CN104537389B - Face identification method and device - Google Patents
Face identification method and device Download PDFInfo
- Publication number
- CN104537389B CN104537389B CN201410837771.6A CN201410837771A CN104537389B CN 104537389 B CN104537389 B CN 104537389B CN 201410837771 A CN201410837771 A CN 201410837771A CN 104537389 B CN104537389 B CN 104537389B
- Authority
- CN
- China
- Prior art keywords
- face
- sample
- picture frame
- user name
- threshold value
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/758—Involving statistics of pixels or of feature values, e.g. histogram matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/164—Detection; Localisation; Normalisation using holistic features
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Collating Specific Patterns (AREA)
Abstract
The embodiment of the present invention provides a kind of face identification method and device, the picture frame sequence of target face is compared with the pictures in Face Sample Storehouse by using image statisticses feature, if judgement knows that the sample face matched with target face is unique, and the picture frame that the match is successful accounts for the ratio of picture frame sequence more than default first threshold value and is less than default second threshold value, wherein, second threshold value is more than the first threshold value, then the picture frame that it fails to match is added in Face Sample Storehouse in pictures corresponding with sample face.Automatic study and the makeover process of Face Sample Storehouse are realized, avoids user from performing cumbersome amendment operation, with the accumulation of usage time, more and more higher is improved the practicality of face identification system by the accuracy rate of recognition of face.
Description
Technical field
The present invention relates to communication technical field, more particularly to a kind of face identification method and device.
Background technology
Recognition of face refers to the technology or system that identity validation or identity finder are carried out by face, and face recognition technology is
Face feature based on people, to the facial image recognition or video flowing of input, first determine whether that it whether there is face, if there is people
Face, then the positional information of the position of each face, size and each major facial organ is further provided, and according to these letters
Breath, further extracts the identity characteristic contained in each face, and itself and known face are contrasted, every so as to identify
The identity of individual face.
Face recognition process typically point three steps：
(1) the image surface archives of face are initially set up.I.e. with the image surface file of the face of camera acquisition unit personnel or take
Their photo forming face is stored up as file, and by these image surface file generated face lines (Faceprint) coding.
(2) current human body image surface is obtained.The image surface of the current discrepancy personnel of cameras capture is used, or takes photo defeated
Enter, and current image surface file generated face line is encoded.
(3) with current face line coding and the comparison of archives stock.The face line coding and file store of image surface that will be current
Face line coding in depositing carries out retrieval comparison.
However, in common face recognition algorithms, hair style, clothing and the environmental background of detected object, illumination etc. are all
The degree of accuracy of detection can be influenceed, and after once having gathered sample, the accuracy rate of identification just secures substantially, it is desirable to improves again
Performance is difficult that therefore, adaptability and practicality are poor, has certain limitation.
The content of the invention
The present invention provides a kind of face identification method and device, can not be learned automatically with solving Face Sample Storehouse in the prior art
Practise, dynamic updates, and causes the accuracy rate of recognition of face low, adaptability and the poor technological deficiency of practicality.
According to the first aspect of the invention, there is provided a kind of face identification method, including：
Obtain the picture frame sequence of target face to be identified；
The picture frame sequence and the pictures in the Face Sample Storehouse that pre-establishes are entered using image statisticses feature
Row compares；
If judgement knows that the sample face matched with the target face is unique, and the picture frame that the match is successful accounts for the figure
The ratio of piece frame sequence is more than default first threshold value and is less than default second threshold value, wherein, second threshold value
More than first threshold value, then the picture frame that it fails to match is added in the Face Sample Storehouse and the sample face pair
In the pictures answered.
According to the second aspect of the invention, there is provided a kind of face identification device, including：Equipment body, in addition to：
Acquisition module, for obtaining the picture frame sequence of target face to be identified；
Judge module, for using image statisticses feature by the picture frame sequence and the Face Sample Storehouse pre-established
In pictures be compared；
Processing module, if for judging to know that the sample face matched with the target face is unique, and the match is successful
The ratio that picture frame accounts for the picture frame sequence is more than default first threshold value and is less than default second threshold value, wherein,
Second threshold value is more than first threshold value, then by the picture frame that it fails to match be added in the Face Sample Storehouse with
In pictures corresponding to the sample face.
Face identification method and device provided in an embodiment of the present invention, by using image statisticses feature by target face
Picture frame sequence be compared with the pictures in Face Sample Storehouse, if judging to know the sample face that matches with target face
Uniquely, and the picture frame that the match is successful accounts for the ratio of picture frame sequence and is more than default first threshold value and less than default second
Threshold value, wherein, the second threshold value is more than the first threshold value, then by the picture frame that it fails to match be added in Face Sample Storehouse with
In pictures corresponding to sample face.Automatic study and the makeover process of Face Sample Storehouse are realized, avoids user from performing cumbersome
Amendment operation, with the accumulation of usage time, more and more higher is improved face identification system by the accuracy rate of recognition of face
Practicality.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are this hairs
Some bright embodiments, for those of ordinary skill in the art, without having to pay creative labor, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is the schematic flow sheet of the present inventor's face recognition method embodiment one；
Fig. 2 is the schematic flow sheet of the present inventor's face recognition method embodiment two；
Fig. 3 is the structural representation of face identification device embodiment one of the present invention.
Embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
Part of the embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
Fig. 1 is the schematic flow sheet of the present inventor's face recognition method embodiment one, as shown in figure 1, this method includes：
Step 100, the picture frame sequence of target face to be identified is obtained；
Terminal device obtains the picture frame sequence of target face to be identified, and specifically acquisition modes can be set according to terminal
Standby software and hardware configuration situation is flexibly set, specific as follows：
Application scenarios one, if terminal device can carry out communication interaction, such as mobile phone, computer etc. with internet, it is arranged on
Associated server in internet can directly provide the picture frame sequence of target face to be identified to terminal device.Or
Associated server can provide multimedia data stream to terminal device, and terminal device passes through Face datection and the mathematics of face tracking
Model inspection multimedia data stream, from IMAQ is proceeded by the time of determining and target person face occur, to default time point
Terminate IMAQ, so as to obtain the picture frame sequence of the target face within the default period.
Application scenarios two, if being provided with the such as camera of the equipment with image collecting function on terminal device and/or sweeping
Instrument is retouched, terminal device can obtain multimedia data stream by the equipment with image collecting function, and terminal device specifically includes：
Camera, video camera, the mobile phone with camera, the computer with camera and the gate control system with camera.It is specific next
Say, terminal device can be by scanner scanning photo or pending image information acquisition multimedia data stream, when passing through
At the time of the mathematical modeling of Face datection and face tracking determines target person face occur, proceed by image from current time and adopt
Collection, terminates IMAQ to default time point, so as to gather the picture frame sequence of target face within the default period.Or
Person, the time that terminal device can be pre-set automatically turn on camera, if camera is more, can selectively opened,
Or the setting that the angle of camera can be automatically adjusted, obtain terminal device so as to shoot surrounding enviroment by camera
Multimedia data stream is taken, at the time of determining target person face occur by the mathematical modeling of Face datection and face tracking, from working as
The preceding moment proceeds by IMAQ, terminates IMAQ to default time point, so as to gather mesh within the default period
Mark the picture frame sequence of face.
Set, can also enter as needed it should be noted that the above-mentioned two kinds of application scenarios specifically introduced can combine
Row selection, the present embodiment are without limitation.
Step 101, using image statisticses feature by the picture frame sequence and the Face Sample Storehouse pre-established
Pictures are compared；
After terminal device obtains the picture frame sequence of target face to be identified, using image statisticses feature by picture
Frame sequence is compared with the pictures in the Face Sample Storehouse pre-established, and specifically, terminal device is in picture frame sequence
The profile of target face, eyebrow, eyes, nose, the feature such as lip extracted, using image statisticses feature will with it is advance
Pictures in the Face Sample Storehouse of foundation are compared, it is necessary to illustrate, image statisticses feature includes：Haar features,
FisherFace features and LBPH features, those skilled in the art can be selected according to using needs, Haar features,
FisherFace features and LBPH features belong to prior art, and this, which is in, repeats.
Wherein, the Face Sample Storehouse pre-established in terminal device can be that terminal device has included goods producer admittedly
Those Face Sample Storehouses changed, are more flexibly that the Face Sample Storehouse in the present embodiment can not only include goods producer
Those Face Sample Storehouses having been cured, it can also be that user needs to store Face Sample Storehouse according to the application of oneself.Specifically,
If being provided with the equipment with image collecting function such as camera and/or scanner on terminal device, terminal device can lead to
Cross the equipment with image collecting function and obtain multimedia data stream, terminal device specifically includes：Camera, video camera, with taking the photograph
Mobile phone as head, the computer with camera and the gate control system with camera.Terminal device passes through with IMAQ
The equipment of function samples to face, and passes through the setting of the website of the APP application software on terminal device or correlation and institute
User name corresponding to the pictures sample of collection, establishes Face Sample Storehouse.
Step 102, if judging to know that the sample face matched with the target face is unique, and the picture frame that the match is successful
The ratio for accounting for the picture frame sequence is more than and default first threshold value and is less than default second threshold value, wherein, described the
Two threshold values are more than first threshold value, then are added to the picture frame that it fails to match in the Face Sample Storehouse and the sample
In pictures corresponding to this face.
Terminal device uses image statisticses feature by the picture in picture frame sequence and the Face Sample Storehouse pre-established
After collection is compared, terminal device judges whether the sample face that is matched in Face Sample Storehouse with target face is unique, also
It is to say, whether there was only one in Face Sample Storehouse, sample face can the match is successful with target face, and other sample faces are all
It fails to match with target face.If terminal device judges to know the sample face matched with target face in Face Sample Storehouse only
One, then continue to judge the ratio for the picture frame sequence that the picture frame that the match is successful accounts for target face whether between default first
Between limit value and the second threshold value, wherein, the second threshold value is more than the first threshold value, i.e., the ratio is more than the first threshold value and small
In the second threshold value, if judging to know that the ratio is more than the first threshold value and is less than the second threshold value, illustrate target face and
The matching degree of sample face is medium, but sample corresponding with the sample face is insufficient in Face Sample Storehouse, so that will basis
Sample corresponding with sample face in the picture frame sequence renewal Face Sample Storehouse of target face, i.e., by the picture frame of target face
The picture frame that it fails to match in sequence is added in Face Sample Storehouse in pictures corresponding with the sample face, and former
Sample re -training together, to improve the accuracy rate of identification next time.
It should be noted that the first threshold value and the second threshold value in the present embodiment are those skilled in the art according to reality
Border is configured using needs, and can dynamically be adjusted according to the change of practical application request.
The face identification method that the present embodiment provides, by using image statisticses feature by the picture frame sequence of target face
Row are compared with the pictures in Face Sample Storehouse, if judge to know that the sample face matched with target face is unique, and
The ratio of picture frame sequence is accounted for more than default first threshold value with successful picture frame and is less than default second threshold value, its
In, the second threshold value is more than the first threshold value, then is added to the picture frame that it fails to match in Face Sample Storehouse and sample face
In corresponding pictures.Automatic study and the makeover process of Face Sample Storehouse are realized, avoids user from performing cumbersome amendment behaviour
Make, with the accumulation of usage time, more and more higher is improved the practicality of face identification system by the accuracy rate of recognition of face.
Based on above-described embodiment, further, methods described also includes：
If terminal device judges to know that the sample face matched with target face in Face Sample Storehouse is unique, and judges to know
The ratio that the picture frame that the match is successful accounts for the picture frame sequence of target face is less than or equal to default first threshold value, then illustrates mesh
Mark face is low with the matching degree of sample face, and sample corresponding with the sample face is also insufficient in Face Sample Storehouse, so as to want
Sample corresponding with the sample face in Face Sample Storehouse is updated according to the picture frame sequence of target face.
Specifically renewal process is that terminal device prompting user inputs user name first, and specific prompting mode can lead to
The mode of voice message or dialog box prompting is crossed, the present embodiment is without limitation.When terminal device receives user's input
User name after, judge user input user name whether with from Face Sample Storehouse and match come sample face corresponding to
User name it is consistent, if judging to know consistent, illustrate matching correctly, but corresponding with the sample face in Face Sample Storehouse
Sample is insufficient, so as to which the picture frame that it fails to match be added in Face Sample Storehouse in pictures corresponding with sample face,
With former sample together re -training, with improve next time identification accuracy rate.
If terminal device judges to know the user name of user's input and from Face Sample Storehouse with matching the sample face come
Corresponding user name is inconsistent, then illustrates matching error, sample corresponding to the target face is not stored in Face Sample Storehouse,
It is newly-built, corresponding with the user name of user's input in Face Sample Storehouse so as to which the picture frame sequence of the target face be added to
In pictures, so as to improve Face Sample Storehouse, further to improve the accuracy rate of identification next time.
Based on above-described embodiment, further, methods described also includes：
If terminal device judges to know that the sample face matched with target face in Face Sample Storehouse is unique, and judges to know
The ratio that the picture frame that the match is successful accounts for the picture frame sequence of target face is more than or equal to default second threshold value, then illustrates mesh
The matching degree height of face and sample face is marked, sample corresponding with the sample face is abundant in Face Sample Storehouse, can accurately know
Do not go out the target face, matching terminates.
Based on above-described embodiment, further, methods described also includes：
Terminal device uses image statisticses feature by the picture in picture frame sequence and the Face Sample Storehouse pre-established
After collection is compared, if terminal device judges to know that the sample face matched with target face in Face Sample Storehouse is not unique,
That is, in Face Sample Storehouse at least two sample face can the match is successful with target face, then illustrate matching error,
Sample in Face Sample Storehouse is wrong, so as to will according to the picture frame sequence of target face update Face Sample Storehouse in the sample
Sample corresponding to this face.
Specifically renewal process is that terminal device prompting user inputs user name first, and specific prompting mode can lead to
The mode of voice message or dialog box prompting is crossed, the present embodiment is without limitation.When terminal device receives user's input
User name after, judge user input user name whether belong in Face Sample Storehouse match come at least two sample people
User name corresponding to face, if the user name for judging to know user's input belongs in Face Sample Storehouse at least two samples for matching and
User name corresponding to this face, then illustrate sample mistake corresponding with the sample face in Face Sample Storehouse, so as to by target person
The picture frame sequence of face is added in Face Sample Storehouse, replaces and schemes corresponding to sample face corresponding with the user name of user's input
Piece collection, to improve the accuracy rate of identification next time.
If terminal device judges to know that the user name of user's input is not belonging to match at least two come in Face Sample Storehouse
User name corresponding to individual sample face, then illustrate not storing sample corresponding to the target face in Face Sample Storehouse, so that will
The picture frame sequence of the target face is added to pictures newly-built, corresponding with the user name of user's input in Face Sample Storehouse
In, so as to improve Face Sample Storehouse, further to improve the accuracy rate of identification next time.
Fig. 2 is the schematic flow sheet of the present inventor's face recognition method embodiment two, and the present embodiment is with the family with camera
The process of face identification method is described exemplified by the gate control system of front yard in detail, as shown in Fig. 2 this method includes：
The face identification method of the present embodiment introduction is applied to the Household access control system with camera, it is assumed that one family
In have tri- people of A, B, C, when use for the first time, Household access control system is sampled by camera to everyone face,
And user name corresponding to the sample face of sampling is set by APP application software or webpage, so as to establish face sample
This storehouse.
Step 200, when A face (target face) is appeared in camera, Household access control system will since current time,
With the speed acquisition 3 seconds of 30 two field picture per second, the picture frame sequence of 90 frames is obtained；
Step 201, using image statisticses feature by the picture frame sequence and the Face Sample Storehouse pre-established
Pictures are compared；
Step 202, whether the sample face that judgement matches with the target face is unique, if so, step 203 is then performed,
Otherwise, step 207 is performed；
Step 203, judge the picture frame that the match is successful account for the picture frame sequence ratio it is whether (default between 50%
First threshold value) between -90% (default second threshold value)；
Step 204, if judging to know, the picture frame that the match is successful accounts for the ratio of the picture frame sequence and is more than 50% and small
In 90%, then the picture frame that it fails to match is added in Face Sample Storehouse in pictures corresponding with sample face.
Step 205, if judgement knows that the picture frame that the match is successful accounts for the ratio of the picture frame sequence more than or equal to 90%,
Then matching terminates.
Step 206, if judgement knows that the picture frame that the match is successful accounts for the ratio of the picture frame sequence less than or equal to 50%,
Then prompting user inputs user name, judges user name corresponding with the sample face and the user in the Face Sample Storehouse
Whether the user name of input is consistent, if so, being then added to the picture frame that it fails to match in the Face Sample Storehouse and the sample
In pictures corresponding to this face；Otherwise, by the picture frame sequence be added in the Face Sample Storehouse it is newly-built, with it is described
In pictures corresponding to the user name of user's input.
Step 207, prompting user inputs user name, judges whether the user name of user's input belongs to from the face
The user name corresponding at least two sample faces matched in Sample Storehouse, if so, being then added to the picture frame sequence
In Face Sample Storehouse, pictures corresponding to sample face corresponding with the user name of user input are replaced, otherwise, by described in
Picture frame sequence is added in the Face Sample Storehouse in pictures newly-built, corresponding with the user name of user input.
Compared with prior art, the advantages of face identification method that the present embodiment provides, is：In face recognition process,
It is the automatic process for performing a study in system nature.The time used with this set system is increasingly longer, recognition of face
Accuracy rate by more and more higher, and do not need user to perform cumbersome amendment operation, system will learn automatically, autonomous amendment.
One of ordinary skill in the art will appreciate that：Realizing all or part of step of above method embodiment can pass through
Programmed instruction related hardware is completed, and foregoing program can be stored in a computer read/write memory medium, the program
Upon execution, the step of execution includes above method embodiment；And foregoing storage medium includes：ROM, RAM, magnetic disc or light
Disk etc. is various can be with the medium of store program codes.
Fig. 3 is the structural representation of face identification device embodiment one of the present invention, as shown in figure 3, the device includes：Equipment
Main body 11, in addition to：Acquisition module 12, judge module 13 and processing module 14, wherein, acquisition module 12, wait to know for obtaining
The picture frame sequence of other target face；Judge module 13, for using image statisticses feature by the picture frame sequence and
Pictures in the Face Sample Storehouse pre-established are compared；Processing module 14, if for judging to know and the target person
The sample face of face matching is unique, and the picture frame that the match is successful accounts for the ratio of the picture frame sequence more than default first
Limit value and it is less than default second threshold value, wherein, second threshold value is more than first threshold value, then will it fails to match
Picture frame be added in the Face Sample Storehouse in pictures corresponding with the sample face.
Specifically, terminal device obtains the picture frame sequence of target face to be identified, and specifically acquisition modes can root
Flexibly set according to the software and hardware configuration situation of terminal device, it is specific as follows：
Application scenarios one, if terminal device can carry out communication interaction, such as mobile phone, computer etc. with internet, it is arranged on
Associated server in internet can directly provide the picture frame sequence of target face to be identified to terminal device.Or
Associated server can provide multimedia data stream to terminal device, and terminal device passes through Face datection and the mathematics of face tracking
Model inspection multimedia data stream, from IMAQ is proceeded by the time of determining and target person face occur, to default time point
Terminate IMAQ, so as to obtain the picture frame sequence of the target face within the default period.
Application scenarios two, if being provided with the such as camera of the equipment with image collecting function on terminal device and/or sweeping
Instrument is retouched, terminal device can obtain multimedia data stream by the equipment with image collecting function, and terminal device specifically includes：
Camera, video camera, the mobile phone with camera, the computer with camera and the gate control system with camera.It is specific next
Say, terminal device can be by scanner scanning photo or pending image information acquisition multimedia data stream, when passing through
At the time of the mathematical modeling of Face datection and face tracking determines target person face occur, proceed by image from current time and adopt
Collection, terminates IMAQ to default time point, so as to gather the picture frame sequence of target face within the default period.Or
Person, the time that terminal device can be pre-set automatically turn on camera, if camera is more, can selectively opened,
Or the setting that the angle of camera can be automatically adjusted, obtain terminal device so as to shoot surrounding enviroment by camera
Multimedia data stream is taken, at the time of determining target person face occur by the mathematical modeling of Face datection and face tracking, from working as
The preceding moment proceeds by IMAQ, terminates IMAQ to default time point, so as to gather mesh within the default period
Mark the picture frame sequence of face.
Set, can also enter as needed it should be noted that the above-mentioned two kinds of application scenarios specifically introduced can combine
Row selection, the present embodiment are without limitation.
Further,
Processing module 14, it is additionally operable to sample face by the equipment with image collecting function, and sets
User name corresponding with the pictures sample gathered, establishes Face Sample Storehouse.
Further,
Judge module 13, if be additionally operable to judge know that sample face match with the target face is unique, and matching into
The ratio that the picture frame of work(accounts for the picture frame sequence is more than or equal to second threshold value, then matching terminates.
Further,
Judge module 13, if be additionally operable to judge know that sample face match with the target face is unique, and matching into
The ratio that the picture frame of work(accounts for the picture frame sequence is less than or equal to first threshold value, then prompts user to input user name；
Processing module 14, be additionally operable to judge in the Face Sample Storehouse user name corresponding with the sample face with it is described
Whether the user name of user's input is consistent, if so, being then added to the picture frame that it fails to match in the Face Sample Storehouse and institute
State in pictures corresponding to sample face；Otherwise, by the picture frame sequence be added in the Face Sample Storehouse it is newly-built, with
In pictures corresponding to the user name of user's input.
Further,
Judge module 13, if being additionally operable to judge to know that the sample face matched with the target face is not unique, prompt
User inputs user name；
Processing module 14, is additionally operable to judge whether the user name of user's input belongs to from the Face Sample Storehouse
The user name corresponding at least two sample faces allotted, if so, the picture frame sequence then is added into Face Sample Storehouse
In, pictures corresponding to sample face corresponding with the user name of user input are replaced, otherwise, by the picture frame sequence
It is added in the Face Sample Storehouse in pictures newly-built, corresponding with the user name of user input.
The function and handling process of each module, may refer to shown in above-mentioned Fig. 1 in the terminal device that the present embodiment provides
Embodiment of the method, its implementing principle and technical effect is similar, and here is omitted.
Finally it should be noted that：Various embodiments above is merely illustrative of the technical solution of the present invention, rather than its limitations；To the greatest extent
The present invention is described in detail with reference to foregoing embodiments for pipe, it will be understood by those within the art that：Its according to
The technical scheme described in foregoing embodiments can so be modified, either which part or all technical characteristic are entered
Row equivalent substitution；And these modifications or replacement, the essence of appropriate technical solution is departed from various embodiments of the present invention technology
The scope of scheme.
Claims (16)
- A kind of 1. face identification method, it is characterised in that including：Obtain the picture frame sequence of target face to be identified；The picture frame sequence and the pictures in the Face Sample Storehouse that pre-establishes are compared using image statisticses feature It is right；If judgement knows that the sample face matched with the target face is unique, and the picture frame that the match is successful accounts for the picture frame The ratio of sequence is more than default first threshold value and is less than default second threshold value, wherein, second threshold value is more than First threshold value, then the picture frame that it fails to match is added in the Face Sample Storehouse corresponding with the sample face In pictures；If judgement knows that the sample face matched with the target face is not unique, user is prompted to input user name；Judge whether the user name of user's input belongs to from the Face Sample Storehouse at least two sample people matched User name corresponding to face, if so, then the picture frame sequence is added in Face Sample Storehouse, replaces and inputted with the user User name corresponding to pictures corresponding to sample face, otherwise, the picture frame sequence is added to the Face Sample Storehouse In it is newly-built, with the corresponding pictures of user name of user input.
- 2. face identification method according to claim 1, it is characterised in that methods described also includes：If judgement knows that the sample face matched with the target face is unique, and the picture frame that the match is successful accounts for the picture frame The ratio of sequence is more than or equal to second threshold value, then matching terminates.
- 3. face identification method according to claim 2, it is characterised in that methods described also includes：If judgement knows that the sample face matched with the target face is unique, and the picture frame that the match is successful accounts for the picture frame The ratio of sequence is less than or equal to first threshold value, then prompts user to input user name；Judge user name corresponding with the sample face and the user input in the Face Sample Storehouse user name whether Unanimously, if so, the picture frame that it fails to match is then added in the Face Sample Storehouse into picture corresponding with the sample face Concentrate；Otherwise, the picture frame sequence is added to user name newly-built, with user input in the Face Sample Storehouse In corresponding pictures.
- 4. according to any described face identification methods of claim 1-3, it is characterised in thatDescribed image statistics feature includes：Haar features, FisherFace features and LBPH features.
- 5. according to any described face identification methods of claim 1-3, it is characterised in that described to obtain target person to be identified The picture frame sequence of face, is specifically included：Recognizable target face is determined whether by the multimedia data stream for detecting input；From at the time of determining target person face occur, picture frame sequence of the target face in preset time period is gathered.
- 6. face identification method according to claim 5, it is characterised in that the multimedia data stream of the input, specifically Including：The multimedia data stream is obtained by the equipment with image collecting function.
- 7. face identification method according to claim 6, it is characterised in that the equipment bag with image collecting function Include：Camera and/or scanner.
- 8. face identification method according to claim 7, it is characterised in that methods described also includes：Face is sampled by the equipment with image collecting function, and the pictures sample pair with being gathered is set The user name answered, establishes Face Sample Storehouse.
- 9. a kind of face identification device, including：Equipment body, it is characterised in that also include：Acquisition module, for obtaining the picture frame sequence of target face to be identified；Judge module, for using image statisticses feature by the picture frame sequence and the Face Sample Storehouse pre-established Pictures are compared；Processing module, if for judging to know that the sample face matched with the target face is unique, and the picture that the match is successful The ratio that frame accounts for the picture frame sequence is more than default first threshold value and is less than default second threshold value, wherein, it is described Second threshold value is more than first threshold value, then by the picture frame that it fails to match be added in the Face Sample Storehouse with it is described In pictures corresponding to sample face；The judge module, is additionally operable to：When judging to know that the sample face that matches with the target face is not unique, then prompt User inputs user name；The processing module, it is additionally operable to judge whether the user name of user's input belongs to from the Face Sample Storehouse and matches The user name corresponding at least two sample faces gone out, if so, then the picture frame sequence is added in Face Sample Storehouse, Pictures corresponding to replacing sample face corresponding with the user name of user input, otherwise, the picture frame sequence is added It is added in the Face Sample Storehouse in pictures newly-built, corresponding with the user name of user input.
- 10. face identification device according to claim 9, it is characterised in thatThe judge module, if it is additionally operable to judge to know that the sample face matched with the target face is unique, and the match is successful Picture frame account for the ratio of the picture frame sequence and be more than or equal to second threshold value, then matching terminates.
- 11. face identification device according to claim 10, it is characterised in thatThe judge module, if it is additionally operable to judge to know that the sample face matched with the target face is unique, and the match is successful Picture frame account for the ratio of the picture frame sequence and be less than or equal to first threshold value, then prompt user to input user name；The processing module, it is additionally operable to judge user name corresponding with the sample face and the use in the Face Sample Storehouse Family input user name it is whether consistent, if so, then by the picture frame that it fails to match be added in the Face Sample Storehouse with it is described In pictures corresponding to sample face；Otherwise, the picture frame sequence is added to newly-built in the Face Sample Storehouse and institute State in pictures corresponding to the user name of user's input.
- 12. according to any described face identification devices of claim 9-11, it is characterised in that the acquisition module is specifically used In：Recognizable target face is determined whether by the multimedia data stream for detecting input；From at the time of determining target person face occur, picture frame sequence of the target face in preset time period is gathered.
- 13. face identification device according to claim 12, it is characterised in that be provided with figure on the equipment body As the equipment of acquisition function,The acquisition module obtains the multimedia data stream by the equipment with image collecting function.
- 14. face identification device according to claim 13, it is characterised in that the equipment with image collecting function Including：Camera and/or scanner.
- 15. face identification device according to claim 13, it is characterised in thatThe processing module, be additionally operable to sample face by the equipment with image collecting function, and set with User name corresponding to the pictures sample gathered, establishes Face Sample Storehouse.
- 16. according to any described face identification devices of claim 13-15, it is characterised in that described device includes：Camera, video camera, mobile phone, computer and gate control system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410837771.6A CN104537389B (en) | 2014-12-29 | 2014-12-29 | Face identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410837771.6A CN104537389B (en) | 2014-12-29 | 2014-12-29 | Face identification method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104537389A CN104537389A (en) | 2015-04-22 |
| CN104537389B true CN104537389B (en) | 2018-03-27 |
Family
ID=52852909
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410837771.6A Active CN104537389B (en) | 2014-12-29 | 2014-12-29 | Face identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104537389B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105844267A (en) * | 2016-06-14 | 2016-08-10 | 皖西学院 | Face recognition algorithm |
| CN107526999B (en) * | 2016-06-22 | 2018-10-19 | 腾讯科技（深圳）有限公司 | A kind of standard faces picture update method, data processing equipment and system |
| CN106067013B (en) * | 2016-06-30 | 2022-04-12 | 美的集团股份有限公司 | Face recognition method and device for embedded system |
| CN106295617A (en) * | 2016-08-25 | 2017-01-04 | 广东云海云计算科技有限公司 | Recognition of face server cluster based on degree of depth study |
| CN108647651A (en) * | 2018-05-14 | 2018-10-12 | 深圳市科发智能技术有限公司 | A kind of face identification method, system and device improving the rate that is identified by |
| CN108805046B (en) * | 2018-05-25 | 2022-11-04 | 京东方科技集团股份有限公司 | Method, apparatus, device and storage medium for face matching |
| CN108846676B (en) * | 2018-08-02 | 2023-07-11 | 平安科技（深圳）有限公司 | Biological feature auxiliary payment method, device, computer equipment and storage medium |
| CN109741380B (en) * | 2018-12-27 | 2021-09-14 | 广州华迅网络科技有限公司 | Textile picture fast matching method and device |
| CN109916017A (en) * | 2019-03-08 | 2019-06-21 | 广东美的制冷设备有限公司 | Control method, air conditioner, intelligent mobile terminal and the storage medium of air conditioner |
| CN109961046B (en) * | 2019-03-26 | 2022-03-15 | 武汉大学 | Video stream face identification method for building dynamic sample set based on keyframe backtracking |
| CN110217270B (en) * | 2019-05-29 | 2021-08-27 | 成都希格玛光电科技有限公司 | Method and system for detecting rail invasion foreign matters at fixed distance |
| CN110188722A (en) * | 2019-06-05 | 2019-08-30 | 福建深视智能科技有限公司 | A kind of method and terminal of local recognition of face image duplicate removal |
| CN110516597A (en) * | 2019-08-27 | 2019-11-29 | 睿云联(厦门)网络通讯技术有限公司 | Off-line learning method, system, equipment and the storage medium of lifting feature resolution |
| CN113537666B (en) * | 2020-04-16 | 2024-05-03 | 马上消费金融股份有限公司 | Evaluation model training method, evaluation and business auditing method, device and equipment |
| CN111859000A (en) * | 2020-06-24 | 2020-10-30 | 天津大学 | Method for constructing and updating human face feature database under deep learning model |
| CN111914637B (en) * | 2020-06-28 | 2021-05-04 | 普瑞达建设有限公司 | Intelligent face recognition integrated management method and system |
| CN113808354A (en) * | 2021-11-16 | 2021-12-17 | 深圳市思拓通信系统有限公司 | Method, device and medium for early warning of construction site dangerous area |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7024033B2 (en) * | 2001-12-08 | 2006-04-04 | Microsoft Corp. | Method for boosting the performance of machine-learning classifiers |
| CN101216884B (en) * | 2007-12-29 | 2012-04-18 | 北京中星微电子有限公司 | A method and system for face authentication |
| CN102004905B (en) * | 2010-11-18 | 2012-11-21 | 无锡中星微电子有限公司 | Human face authentication method and device |
-
2014
- 2014-12-29 CN CN201410837771.6A patent/CN104537389B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104537389A (en) | 2015-04-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104537389B (en) | Face identification method and device | |
| CN109034013B (en) | Face image recognition method, device and storage medium | |
| CN110602527B (en) | Video processing method, device and storage medium | |
| CN111862296B (en) | Three-dimensional reconstruction method, three-dimensional reconstruction device, three-dimensional reconstruction system, model training method and storage medium | |
| CN108229376B (en) | Method and device for detecting blinking | |
| CN112364827B (en) | Face recognition method, device, computer equipment and storage medium | |
| CN109829381A (en) | A kind of dog only identifies management method, device, system and storage medium | |
| CN108269333A (en) | Face identification method, application server and computer readable storage medium | |
| CN110472487B (en) | Living user detection method, living user detection device, computer device, and storage medium | |
| CN105518711A (en) | In-vivo detection method, in-vivo detection system, and computer program product | |
| WO2021082692A1 (en) | Pedestrian picture labeling method and device, storage medium, and intelligent apparatus | |
| CN112702521B (en) | Image shooting method and device, electronic equipment and computer readable storage medium | |
| CN106101540B (en) | Focus point determines method and device | |
| CN106295499A (en) | Age estimation method and device | |
| CN108182746A (en) | Control system, method and apparatus | |
| US20230041573A1 (en) | Image processing method and apparatus, computer device and storage medium | |
| WO2021104128A1 (en) | Feature library update method and apparatus, inference server and storage medium | |
| CN109711312A (en) | A kind of testimony of a witness verifying system and method based on silent In vivo detection recognition of face | |
| CN110363790A (en) | Target tracking method, device and computer readable storage medium | |
| CN111275901B (en) | Control method and device of express delivery cabinet, storage medium and computer equipment | |
| CN109697389B (en) | Identity recognition method and device | |
| CN114861241A (en) | Anti-peeping screen method based on intelligent detection and related equipment thereof | |
| CN110992500A (en) | Attendance checking method and device, storage medium and server | |
| US9286707B1 (en) | Removing transient objects to synthesize an unobstructed image | |
| CN107992816A (en) | Photographing search method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Face recognition method and device Effective date of registration: 20201231 Granted publication date: 20180327 Pledgee: China CITIC Bank Co.,Ltd. Jiaxing Tongxiang sub branch Pledgor: SENGLED OPTOELECTRONICS Co.,Ltd. Registration number: Y2020330001361 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20230208 Granted publication date: 20180327 Pledgee: China CITIC Bank Co.,Ltd. Jiaxing Tongxiang sub branch Pledgor: SENGLED OPTOELECTRONICS Co.,Ltd. Registration number: Y2020330001361 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Face recognition method and device Effective date of registration: 20230302 Granted publication date: 20180327 Pledgee: Tongxiang Yunbei Investment Construction Co.,Ltd. Pledgor: SENGLED OPTOELECTRONICS Co.,Ltd. Registration number: Y2023980033494 |