, and is not considered as limiting the invention.On the contrary, this
All changes in the range of spirit and intension that the embodiment of invention includes falling into attached claims, modification and equivalent
Thing.
In the description of the invention, it is to be understood that term " first ", " second " etc. be only used for describe purpose, without
It is understood that to indicate or imply relative importance.In the description of the invention, it is necessary to which explanation, is provided unless otherwise clear and definite
And restriction, term " connected ", " connection " should be interpreted broadly, for example, it may be fixedly connected or be detachably connected,
Or be integrally connected;Can be mechanical connection or electrical connection;Can be joined directly together, intermediary can also be passed through
It is indirectly connected to.For the ordinary skill in the art, the tool of above-mentioned term in the present invention can be understood with concrete condition
Body implication.In addition, in the description of the invention, unless otherwise indicated, " multiple " are meant that two or more.
Any process described otherwise above or System describe are construed as in flow chart or herein, represent to include
Module, fragment or the portion of the code of one or more executable instructions for the step of realizing specific logical function or process
Point, and the scope of the preferred embodiment of the present invention includes other realization, wherein can not be by shown or discussion suitable
Sequence, including according to involved function by it is basic simultaneously in the way of or in the opposite order, carry out perform function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
The system applied of the present invention, share equipment body, cargo scanner, identification authentication system, gateway, goods cabinet and
Charging Detail Record unit;
The cargo scanner and Charging Detail Record unit are all connected with a control unit;Goods is placed on the goods cabinet, described
Cargo scanner scans goods, obtains goods information and transmits to described control unit;User is filled by the authentication
Put and the shared equipment is opened after confirmation identity, user takes goods and closed after the cabinet door of shared equipment, the cargo scanner
Goods is scanned, goods information is obtained and transmits to described control unit;The Charging Detail Record unit is carried out by described control unit
Charging, and deducted fees from user's registration account.
In the embodiment that the present invention is provided, identification authentication system is arranged on shared equipment body, authentication is utilized
The cabinet door and Charging Detail Record unit of the shared equipment of device control, wherein, the identification authentication system in the application is face identification device, profit
With the face image of collection consumer, customer identification is judged, first, consumer's enrollment status information is shared equipment backstage and protected
Consumer identification information is deposited, when consumer is using shared equipment, the identification authentication system collection being arranged in shared equipment disappears
The person's of expense facial image information, and matched with the consumer identification information that shared equipment backstage is stored, determine after user identity,
The shared equipment cabinet door of unblock, consumer takes goods and closed after the cabinet door of shared equipment, shares the cargo scanner scanning of equipment
Goods, obtains goods information and transmits to described control unit;The Charging Detail Record unit carries out charging by described control unit,
And deducted fees from user's registration account.
Wherein, user's registration account can be by setting the binding payer such as bank card or Alipay wechat accordingly
Formula, is paid after consumer is using shared equipment.
Referring to accompanying drawing 1, it illustrates the identification authentication system, including:Cloud server, image acquisition units, detection
Unit and result output unit, the cloud server include database, for preserving user's training sample, described image collection
Unit is arranged on the shared device end, and detection unit and the result output unit is set in a gateway, the detection
Feature classifiers are provided with unit, the feature classifiers are trained based on the training sample, the detection unit pair
The user images of described image collecting unit collection are identified, and obtain identity authentication result, the result output unit will be used
Family identity authentication result is transferred to gateway, and gateway extracts the use habit of the shared device end of the user, sends to each common
Enjoy device end, the shared device end perform automatically meet the operation of user's use habit or provide respective selection for
Family is selected.
Referring to accompanying drawing 2, the system flow chart of the present invention is shown.Wherein, the feature classifiers are based on improved Boost
System is trained, and the training process of feature classifiers comprises the following steps in the detection unit:
A1. user's training sample is useful in the whole body images that the user that gathers in advance continuously moves, training sample
What family was present includes N number of, i.e., N number of positive sample, and user is non-existent including L, i.e., L negative sample;
A2. the feature vector, X of training sample, i.e. X=(f are obtained1(x),f2(x),...fk(x))T, f (x) expression image samples
Eigen;Sample label is expressed as y, and y=1 represents positive sample label, and y=0 represents negative sample label, and X is the posteriority of positive sample
Probability can be expressed as
Wherein function δ (z) is defined as
So as to set up sorter model
Convolution (1) and formula (2), posterior probability can be expressed as
P (y=1 | X)=δ (Hk(X)) (4)
For the grader H of feature vector, Xk(X) it is represented by
Wherein,hk(fk(x) Weak Classifier) is represented, by K weak typing
Device can constitute strong classifier Hk(X)。
Positive negative sample is respectively put into two set:Positive sample set { X1j, j=0 ... N-1 } and negative sample set
{X0j, j=N ... N+L-1 }, constantly Weak Classifier is selected using multigroup sample from positive and negative sample set, and then structure
Produce discrimination highest assembled classifier, it is known that the posterior probability of single sample is expressed as
Pij=δ (Hk(Xij)) (6)
Wherein, the numbering of i value representative sample set, i=1 represents positive sample set, and i=0 represents negative sample set, j
It is sample number, setting grader hk(fk(xij)) in conditional probability be Gaussian Profile, i.e. conditional probability is
p(fk(xij) | y=1)~N (μ1,σ1)
p(fk(xij) | y=0)~N (μ0,σ0) (7)
Wherein, μ1,σ1,μ0,σ0Incremental update can be carried out
μ0,σ0Renewal it is identical with above formula;Can be in the hope of P by formula (7) and formula (8)ij, so, sample set i posteriority
Probability can be expressed as
A3. handmarking's human face region is carried out to first frame positive sample image, then obtained using Kalman filter per frame
Human face region in image, and handmarking's amendment is carried out every 10 two field pictures, missed with the accumulation for reducing Kalman filter
Difference;
A4. the human face region exported for Kalman filter, extracts the human face region feature, generates characteristic vector,
Characteristic vector set is formed, and weight is set, the posterior probability of this feature vector set is improved:
Wherein, wj0It is weighting function, monotone decreasing is expressed asWherein, | d (X1j)-d
(X10) | represent sample x1jThe Euclidean distance in region is manually demarcated to first frame, c is a constant;
A5. grader is selected after the posterior probability for determining sample set, the system representation of selection is
Wherein,It is the strong classifier for including k-1 Weak Classifier;L is the log-likelihood function of set, fixed
Justice is
L=∑si(yilogPi+(1-yi)log(1-Pi)) (11)
Preferably, the user images of the extraction are characterized as Haar-Li ke features.
Preferably, the user images that the detection unit is gathered to described image collecting unit are identified specifically, obtaining
User images are taken, the feature classifiers generated in characteristic vector, input detection unit are identified;
The identification process also includes, the face that the characteristic vector for being identified as user is exported with the Kalman filter
The characteristic vector in region carries out arest neighbors matching, and the arest neighbors matching is based on Euclidean distance, distance value is more than into predetermined threshold
Characteristic vector be added in the characteristic vector set of Kalman filter output.
The purpose so done is, is identified as the characteristic vector of user by the certification of feature classifiers, that is, is recognized
It is set to user, then is matched with the characteristic vector of human face region, it is assumed that the Euclidean distance between two samples is less than pre-
If threshold value, then the two samples just belong to same class, i.e., belong to same category with the sample in characteristic vector set;And if
Euclidean distance between two samples is more than predetermined threshold value, then it can be assumed that currently there is new sample, can be added
Enter into characteristic vector set, feature classifiers are trained again.
Described above is only the better embodiment of the present invention, therefore all constructions according to described in present patent application scope,
The equivalent change or modification that feature and principle are done, is included in the range of present patent application.
Claims (6)
1. a kind of face identification system for being used to share equipment, it is characterised in that the system includes:Shared equipment body, goods
Scanner, identification authentication system, gateway, goods cabinet and Charging Detail Record unit;
The cargo scanner and Charging Detail Record unit are all connected with a control unit;Goods is placed on the goods cabinet, the goods
Scanner scans goods, obtains goods information and transmits to described control unit;User is true by the identification authentication system
Recognize after identity and open the shared equipment, user takes goods and closed after the cabinet door of shared equipment, the cargo scanner scanning
Goods, obtains goods information and transmits to described control unit;The Charging Detail Record unit carries out charging by described control unit,
And deducted fees from user's registration account.
2. according to the system in claim 1, it is characterised in that:The identification authentication system includes:Cloud server, image are adopted
Collect unit, detection unit and result output unit, the cloud server includes database, for preserving user's training sample,
Described image collecting unit is arranged on the shared device end, and detection unit and the result output unit is arranged on gateway
In, feature classifiers are provided with the detection unit, the feature classifiers are trained based on the training sample, described
The user images that detection unit is gathered to described image collecting unit are identified, and obtain identity authentication result, and the result is defeated
Go out unit and authenticating user identification result be transferred to gateway, gateway extracts the use habit of the shared device end of the user,
Send to each shared device end, the shared device end is performed automatically to be met the operation of user's use habit or provide corresponding
Option is selected for user.
3. according to the system in claim 2, it is characterised in that:The feature classifiers are carried out based on improved Boost systems
The training process of feature classifiers comprises the following steps in training, the detection unit:
A1. user's training sample is to have user to deposit in the whole body images that the user that gathers in advance continuously moves, training sample
Include N number of, i.e., N number of positive sample, the non-existent L that includes of user is individual, i.e., L negative sample;
A2. the feature vector, X of training sample, i.e. X=(f are obtained1(x),f2(x),...fk(x))T, f (x) expression image pattern spies
Levy;Sample label is expressed as y, and y=1 represents positive sample label, and y=0 represents negative sample label, and X is the posterior probability of positive sample
It can be expressed as
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
<mo>|</mo>
<mi>X</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>|</mo>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>Y</mi>
<mo>=</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msub>
<mi>&Sigma;</mi>
<mrow>
<mi>y</mi>
<mo>&Element;</mo>
<mo>{</mo>
<mn>0</mn>
<mo>,</mo>
<mn>1</mn>
<mo>}</mo>
</mrow>
</msub>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>|</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>=</mo>
<mi>&delta;</mi>
<mrow>
<mo>(</mo>
<mi>l</mi>
<mi>n</mi>
<mo>(</mo>
<mfrac>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>|</mo>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>|</mo>
<mi>y</mi>
<mo>=</mo>
<mn>0</mn>
<mo>)</mo>
</mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>y</mi>
<mo>=</mo>
<mn>0</mn>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein function δ (z) is defined as
<mrow>
<mi>&delta;</mi>
<mrow>
<mo>(</mo>
<mi>z</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mrow>
<mn>1</mn>
<mo>+</mo>
<msup>
<mi>e</mi>
<mrow>
<mo>-</mo>
<mi>z</mi>
</mrow>
</msup>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</mrow>
So as to set up sorter model
<mrow>
<msub>
<mi>H</mi>
<mi>k</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>l</mi>
<mi>n</mi>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>|</mo>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>|</mo>
<mi>y</mi>
<mo>=</mo>
<mn>0</mn>
<mo>)</mo>
</mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>y</mi>
<mo>=</mo>
<mn>0</mn>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
</mrow>
Convolution (1) and formula (2), posterior probability can be expressed as
P (y=1 | X)=δ (Hk(X)) (4)
For the grader H of feature vector, Xk(X) it is represented by
<mrow>
<msub>
<mi>H</mi>
<mi>k</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>l</mi>
<mi>n</mi>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>|</mo>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>|</mo>
<mi>y</mi>
<mo>=</mo>
<mn>0</mn>
<mo>)</mo>
</mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>y</mi>
<mo>=</mo>
<mn>0</mn>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>K</mi>
</munderover>
<msub>
<mi>h</mi>
<mi>k</mi>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>5</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein,hk(fk(x) Weak Classifier) is represented, can be with by K Weak Classifier
Constitute strong classifier Hk(X);
Positive negative sample is respectively put into two set:Positive sample set { X1j, j=0 ... N-1 } and negative sample set { X0j, j=
N ... N+L-1 }, constantly Weak Classifier is selected using multigroup sample from positive and negative sample set, and then construct identification
Rate highest assembled classifier, it is known that the posterior probability of single sample is expressed as
Pij=δ (Hk(Xij)) (6)
Wherein, the numbering of i value representative sample set, i=1 represents positive sample set, and i=0 represents negative sample set, and j is sample
This numbering, setting grader hk(fk(xij)) in conditional probability be Gaussian Profile, i.e. conditional probability is
p(fk(xij) | y=1)~N (μ1,σ1)
p(fk(xij) | y=0)~N (μ0,σ0) (7)
Wherein, μ1,σ1,μ0,σ0Incremental update can be carried out
<mrow>
<msub>
<mi>&mu;</mi>
<mn>1</mn>
</msub>
<mo>&LeftArrow;</mo>
<msub>
<mi>&eta;&mu;</mi>
<mn>1</mn>
</msub>
<mo>+</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>-</mo>
<mi>&eta;</mi>
<mo>)</mo>
</mrow>
<mfrac>
<mn>1</mn>
<mi>N</mi>
</mfrac>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>|</mo>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mn>1</mn>
</mrow>
</munder>
<msub>
<mi>f</mi>
<mi>k</mi>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msub>
<mi>&sigma;</mi>
<mn>1</mn>
</msub>
<mo>&LeftArrow;</mo>
<msub>
<mi>&eta;&sigma;</mi>
<mn>1</mn>
</msub>
<mo>+</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>-</mo>
<mi>&eta;</mi>
<mo>)</mo>
</mrow>
<msqrt>
<mrow>
<mfrac>
<mn>1</mn>
<mi>N</mi>
</mfrac>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>|</mo>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mn>1</mn>
</mrow>
</munder>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<msub>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>)</mo>
<mo>-</mo>
<msub>
<mi>&mu;</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>8</mn>
<mo>)</mo>
</mrow>
</mrow>
μ0,σ0Renewal it is identical with above formula;Can be in the hope of P by formula (7) and formula (8)ij, so, sample set i posterior probability
It can be expressed as
<mrow>
<msub>
<mi>P</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mn>1</mn>
<mo>-</mo>
<munder>
<mi>&Pi;</mi>
<mi>j</mi>
</munder>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>-</mo>
<msub>
<mi>P</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>9</mn>
<mo>)</mo>
</mrow>
</mrow>
A3. handmarking's human face region is carried out to first frame positive sample image, then obtained using Kalman filter per two field picture
In human face region, and every 10 two field pictures carry out handmarking's amendment, to reduce the accumulated error of Kalman filter;
A4. the human face region exported for Kalman filter, extracts the human face region feature, generates characteristic vector, is formed
Characteristic vector set, and weight is set, improve the posterior probability of this feature vector set:
<mrow>
<msub>
<mi>P</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>=</mo>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
<mo>|</mo>
<msup>
<mi>X</mi>
<mo>+</mo>
</msup>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<mrow>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</munderover>
<msub>
<mi>w</mi>
<mrow>
<mi>j</mi>
<mn>0</mn>
</mrow>
</msub>
<msub>
<mi>P</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>12</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, wj0It is weighting function, monotone decreasing is expressed asWherein, | d (X1j)-d(X10) | table
This x of sample1jThe Euclidean distance in region is manually demarcated to first frame, c is a constant;
A5. grader is selected after the posterior probability for determining sample set, the system representation of selection is
<mrow>
<msub>
<mi>h</mi>
<mi>k</mi>
</msub>
<mo>=</mo>
<munder>
<mi>argmax</mi>
<mrow>
<mi>h</mi>
<mo>&Element;</mo>
<mrow>
<mo>{</mo>
<mrow>
<msub>
<mi>h</mi>
<mn>1</mn>
</msub>
<mn>...</mn>
<msub>
<mi>h</mi>
<mi>M</mi>
</msub>
</mrow>
<mo>}</mo>
</mrow>
</mrow>
</munder>
<mi>l</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>H</mi>
<mrow>
<mi>k</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>+</mo>
<mi>h</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>10</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein,It is the strong classifier for including k-1 Weak Classifier;L is the log-likelihood function of set, is defined as
L=∑si(yilog Pi+(1-yi)log(1-Pi)) (11)。
4. according to the system in claim 2, it is characterised in that:The user images of the extraction are characterized as Haar-Like features.
5. the system in Claims 2 or 3, it is characterised in that:The detection unit is gathered to described image collecting unit
User images be identified specifically, obtain user images, generate the feature classifiers in characteristic vector, input detection unit
It is identified;
The identification process also includes, the human face region that the characteristic vector for being identified as user is exported with the Kalman filter
Characteristic vector carry out arest neighbors matching, arest neighbors matching is based on Euclidean distance, distance value is more than to the spy of predetermined threshold
Vector is levied to be added in the characteristic vector set of the Kalman filter output.
6. according to the system in claim 2, it is characterised in that:The use that the detection unit is gathered to described image collecting unit
Family face is identified.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710501944.0A CN107316023A (en) | 2017-06-27 | 2017-06-27 | A kind of face identification system for being used to share equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710501944.0A CN107316023A (en) | 2017-06-27 | 2017-06-27 | A kind of face identification system for being used to share equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107316023A true CN107316023A (en) | 2017-11-03 |
Family
ID=60180370
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710501944.0A Pending CN107316023A (en) | 2017-06-27 | 2017-06-27 | A kind of face identification system for being used to share equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107316023A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109147199A (en) * | 2018-07-09 | 2019-01-04 | 保山市质量技术监督综合检测中心 | A kind of device management method and system based on fingerprint recognition |
CN109711473A (en) * | 2018-12-29 | 2019-05-03 | 北京沃东天骏信息技术有限公司 | Item identification method, equipment and system |
CN111080917A (en) * | 2019-12-04 | 2020-04-28 | 万翼科技有限公司 | Self-help article borrowing and returning method and related equipment |
CN111209567A (en) * | 2019-12-30 | 2020-05-29 | 北京邮电大学 | Method and device for judging perceptibility of improving robustness of detection model |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102609678A (en) * | 2011-01-24 | 2012-07-25 | 台湾色彩与影像科技股份有限公司 | Intelligent self-service system for face recognition |
CN103544775A (en) * | 2013-10-25 | 2014-01-29 | 上海煦荣信息技术有限公司 | Vending machines, vending system and vending method for same |
CN203773629U (en) * | 2013-11-06 | 2014-08-13 | 上海煦荣信息技术有限公司 | Intelligent self-service vending system |
CN105427459A (en) * | 2014-09-17 | 2016-03-23 | 黄浩庭 | Self-help vending method and vending machine |
CN205722151U (en) * | 2016-04-06 | 2016-11-23 | 上海英内物联网科技股份有限公司 | A kind of self-selecting type Intelligent vending machine with face identification functions |
CN106204918A (en) * | 2016-03-08 | 2016-12-07 | 青岛海尔特种电冰柜有限公司 | Intelligence selling cabinet and intelligence selling system |
CN106469369A (en) * | 2016-11-03 | 2017-03-01 | 林杰 | A kind of automatic selling method based on Internet of Things and free supermarket |
CN206097289U (en) * | 2016-04-26 | 2017-04-12 | 上海英内物联网科技股份有限公司 | Intelligent vending machine who has face identification function based on internet of things |
-
2017
- 2017-06-27 CN CN201710501944.0A patent/CN107316023A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102609678A (en) * | 2011-01-24 | 2012-07-25 | 台湾色彩与影像科技股份有限公司 | Intelligent self-service system for face recognition |
CN103544775A (en) * | 2013-10-25 | 2014-01-29 | 上海煦荣信息技术有限公司 | Vending machines, vending system and vending method for same |
CN203773629U (en) * | 2013-11-06 | 2014-08-13 | 上海煦荣信息技术有限公司 | Intelligent self-service vending system |
CN105427459A (en) * | 2014-09-17 | 2016-03-23 | 黄浩庭 | Self-help vending method and vending machine |
CN106204918A (en) * | 2016-03-08 | 2016-12-07 | 青岛海尔特种电冰柜有限公司 | Intelligence selling cabinet and intelligence selling system |
CN205722151U (en) * | 2016-04-06 | 2016-11-23 | 上海英内物联网科技股份有限公司 | A kind of self-selecting type Intelligent vending machine with face identification functions |
CN206097289U (en) * | 2016-04-26 | 2017-04-12 | 上海英内物联网科技股份有限公司 | Intelligent vending machine who has face identification function based on internet of things |
CN106469369A (en) * | 2016-11-03 | 2017-03-01 | 林杰 | A kind of automatic selling method based on Internet of Things and free supermarket |
Non-Patent Citations (1)
Title |
---|
孟繁杰 等: "基于球形摄像机模型的全景三维跟踪算法的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109147199A (en) * | 2018-07-09 | 2019-01-04 | 保山市质量技术监督综合检测中心 | A kind of device management method and system based on fingerprint recognition |
CN109711473A (en) * | 2018-12-29 | 2019-05-03 | 北京沃东天骏信息技术有限公司 | Item identification method, equipment and system |
CN111080917A (en) * | 2019-12-04 | 2020-04-28 | 万翼科技有限公司 | Self-help article borrowing and returning method and related equipment |
CN111080917B (en) * | 2019-12-04 | 2021-12-17 | 万翼科技有限公司 | Self-help article borrowing and returning method and related equipment |
CN111209567A (en) * | 2019-12-30 | 2020-05-29 | 北京邮电大学 | Method and device for judging perceptibility of improving robustness of detection model |
CN111209567B (en) * | 2019-12-30 | 2022-05-03 | 北京邮电大学 | Method and device for judging perceptibility of improving robustness of detection model |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ferrari et al. | On the personalization of classification models for human activity recognition | |
US8320643B2 (en) | Face authentication device | |
CN107122641B (en) | Intelligent equipment owner identification method and intelligent equipment owner identification device based on use habit | |
CN107316023A (en) | A kind of face identification system for being used to share equipment | |
CN106096576B (en) | A kind of intelligent Service method of robot | |
Yang et al. | Sieving regression forest votes for facial feature detection in the wild | |
Qin et al. | A fuzzy authentication system based on neural network learning and extreme value statistics | |
CN105184304B (en) | Image identification device and characteristic quantity data register method to image identification device | |
CN107220582A (en) | Recognize the driver of vehicle | |
CN108875491A (en) | Data-updating method, authenticating device and the system and non-volatile memory medium of face unlock certification | |
US9798942B2 (en) | Biometric recognition method with speed and security feature suitable for POS/ATM applications | |
CN107644218B (en) | The working method that crowded region behavior analyzes and determines is realized based on image collecting function | |
CN106303599A (en) | A kind of information processing method, system and server | |
WO2017192719A1 (en) | User specific classifiers for biometric liveness detection | |
CN102968645A (en) | Method for improving face recognition accuracy rate and adaptability through updating of images | |
Kaur et al. | Fusion in multimodal biometric system: A review | |
Sastry et al. | A team of continuous-action learning automata for noise-tolerant learning of half-spaces | |
CN111831989A (en) | Identity authentication method and device based on multi-mode PPG signal | |
Fu et al. | A feature-based on-line detector to remove adversarial-backdoors by iterative demarcation | |
Yang et al. | Retraining and dynamic privilege for implicit authentication systems | |
Takahashi et al. | A review of off-line mode dataset shifts | |
CN107330398A (en) | The method and device of iris recognition in a kind of shared selling apparatus | |
CN109308782B (en) | Behavior detection method and system for target object | |
Fernandez | Biometric sample quality and its application to multimodal authentication systems | |
CN114971628A (en) | Digital RMB biological payment method, system, terminal equipment and storage medium |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171103 |
|
RJ01 | Rejection of invention patent application after publication |