CN107527008A - Face identification system and its control method and control device - Google Patents
Face identification system and its control method and control device Download PDFInfo
- Publication number
- CN107527008A CN107527008A CN201710552991.8A CN201710552991A CN107527008A CN 107527008 A CN107527008 A CN 107527008A CN 201710552991 A CN201710552991 A CN 201710552991A CN 107527008 A CN107527008 A CN 107527008A
- Authority
- CN
- China
- Prior art keywords
- face
- image
- sample
- new images
- identification system
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Processing Or Creating Images (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a kind of face identification system and its control method and control device.When the run time that this method is included in face identification system meets predetermined condition, the face image for gathering face is used as more new images;Using updating image update sample database, wherein, sample database is used to store sample image, and sample image is the face image as face identification system identification sample;The face image of face is gathered as identification image;And match cognization image carries out recognition of face with the sample image in sample database.The present invention solves the problems, such as cumbersome during the sample image renewal of face identification system in the prior art.
Description
Technical field
The present invention relates to mode identification technology, in particular to a kind of face identification system and its control method
And control device.
Background technology
Face identification system is an emerging biological identification technology using face recognition technology as core, is the current world
The high-quality precision and sophisticated technology of sciemtifec and technical sphere tackling key problem.Its widely used regional characteristics analysis algorithm, has merged computer image processing technology
With biostatistics principle in one, portrait characteristic point is extracted from video using computer image processing technology, utilizes biology
Statistical principle carries out analysis founding mathematical models, has vast potential for future development.
Face identification system has a wide range of applications, including recognition of face access management system, recognition of face access control and attendance
System, recognition of face monitoring management etc..Wherein, with the rise of smart home, face identification system is in lifting house security
Property and convenience in terms of play an important role.Common face identification system is before first use, it is necessary to manual more new samples
Database, the face image of face is comparatively laborious as sample image typing, usual Input Process.After typing, lead to
Face corresponding to overmatching sample image and the face image identification collected, and then according to the sound of recognition result progress gate inhibition
Should.Inventor has found that face identification system of the prior art is after typing sample image first, not to sample database
In sample image be updated, and during face identification system long-time use, with advancing age, same people
The face image of face may change, and now may result in wrong identification, if it is desired to update sample database, still
It can only be manually entered.
For in correlation technique face identification system sample image update when it is cumbersome the problem of, not yet propose at present
Efficiently solve scheme.
The content of the invention
The invention provides a kind of face identification system and its control method and control device, at least to solve prior art
The problem of cumbersome during the sample image renewal of middle face identification system.
In order to solve the above technical problems, one side according to embodiments of the present invention, the invention provides a kind of knowledge of face
The control method of other system, the control method include:When the run time of face identification system meets predetermined condition, face is gathered
Face image be used as more new images;Using updating image update sample database, wherein, sample database is used to store sample
Image, sample image are the face images as face identification system identification sample;The face image of face is gathered as identification
Image;And match cognization image carries out recognition of face with the sample image in sample database.
Further, when the run time of face identification system meets predetermined condition, the face image for gathering face is made
Include for the step of more new images:, within a predetermined period of time, will when the run time of face identification system meets predetermined condition
The face image collected every time is used as more new images, obtains multiple more new images;Using updating image update sample data
The step of storehouse, includes:When the number of the more new images of same face reaches predetermined number, using belonging to the predetermined of same face
The renewal image update sample database of number, wherein, predetermined number is to belong to the sample graph of same face in sample database
The number of picture.
Further, after the face image of face is gathered as the step of more new images, using updating image update
Before the step of sample database, method also includes:The first more new images and the second more new images are matched, wherein, the first renewal
Image and the second more new images are any two in multiple more new images;As of the first more new images and the second more new images
When exceeding first threshold with degree, determine that the first more new images and the second more new images belong to same face.
Further, after the face image of face is gathered as the step of more new images, using updating image update
Before the step of sample database, method also includes:The depth of the night of matching the new images with first sample image, the 4th more new images with
First sample image, wherein, new images and the 4th more new images are any two in multiple more new images the depth of the night the, the first sample
This image is any one sample image in sample database;The matching degree of new images and first sample image surpasses when the depth of the night
Cross Second Threshold, and when the matching degree of the 4th more new images and first sample image exceedes Second Threshold, determine the 3rd renewal figure
Picture and the 4th more new images belong to same face.
Further, within a predetermined period of time, will be each when the run time of face identification system meets predetermined condition
The face image collected is used as more new images, and the step of obtaining multiple more new images includes:In working as face identification system
When preceding run time is zero, within a predetermined period of time, using the face image collected every time as more new images, obtain multiple
More new images.
Further, when the number of the more new images of same face reaches predetermined number, using belonging to same face
The renewal image update sample database of predetermined number includes:When the number of the more new images of same face reaches predetermined number
When, the more new images for the predetermined number for belonging to same face are stored to sample database.
Further, within a predetermined period of time, will be each when the run time of face identification system meets predetermined condition
The face image collected is used as more new images, and the step of obtaining multiple more new images includes:In working as face identification system
The interval of preceding run time and the run time of preceding face identification system when once updating sample database reached between the scheduled time
Every when, within a predetermined period of time, using the face image collected every time as more new images, obtain multiple more new images.
Further, the sample image that same face is belonged in sample database is a sample group, sample database bag
Multiple sample groups are included, the more new images for belonging to the predetermined number of same face are a renewal group, using belonging to same face
The step of renewal image update sample database of predetermined number, includes:Matching the new images and the sample in sample database just before dawn
This image, wherein, the just before dawn new images be any one more new images in renewal group;Judgement sample database is with the presence or absence of the
Two sample images, wherein, the second sample image is matching degree in sample database with the new images just before dawn more than the 3rd threshold value
Sample image;And when the second sample image in sample database be present, the second sample image institute is substituted using renewal group
Sample group.
Another aspect according to embodiments of the present invention, there is provided a kind of control device of face identification system, control dress
Put including:Update module, for when the run time of face identification system meets predetermined condition, gathering the face image of face
As more new images, using updating image update sample database, wherein, sample database is used to store sample image, sample
Image is the face image as face identification system identification sample;Identification module, for gathering the face image conduct of face
Image is identified, match cognization image carries out recognition of face with the sample image in sample database.
The third aspect according to embodiments of the present invention, there is provided a kind of face identification system, the face identification system include:
Camera device and control device, the control device are any one control device provided by the invention.
In the present invention, face identification system gathers the face of face when the run time of itself meets predetermined condition
Image is updated as more new images to the sample image in sample database automatically, without manual entry, saves user's hand
Dynamic operation, lifts Consumer's Experience.
Brief description of the drawings
Fig. 1 is the flow chart of the control method for the face identification system that the embodiment of the present invention 1 provides;
Fig. 2 is the flow chart of the control method for the face identification system that the embodiment of the present invention 2 provides;
Fig. 3 is the flow chart of the control method for the face identification system that the embodiment of the present invention 3 provides;
Fig. 4 is the flow chart of the control method for the face identification system that the embodiment of the present invention 4 provides;
Fig. 5 is the flow chart of the control method for the face identification system that the embodiment of the present invention 5 provides;
Fig. 6 is the block diagram of the control device for the face identification system that the embodiment of the present invention 6 provides;And
Fig. 7 is the control device block diagram for the face identification system that the embodiment of the present invention 7 provides.
Embodiment
Here exemplary embodiment will be illustrated in detail, its example is illustrated in the accompanying drawings.Following description is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent and the consistent all embodiments of the present invention.On the contrary, they be only with it is such as appended
The example of the consistent apparatus and method of some aspects being described in detail in claims, of the invention.
The present invention proposes a kind of face identification system and its control method and control device, face identification system of the invention
The certification scene of the indoor and outdoor scope such as factory, supermarket, school, hospital, residential quarter is widely portable to, particularly in smart home
In application, such as using the recognition result of face identification system as gate inhibition respond foundation, by the identification knot of face identification system
The foundation of fruit as operation intelligent appliance authority etc..Or other scenes are can also be applied to, the present invention is not defined to this.
Preset sample database is used to store sample image inside face identification system, and sample image is as recognition of face
The face image of system identification sample.In identification process, by the face image and sample database of the face collected
Sample image is matched, and then by matching degree compared with preset threshold value, draws recognition result.It is applied in face identification system
During smart home, if matching degree exceedes preset threshold value, gate inhibition opens or intelligent appliance authorization, for example, to air-conditioning
Authorization.
Wherein, the renewal passage time of sample database is monitored to automatically update, in recognition of face in face identification system
When the run time of system meets predetermined condition, start generation patterns, using the face image of the face collected as renewal figure
Picture, sample database is updated.
It is the face of face it should be noted that more new images mentioned in the present invention, identification image and sample image
Image, three titles are only used for distinguishing different purposes.More new images are the face for updating sample image in sample database
Portion's image;Identification image is face image to be identified;Sample image is used for match cognization figure to be stored in sample database
As and complete the face image of identification process.
Specifically, face identification system proposed by the present invention and its control method and each embodiment of control device will be
It is described in detail in following content.
Embodiment 1
Fig. 1 is the flow chart of the control method for the face identification system that the embodiment of the present invention 1 provides, as shown in figure 1, the people
The control method of face identifying system includes steps S102 to step S108.
Step S102:When the run time of face identification system meets predetermined condition, the face image for gathering face is made
For more new images.
In the present invention, when the zero point of the run time of face identification system can be that system makes to have access to electricity for the first time, or
Can also be system in use for some time by it is initial when, the timing by zero point the during operation of system.Default sample data
The update condition in storehouse, for example, when to set update condition be that system is upper for the first time electric, namely when run time is zero, by what is collected
Face image is used as more new images, to realize the first time of sample database renewal (namely first typing sample image).And for example,
After setting system operation for a period of time, after specific such as operation half a year, using the face image collected as more new images, with to existing
Sample image in some sample databases is updated.
If during the first time renewal of sample database, the face image for gathering face is only used as more new images, if
, can be simultaneously using face image as identification image, recognition of face system when being the renewal for the second time or afterwards of sample database
System performs normal recognition of face step.
Step S104:Using updating image update sample database.
After more new images are obtained, sample database is updated using more new images.For example, updated in first time
When, more new images are stored to sample database and complete to update;In renewal for the second time or afterwards, replaced using more new images
Existing sample image completes renewal in sample database.
Preferably, in order to improve the robustness of identification, in sample database, multiple face images of same face are stored
As sample image, for example, belonging to the number of the sample image of same face can set as system preset parameter before dispatching from the factory
For a definite value, or, the number can also be used as an adjustable systematic parameter, adjusted by user and determine a definite value,
Specifically, such as the definite value is set for 10.Based on this, when carrying out the renewal of sample database, in step s 102, at one
In predetermined amount of time, using the face image collected every time as more new images, multiple more new images, the scheduled time are obtained
Section can also be set to a definite value as system preset parameter before dispatching from the factory, or, the predetermined amount of time can as one
The systematic parameter of regulation, adjusted by user and determine a definite value, specifically, such as it is 10 days to set the predetermined amount of time.
In the multiple more new images collected, when the number of the more new images of same face reaches predetermined number,
Using the renewal image update sample database for belonging to the same face predetermined number, now, in the sample database after renewal
The number for belonging to the sample image of same face is predetermined number, so as in identification process, it is possible to increase system robustness, drop
Low identification error rate.
Wherein, after above-mentioned steps S102, before step S104, following method can be used to judge two more new images
Whether same face is belonged to.
In the first optional embodiment, using the first more new images and the second more new images as in multiple more new images
Any two exemplified by illustrate:Match the first more new images and the second more new images, when the first more new images and second more
When the matching degree of new images exceedes first threshold, determine that the first more new images and the second more new images belong to same face, for example,
It is 70% to set first threshold.
In second of optional embodiment, using new images depth of the night the and the 4th more new images as in multiple more new images
Any two, by first sample image be sample database in any one sample image exemplified by illustrate:Matching the
The depth of the night new images and first sample image, the 4th more new images and first sample image, new images and first sample when the depth of the night
The matching degree of image exceedes Second Threshold, and when the matching degree of the 4th more new images and first sample image exceedes Second Threshold,
The depth of the night of determining the new images and the 4th more new images belong to same face, for example, it is 70% to set first threshold.
Step S106:The face image of face is gathered as identification image.
Step S108:Match cognization image carries out recognition of face with the sample image in sample database.
In this embodiment, face identification system gathers the face of face when the run time of itself meets predetermined condition
Portion's image is updated as more new images to the sample image in sample database automatically, without manual entry, saves user
It is manually operated, lift Consumer's Experience.
Embodiment 2
The embodiment 2 provides a kind of preferred embodiment on the basis of above-described embodiment 1, and something in common refers to
State the associated description of embodiment 1.Fig. 2 is the flow chart of the control method for the face identification system that the embodiment of the present invention 2 provides, such as
Shown in Fig. 2, the control method of the face identification system includes steps S202 to step S208.
Step S202:When the current run time of face identification system is zero, within a predetermined period of time, will gather every time
The face image arrived is used as more new images, obtains multiple more new images.
In electricity operation on face identification system is first, the current run time of face identification system is zero, now pre-
In fixed a period of time, multi collect face image obtains multiple more new images;Or face identification system is in use
When being performed initialization, sample database is cleared, meanwhile, the current run time of face identification system is initialized to zero,
Now in predetermined period of time, multi collect face image obtains multiple more new images.
Step S204:When the number of the more new images of same face reaches predetermined number, the pre- of same face will be belonged to
The more new images for determining number are stored to sample database.
The more new images of same face are added up, after reaching predetermined number, the predetermined number of same face will be belonged to
More new images store to sample database, completed sample according to the sample image of a face in storehouse typing.
Wherein, directly will the storage of more new images when obtaining first more new images., will when subsequently obtaining more new images
The more new images newly obtained are matched with the more new images stored, when the matching degree of the two exceedes first threshold, really
Belong to same face both fixed, will be stored both as a renewal group;When the matching degree of the two is not less than first threshold, it is determined that
The two is not belonging to same face.If that the more new images newly obtained are matched with any more new images stored
During with degree not less than first threshold, the more new images newly obtained are individually stored.Reach predetermined for the number of more new images
The renewal group of number, the more new images of predetermined number are stored to sample database.
Step S206:After predetermined amount of time, the face image of face is gathered as identification image.
Step S208:Match cognization image carries out recognition of face with sample image.
In this embodiment, in electricity operation on face identification system is first, or, face identification system is using process
In when being performed initialization, first automatic data collection face image is used as more new images in a predetermined amount of time, updates sample data
Sample image in storehouse, after predetermined amount of time, automatic data collection face image is identified as identification image, updates and identifies
Process without manually typing, paid close attention to without user, save the operation being manually entered, lift Consumer's Experience.
Embodiment 3
The embodiment 23 provides a kind of preferred embodiment on the basis of above-described embodiment 1, and something in common refers to
The associated description of above-described embodiment 1.Fig. 3 is the flow chart of the control method for the face identification system that the embodiment of the present invention 3 provides,
As shown in figure 3, the control method of the face identification system includes steps S302 to step S312.
Step S302:The recognition of face when the current run time of face identification system is with preceding once renewal sample database
When the interval of the run time of system reaches predetermined time interval, within a predetermined period of time, the face image that will be collected every time
More new images are used as, obtain multiple more new images.
The embodiment sets the face identification system interval scheduled time once to be automatically updated.Wherein, recognition of face system
Since zero point after timing, run time gradually increases the run time of system.Every time after renewal sample database, to sample data
Run time when storehouse updates is recorded, the face identification system when current run time is with preceding once renewal sample database
The interval of run time when reaching predetermined time interval, in predetermined period of time, multi collect face image obtains more
Individual more new images.
Wherein, in sample database, the sample image for belonging to same face is a sample group, and sample database includes
Multiple sample groups.For obtained multiple more new images, will belong to same face predetermined number more new images for one more
New group.
Step S304:Matching the new images and the sample image in sample database just before dawn.
Wherein, the just before dawn new images be any one more new images in certain renewal group.
Step S306:Judgement sample database whether there is the second sample image.
Wherein, the second sample image be in sample database with the just before dawn new images matching degree more than the 3rd threshold value sample
This image.
Step S308:When the second sample image in sample database be present, the second sample image is substituted using renewal group
The sample group at place.
Wherein, renewal group can be at the end of predetermined amount of time, by being matched between obtained multiple more new images
Formed, matching degree is more than the more new images of first threshold as a renewal group.
Step S310:The face image of face is gathered as identification image.
Wherein, the step can be carried out after predetermined amount of time terminates, and can also be carried out simultaneously with above-mentioned steps S3302.
Step S312:Match cognization image carries out recognition of face with sample image.
In this embodiment, during face identification system is run, at interval of the scheduled time carry out once it is automatic more
Newly, avoid changing caused identification mistake when same user's face image changes over time, also, in renewal,
Automatic data collection face image is used as more new images in one predetermined amount of time, updates the sample image in sample database, updated
Cheng Wuxu manually typings, paid close attention to without user, save the operation being manually entered, lift Consumer's Experience.
Embodiment 4
The embodiment 4 provides a kind of preferred embodiment on the basis of above-described embodiment 1-3, and something in common refers to
Above-described embodiment 1-3 associated description.Fig. 4 is the flow of the control method for the face identification system that the embodiment of the present invention 4 provides
Figure, as shown in figure 4, the control method of the face identification system includes steps S402 to step S430.
Step S402:In the upper electric initial launch of face identification system, start the camera device of face identification system, with
Using the face image of collection face as more new images, and start timer and start timing.
Step S404:When obtaining first more new images, a newly-built renewal group, and first more new images are deposited
Store up to the renewal group.
Step S406:After first, when obtaining more new images every time, updated with one in already present renewal group
Image is matched.
Step S408:When the matching degree of obtained more new images and a more new images in some already present renewal group
During more than first threshold, judge whether the number of the more new images in the renewal group reaches predetermined number.
Step S410:When the more new images in the renewal group are not up to predetermined number, the more new images that this is obtained are deposited
Storage is to the renewal group, and when the more new images in the renewal group reach predetermined number, the more new images that this is obtained abandon, by this
Renewal group is stored to sample database, forms a sample group.
Step S412:When obtained more new images and the more new images in already present any renewal group matching degree not
During more than first threshold, a newly-built renewal group, and the face image that this time is collected is stored to newly-built renewal group.
Step S414:When the time of timer reaching for first scheduled time, timer is reset, and the face that will gather face
Portion's image carries out recognition of face as identification image, match cognization image with the sample image in sample database.
Step S416:When the time of timer reaching for second scheduled time, the face image of face will be gathered as more
New images.
Step S418:When obtaining more new images, more new images are matched with a sample image in sample group.
Step S420:When the matching degree of a sample image in obtained more new images and some sample group is more than second
During threshold value, judge that renewal group whether there is corresponding to the sample group.
Step S422:If in the presence of renewal group corresponding to the sample group, judge the number of the more new images in the renewal group
Whether predetermined number is reached.
Step S424:When the more new images in the renewal group are not up to predetermined number, the more new images that this is obtained are deposited
Storage is to the renewal group, and when the more new images in the renewal group reach predetermined number, the more new images that this is obtained abandon, and utilize
The renewal group substitutes sample group corresponding to the renewal group.
Step S426:If renewal group is not present corresponding to the sample group, renewal group corresponding to newly-built sample group,
And the more new images are stored to the renewal group.
Step S428:When obtained more new images and a sample image in any one sample group matching degree not
During more than Second Threshold, the more new images are abandoned.
Step S430:When the time of timer reaching for first scheduled time, timer is reset, and the face that will gather face
Portion's image carries out recognition of face as identification image, match cognization image with the sample image in sample database.
In this embodiment, when face identification system is upper first electric, automatic data collection facial image is carried out to sample database
Renewal, namely typing first, without manual entry;After face identification system runs a period of time, automatic data collection facial image
Sample database is updated to sample database, without manual entry, while the sample in sample database can be ensured
Image sources occur in the face image of the face of nearlyer a period of time, the face image that can avoid increasing face due to the time
Identification mistake caused by change.
Embodiment 5
The embodiment 5 provides a kind of preferred embodiment on the basis of above-described embodiment 1-4, and something in common refers to
Above-described embodiment 1-4 associated description.Fig. 5 is a kind of control method for face identification system that the embodiment of the present invention 5 provides
Flow chart, the face identification system of the embodiment have main control module and camera, have face identification functions.This system makes
With preceding, the face image of the face without being manually entered user, system can gather use successively within a week by camera
The face image at family, the similarity (namely matching degree) of the face image of 10 faces collected of camera all reach 70%
When, then it is assumed that it is the face of same person, and the storage of these face images is arrived to the face database (namely renewal group) of user 1,
When collecting 10 human face similarity degrees of second people and all reaching 70%, then user 2 is arrived into the face image storage of second people
Face database, the like, the face of all domestic consumers is all automatically stored to the face database of corresponding individual within a week,
Having stored the user of face will not be collected and store again, after a week, the face of the new user of system no longer automatic input, when
When personage occurs in camera picture, system can gather the face image of personage, and with store domestic consumer face image (namely
Sample image) contrasted, when similarity is less than 80%, then it is assumed that picture occurs that stranger, wherein, if a week
After need the new domestic consumer of typing, can use be manually entered.
The method of this automatic input face can save the operation for being manually entered face.
After face automatic input, system can resurvey the face image for having stored user every 6 months, if 10 times are adopted
The face collected all reaches 70% with storing the similarity of face, then the face database of this user is updated, because the increasing with the age
Long, face may change, and update a face database every half a year, it is ensured that recognition of face is accurate.
Using the embodiment, first in use, automatic input face, is saved manually operated;In use, automatically
Update face database so that face identification system can accurately identify face for a long time.
It is the embodiment of the control method of face identification system provided by the invention above, present invention also offers face knowledge
, it is necessary to explanation, the control device of face identification system provided by the invention can be used for performing this hair for the control device of other system
The control method of any one face identification system of bright offer.
Embodiment 6
Fig. 6 is the block diagram of the control device for the face identification system that the embodiment of the present invention 6 provides, as shown in fig. 6, the control
Device includes update module 22 and identification module 24.
Wherein, update module 22 is used for when the run time of face identification system meets predetermined condition, gathers face
Face image is used as more new images, using updating image update sample database, wherein, sample database is used to store sample graph
Picture, sample image are the face images as face identification system identification sample.
The face image that identification module 24 is used to gather face enters as identification image, match cognization image with sample image
Row recognition of face.
In this embodiment, face identification system gathers the face of face when the run time of itself meets predetermined condition
Portion's image is updated as more new images to the sample image in sample database automatically, without manual entry, saves user
It is manually operated, lift Consumer's Experience.
Preferably, update module 22 is used for when the run time of face identification system meets predetermined condition, in pre- timing
Between in section, using the face image collected every time as more new images, obtain multiple more new images, and when same face
When the number of more new images reaches predetermined number, the renewal image update sample data for the predetermined number for belonging to same face is utilized
Storehouse, wherein, predetermined number is the number for the sample image for belonging to same face in sample database.
Using the preferred embodiment, multiple face images are gathered as sample image to the sample image in sample database
It is updated, the sample image of same face is multiple in sample database, lifts the robustness of face identification system, reduces and knows
Not mistake.
Preferably, the control device also includes the first matching module, and the face image for update module collection face is made
After more new images, figure is updated using the first more new images and second before updating image update sample database, are matched
Picture, wherein, the first more new images and the second more new images are any two in multiple more new images, when the first more new images and
When the matching degree of second more new images exceedes first threshold, determine that the first more new images and the second more new images belong to same people
Face.
Using the preferred embodiment, two more new images are determined by being matched between the more new images collected
Whether same face is belonged to, and without access-sample database, data processing speed faster, reduces Installed System Memory consumption.
Preferably, the control device also includes the second matching module, and the face image for update module collection face is made
After more new images, using before updating image update sample database, new images and first sample figure the depth of the night matching the
As, the 4th more new images and first sample image, wherein, new images and the 4th more new images is in multiple more new images the depth of the night the
Any two, first sample image is any one sample image in sample database, the new images and first when the depth of the night
The matching degree of sample image exceedes Second Threshold, and the matching degree of the 4th more new images and first sample image exceedes Second Threshold
When, new images and the 4th more new images belong to same face the depth of the night determining.
Using the preferred embodiment, by the way that the sample image in the more new images collected and sample database is carried out
Match somebody with somebody and determine whether two more new images belong to same face, can be it is determined that whether two more new images belong to same face
Meanwhile it is determined that the more new images target to be updated, reduces matching times, lifting data processing speed.
Preferably, update module 22 is used for when the current run time of face identification system is zero, in predetermined amount of time
It is interior, using the face image collected every time as more new images, obtain multiple more new images.
Using the preferred embodiment, when being zero at runtime, automatic data collection face image is as more new images to sample
Database is updated, and can save user's manually first typing face image, lifts Consumer's Experience.
It is further preferred that update module 22 is used for when the number of the more new images of same face reaches predetermined number,
The more new images for the predetermined number for belonging to same face are stored to sample database.
Using the preferred embodiment, when being zero at runtime, it will directly belong to the renewal of the predetermined number of same face
Image is stored to sample database and is updated, and quickly realizes the renewal of sample database.
Preferably, update module 22 is used to once update sample data with preceding in the current run time of face identification system
When the interval of the run time of face identification system reaches predetermined time interval during storehouse, within a predetermined period of time, it will gather every time
The face image arrived is used as more new images, obtains multiple more new images.
Using the preferred embodiment, at interval of the scheduled time, automatic data collection face image is as more new images to sample number
It is updated according to storehouse, the identification mistake caused by the face image of time growth face changes can be avoided.
It is further preferred that the sample image for belonging to same face in sample database is a sample group, sample data
Storehouse includes multiple sample groups, and the more new images for belonging to the predetermined number of same face are a renewal group, and update module 22 is used for
Matching the new images and the sample image in sample database just before dawn, wherein, the just before dawn new images be any one in renewal group
Individual more new images, judgement sample database whether there is the second sample image, wherein, the second sample image is in sample database
With just before dawn new images matching degree more than the 3rd threshold value sample image, when the second sample image in sample database being present
When, the sample group where the second sample image is substituted using renewal group.
Using the preferred embodiment, when the matching degree of the more new images and the sample image in sample database that collect surpasses
When crossing predetermined value, substitute the sample group where the sample image using the renewal group where more new images, ensure more new images with
Sample image belongs to same face, avoids updating mistake.
Embodiment 7
Fig. 7 is the block diagram for the face identification system that the embodiment of the present invention 7 provides, as shown in fig. 7, the face identification system bag
Camera device 40 and control device 20 are included, wherein, control device 20 is any one recognition of face provided in an embodiment of the present invention
The control device of system.
The face identification system provided using the embodiment, face identification system meet predetermined bar in the run time of itself
During part, the face image for gathering face is updated as more new images to the sample image in sample database automatically, without
Manual entry, it is manually operated to save user, lifts Consumer's Experience.
On the device in above-described embodiment, wherein unit, module performs the concrete mode of operation relevant
It is described in detail in the embodiment of this method, explanation will be not set forth in detail herein.
Those skilled in the art will readily occur to the present invention its after considering specification and putting into practice invention disclosed herein
Its embodiment.The application be intended to the present invention any modification, purposes or adaptations, these modifications, purposes or
Person's adaptations follow the general principle of the present invention and the common knowledge in the art do not invented including the present invention
Or conventional techniques.Description and embodiments are considered only as exemplary, and true scope and spirit of the invention are by following
Claim is pointed out.
It should be appreciated that the invention is not limited in the precision architecture for being described above and being shown in the drawings, and
And various modifications and changes can be being carried out without departing from the scope.The scope of the present invention is only limited by appended claim.
Claims (10)
- A kind of 1. control method of face identification system, it is characterised in thatWhen the run time of face identification system meets predetermined condition, the face image for gathering face is used as more new images;Using the renewal image update sample database, wherein, the sample database is used to store sample image, the sample This image is the face image as face identification system identification sample;The face image of face is gathered as identification image;AndMatch the identification image and carry out recognition of face with the sample image in the sample database.
- 2. the control method of face identification system according to claim 1, it is characterised in thatWhen the run time of face identification system meets predetermined condition, step of the face image of face as more new images is gathered Suddenly include:When the run time of face identification system meets predetermined condition, within a predetermined period of time, the face that will be collected every time Portion's image is used as more new images, obtains multiple more new images;Included using the step of renewal image update sample database:When the number of more new images described in same face reaches During to predetermined number, using sample database described in the renewal image update for the predetermined number for belonging to same face, Wherein, the predetermined number is the number for the sample image for belonging to same face in the sample database.
- 3. the control method of face identification system according to claim 2, it is characterised in that in face's figure of collection face After as the step of more new images, before the step of renewal image update sample database, methods described is also Including:The first more new images and the second more new images are matched, wherein, the described first more new images are with the described second more new images Any two in multiple more new images;When the matching degree of the described first more new images and the described second more new images exceedes first threshold, described first is determined more New images and the described second more new images belong to same face.
- 4. the control method of face identification system according to claim 2, it is characterised in that in face's figure of collection face After as the step of more new images, before the step of renewal image update sample database, methods described is also Including:The depth of the night of matching the new images with first sample image, the 4th more new images and the first sample image, wherein, described the The depth of the night new images and the 4th more new images be any two in multiple more new images, the first sample image is Any one sample image in the sample database;The matching degree of new images and the first sample image exceedes Second Threshold, and the 4th renewal figure when described the depth of the night As and the matching degree of the first sample image exceed the Second Threshold when, new images and the described 4th depth of the night determining described the More new images belong to same face.
- 5. the control method of face identification system according to claim 2, it is characterised in thatWhen the run time of face identification system meets predetermined condition, within a predetermined period of time, the face that will be collected every time Image is used as more new images, and the step of obtaining multiple more new images includes:In the current fortune of the face identification system When the row time is zero, within a predetermined period of time, using the face image collected every time as more new images, obtain multiple described More new images.
- 6. the control method of face identification system according to claim 5, it is characterised in thatWhen the number of more new images described in same face reaches the predetermined number, using belonging to the described pre- of same face Sample database described in determining the renewal image update of number includes:When the number of more new images described in same face reaches During the predetermined number, more new images described in the predetermined number of same face will be belonged to and stored to the sample data Storehouse.
- 7. the control method of face identification system according to claim 2, it is characterised in thatWhen the run time of face identification system meets predetermined condition, within a predetermined period of time, the face that will be collected every time Image is used as more new images, and the step of obtaining multiple more new images includes:In the current fortune of the face identification system The row time with it is preceding once update the sample database when face identification system the interval of run time reach pre- timing Between when being spaced, within a predetermined period of time, using the face image collected every time as more new images, obtain multiple renewals Image.
- 8. the control method of face identification system according to claim 7, it is characterised in that belong in the sample database It is a sample group in the sample image of same face, the sample database includes multiple sample groups, belongs to same The more new images of the predetermined number of one face are a renewal group, utilize the predetermined number for belonging to same face The renewal image update described in sample database the step of include:Matching the new images and the sample image in the sample database just before dawn, wherein, described the just before dawn new images be renewal Any one of more new images in group;Judge that the sample database whether there is the second sample image, wherein, second sample image is the sample number According in storehouse with described just before dawn new images matching degree more than the 3rd threshold value sample image;AndWhen second sample image be present in the sample database, second sample graph is substituted using the renewal group As the sample group at place.
- A kind of 9. control device of face identification system, it is characterised in that including:Update module, the face image for when the run time of face identification system meets predetermined condition, gathering face are made For more new images, image update sample database is updated using described, wherein, the sample database is used to store sample graph Picture, the sample image are the face images as face identification system identification sample;Identification module, the face image for gathering face match the identification image and the sample number as identification image Recognition of face is carried out according to the sample image in storehouse.
- 10. a kind of face identification system, it is characterised in that including camera device and control device, the control device is right It is required that the control device described in 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710552991.8A CN107527008A (en) | 2017-07-07 | 2017-07-07 | Face identification system and its control method and control device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710552991.8A CN107527008A (en) | 2017-07-07 | 2017-07-07 | Face identification system and its control method and control device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107527008A true CN107527008A (en) | 2017-12-29 |
Family
ID=60748915
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710552991.8A Pending CN107527008A (en) | 2017-07-07 | 2017-07-07 | Face identification system and its control method and control device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107527008A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109492560A (en) * | 2018-10-26 | 2019-03-19 | 深圳力维智联技术有限公司 | Facial image Feature fusion, device and storage medium based on time scale |
CN109933265A (en) * | 2018-11-08 | 2019-06-25 | 闽江学院 | A kind of display methods and device of automatic identification |
CN111062342A (en) * | 2019-12-20 | 2020-04-24 | 中国银行股份有限公司 | Debugging method and device of face recognition system |
CN111126147A (en) * | 2019-11-22 | 2020-05-08 | 北京迈格威科技有限公司 | Image processing method, device and electronic system |
CN111639915A (en) * | 2020-05-29 | 2020-09-08 | 广东电网有限责任公司 | Staff data management system |
CN112148908A (en) * | 2020-10-23 | 2020-12-29 | 北京百度网讯科技有限公司 | Image database updating method and device, electronic equipment and medium |
CN112818784A (en) * | 2021-01-22 | 2021-05-18 | 浙江大华技术股份有限公司 | Control method and device of access control equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101231692A (en) * | 2007-12-24 | 2008-07-30 | 浙江金指码科技有限公司 | Learning type intellectual fingerprint identification comparison method adjusted by fingerprint eigenvalue |
CN105869235A (en) * | 2015-01-20 | 2016-08-17 | 阿里巴巴集团控股有限公司 | Safe gate inhibition method and system thereof |
CN106021606A (en) * | 2016-06-21 | 2016-10-12 | 广东欧珀移动通信有限公司 | Fingerprint template updating method and terminal equipment |
-
2017
- 2017-07-07 CN CN201710552991.8A patent/CN107527008A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101231692A (en) * | 2007-12-24 | 2008-07-30 | 浙江金指码科技有限公司 | Learning type intellectual fingerprint identification comparison method adjusted by fingerprint eigenvalue |
CN105869235A (en) * | 2015-01-20 | 2016-08-17 | 阿里巴巴集团控股有限公司 | Safe gate inhibition method and system thereof |
CN106021606A (en) * | 2016-06-21 | 2016-10-12 | 广东欧珀移动通信有限公司 | Fingerprint template updating method and terminal equipment |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109492560A (en) * | 2018-10-26 | 2019-03-19 | 深圳力维智联技术有限公司 | Facial image Feature fusion, device and storage medium based on time scale |
CN109933265A (en) * | 2018-11-08 | 2019-06-25 | 闽江学院 | A kind of display methods and device of automatic identification |
CN109933265B (en) * | 2018-11-08 | 2020-08-14 | 闽江学院 | Display method and device for automatic identification |
CN111126147A (en) * | 2019-11-22 | 2020-05-08 | 北京迈格威科技有限公司 | Image processing method, device and electronic system |
CN111126147B (en) * | 2019-11-22 | 2023-10-24 | 天津极豪科技有限公司 | Image processing method, device and electronic system |
CN111062342A (en) * | 2019-12-20 | 2020-04-24 | 中国银行股份有限公司 | Debugging method and device of face recognition system |
CN111062342B (en) * | 2019-12-20 | 2023-04-07 | 中国银行股份有限公司 | Debugging method and device of face recognition system |
CN111639915A (en) * | 2020-05-29 | 2020-09-08 | 广东电网有限责任公司 | Staff data management system |
CN112148908A (en) * | 2020-10-23 | 2020-12-29 | 北京百度网讯科技有限公司 | Image database updating method and device, electronic equipment and medium |
CN112818784A (en) * | 2021-01-22 | 2021-05-18 | 浙江大华技术股份有限公司 | Control method and device of access control equipment and storage medium |
CN112818784B (en) * | 2021-01-22 | 2024-02-06 | 浙江大华技术股份有限公司 | Control method and device of access control equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107527008A (en) | Face identification system and its control method and control device | |
CN109740787B (en) | Training building air conditioner load prediction model and prediction method and device using same | |
CN110836514B (en) | Control method and device of air conditioning unit | |
Szczurek et al. | Occupancy determination based on time series of CO2 concentration, temperature and relative humidity | |
CN113110082B (en) | Method and device for controlling household appliance and household appliance | |
US8078318B2 (en) | Appliance control apparatus and method | |
JP4791385B2 (en) | Clothes dry state prediction program and washing machine | |
CN105095882A (en) | Method and apparatus for gesture identification | |
CN109284988A (en) | Data analysis system and method | |
CN108181837B (en) | Control method and control device | |
CN105786711A (en) | Data analysis method and device | |
CN105975079A (en) | Air conditioner information processing method and device | |
CN111649465B (en) | Automatic control method and system for air conditioning equipment | |
CN111623498A (en) | Automatic air conditioner temperature determination method and system | |
CN107438019A (en) | Smart home learning control method, device and system | |
CN109883016A (en) | A kind of air pleasant degree adjusting method and equipment | |
CN111896831B (en) | Non-invasive comprehensive energy load monitoring method and system | |
Li et al. | Non-intrusive load monitoring based on convolutional neural network mixed residual unit | |
CN113888005A (en) | Building design dynamic process management system and method | |
CN110096793A (en) | A kind of house energy consumption prediction technique based on residential subscribers' activity pattern | |
CN117010638B (en) | Intelligent management method and system for hotel equipment | |
CN115601081B (en) | Online digital exhibition service management system based on big data analysis | |
CN116951780A (en) | Intelligent central water heater management method and system | |
CN115456162B (en) | Training method and device for user behavior habit learning model based on virtual interaction | |
CN115829662A (en) | Hotel guest room price dynamic updating method |
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: 20171229 |
|
RJ01 | Rejection of invention patent application after publication |