CN106326815B - A kind of facial image recognition method - Google Patents
A kind of facial image recognition method Download PDFInfo
- Publication number
- CN106326815B CN106326815B CN201510381767.8A CN201510381767A CN106326815B CN 106326815 B CN106326815 B CN 106326815B CN 201510381767 A CN201510381767 A CN 201510381767A CN 106326815 B CN106326815 B CN 106326815B
- Authority
- CN
- China
- Prior art keywords
- facial image
- training sample
- user
- intelligent terminal
- photograph album
- 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
- 230000001815 facial effect Effects 0.000 title claims abstract description 96
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012549 training Methods 0.000 claims abstract description 101
- 238000004891 communication Methods 0.000 claims abstract description 28
- 238000012795 verification Methods 0.000 claims abstract description 9
- 238000012790 confirmation Methods 0.000 claims description 4
- 230000018199 S phase Effects 0.000 claims description 2
- 206010048669 Terminal state Diseases 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Processing Or Creating Images (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of facial image recognition methods, belong to image identification technical field；Method mass: step S1 generates the first training sample using communication directory and user's photograph album training；And/or the buddy list and good friend's photograph album training the second training sample of generation of the user account using social network sites；Step S2, in conjunction with the first training sample and/or the second training sample to form an initial training sample；Step S3 acquires facial image, and carries out recognition and verification according to initial training sample, to improve and form a final identification data；Intelligent terminal is operated according to the identification that identification data execute facial image.The beneficial effect of above-mentioned technical proposal is: enabling intelligent terminal according to the relational network automatic identification and training facial recognition data of user, so that the troublesome operation of picture sample needed for avoiding user from needing to be manually entered a large amount of recognitions of face, promotes the usage experience of user.
Description
Technical field
The present invention relates to image identification technical field more particularly to a kind of facial image recognition methods.
Background technique
Recognition of face is increasingly being applied in intelligent terminal as a kind of emerging identification technology.Such as user
It can be by distinguishing that face recalls the user account etc. that some associated datas include name, contact method, social networks, user
It again may be by distinguishing that the mode of face realizes encryption and decryption operation etc..
But existing face recognition technology, it usually needs great amount of images needed for user is manually entered identification in advance
The mode of sample, input can import for batch, or production acquisition in advance etc..But no matter which kind of is manually entered mode, all
So that the whole operation of user is very cumbersome, the usage experience of user is greatly reduced.
Summary of the invention
According to problems of the prior art, a kind of technical solution of facial image recognition method is now provided, it is intended to solve
User certainly existing in the prior art needs to be manually entered the largely image sample data for recognition of face, and setting figure one by one
As the relationship with name, so as to cause very cumbersome defect is operated；
Above-mentioned technical proposal specifically includes:
A kind of facial image recognition method is suitable for intelligent terminal；Wherein, including a pre-training step, corresponding to obtain
Training sample；
The pre-training step specifically includes:
Step S1, the intelligent terminal remotely obtain the communication directory being stored in the mobile terminal of user and user
Photograph album is matched using the facial image in user's photograph album with the personal head portrait in the communication directory, to establish
State facial image and associated first training sample of the personal corresponding personal name of head portrait in the communication directory；
And/or
The intelligent terminal is remotely searched and obtains the corresponding good friend's column of the preset user account for being associated with social networks
Table and good friend's photograph album, and according to the personal head portrait progress in the facial image and the buddy list in good friend's photograph album
Match, to establish the facial image personal name corresponding with the personal head portrait in the buddy list associated the
Two training samples；
Step S2, in conjunction with first training sample and/or second training sample to form an initial training sample；
Step S3 acquires facial image, and carries out recognition and verification according to the initial training sample, to improve and form one
Final identification data；
After forming the identification data, the facial image that the intelligent terminal acquisition needs to identify, and according to the identification
Data are identified, recognition result is exported.
Preferably, facial image recognition method, wherein in the step S1, form the step of first training sample
Suddenly it specifically includes:
Step S11a remotely obtains the facial image in user's photograph album；
Step S12a remotely obtains the communication directory；
Step S13a is matched using the facial image in the communication directory, to find the communication directory
In the personal head portrait that matches；
The facial image personal name corresponding with the personal head portrait to match is associated by step S14a；
Step S15a, circulation executes the step S11a-S14a, according to all faces in user's photograph album
Image training forms first training sample.
Preferably, facial image recognition method, wherein in the step S1, form the step of second training sample
Suddenly it specifically includes:
Step S11b remotely obtains user's account by being associated with the user account of at least one social network sites
The facial image in number associated good friend's photograph album；
Step S12b remotely obtains user's account by being associated with the user account of at least one social network sites
Number associated buddy list；
Step S13b is matched using the facial image in the buddy list, to find the buddy list
In the personal head portrait that matches；
Step S14b, by the corresponding personal name of the personal head portrait to match in the facial image and the buddy list
It is associated；
Step S15b, circulation executes the step S11b-S14b, according to all faces in good friend's photograph album
Image training forms second training sample.
Preferably, facial image recognition method, wherein in the step S1, the intelligent terminal passes through wireless connection
Mode accesses the mobile terminal, and the corresponding communication directory and user's photograph album are obtained out of described mobile terminal.
Preferably, facial image recognition method, wherein the step S3 is specifically included:
Step S31, one facial image of collection in worksite；
Step S32 carries out recognition and verification according to facial image of the initial training sample to the collection in worksite:
If can not identify, the initial training sample is updated according to the facial image of the collection in worksite；It is subsequently diverted to
Step S33；
If can identify, directly to step S33；
Step S33, circulation executes the step S31-S32, with perfect according to the facial image of collection in worksite described in multiple
The initial training sample；
Step S34 will be integrally formed the identification data by the perfect initial training sample, and be stored in described
In intelligent terminal.
Preferably, facial image recognition method, wherein in the step S32, the intelligent terminal is by issuing prompt
Sound is for the corresponding user's confirmation of the facial image of collection in worksite.
Preferably, facial image recognition method, wherein the intelligent terminal is that the intelligence with robot appearance is whole
End.
Preferably, facial image recognition method, wherein the acquisition facial image in the step S3 includes: that acquisition is closed
It is coupled to the man face image acquiring of people in the initial training sample, the people in user's relational network or stranger.
The beneficial effect of above-mentioned technical proposal is: providing a kind of facial image recognition method, enables intelligent terminal root
According to the relational network automatic identification and training facial recognition data of user, so that user be avoided to need to be manually entered a large amount of people
The troublesome operation of picture sample needed for face identification, promotes the usage experience of user.
Detailed description of the invention
Fig. 1-4 is a kind of flow diagram of facial image recognition method in preferred embodiment of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art without creative labor it is obtained it is all its
His embodiment, shall fall within the protection scope of the present invention.
It should be noted that in the absence of conflict, the feature in embodiment and embodiment in the present invention can phase
Mutually combination.
The present invention will be further explained below with reference to the attached drawings and specific examples, but not as the limitation of the invention.
In preferred embodiment of the invention, it is based on the above-mentioned problems in the prior art, a kind of face figure is now provided
As the technical solution of recognition methods, it is suitable for intelligent terminal.
It include a pre-training step in above-mentioned facial image recognition method in preferred embodiment of the invention.It is so-called pre-
Training step refers to before intelligent terminal is put into actual face recognition operation by reality, establishes the required identification number of identification
According to process.In other words, it in preferred embodiment of the invention, by pre-training step, can build in an intelligent terminal in advance
Found an information bank for identification.
Then in preferred embodiment of the invention, as shown in Figure 1, above-mentioned pre-training step specifically includes:
Step S1, intelligent terminal remotely obtain the communication directory being stored in the mobile terminal of user and user's phase
Volume, is matched using the image pattern in user's photograph album with the personal head portrait in communication directory, to establish image pattern and lead to
Interrogate associated first training sample of the corresponding personal name of personal head portrait in register；And/or
Intelligent terminal remotely search and obtain the corresponding buddy list of the preset user account for being associated with social networks with
And good friend's photograph album, and matched according to the image pattern in good friend's photograph album with the personal head portrait in buddy list, to establish figure
Associated second training sample of decent corresponding with the personal head portrait in buddy list personal name；
In a preferred embodiment of the invention, it can choose and only form above-mentioned first training sample or above-mentioned second
Training sample.
In another preferred embodiment of the invention, above-mentioned first training sample and the second training sample can be formed simultaneously
This.
Then in preferred embodiment of the invention, as shown in Fig. 2, forming the step of the first training sample in above-mentioned steps S1
Suddenly it specifically includes:
Step S11a, the long-range facial image obtained in user's photograph album；
In preferred embodiment of the invention, preset facial image can be the facial image of collection in worksite, can also be with
Facial image of acquired completion before being, such as the major part photograph of friend issued on social networks etc..
Step S12a, remotely obtains communication directory；
In preferred embodiment of the invention, so-called communication directory refers to and preset include personal name and close with it
The list of the personal head portrait of connection correspondingly may include that other are associated with personal information, such as correspondent party in register list
Formula and/or the user account of home address and/or social networks etc..In preferred embodiment of the invention, above-mentioned communication name
Record is stored in the mobile terminal of user.
In preferred embodiment of the invention, intelligent terminal can remotely obtain above-mentioned communication directory.Specifically, intelligence is whole
It establishes and is wirelessly connected between end and the mobile terminal of user, and by being wirelessly connected the communication name for obtaining and saving in mobile terminal
Record, i.e. user authorize the communication directory of intelligent terminal remote access user.In preferred embodiment of the invention, due to intelligence
It can be usually short distance connection between terminal and the mobile terminal of user, therefore number can be transmitted using bluetooth connection mode
According to.In the other embodiment of the present invention, equally data can be transmitted using other radio connections.
Step S13a is matched in communication directory using facial image, to find to match in communication directory
Number of people picture；
Facial image personal name corresponding with the personal head portrait to match is associated by step S14a；
Step S15a, circulation execute step S11a-S14a, to be formed according to the face images training in user's photograph album
First training sample.
In preferred embodiment of the invention, the facial image for including in above-mentioned user's photograph album refers to the photograph of user's photograph album
In piece, photo wherein with face shape can be clearly identified.
In preferred embodiment of the invention, using in each of above-mentioned user's photograph album face image, with communication directory
Including personal head portrait matched one by one, to find the higher personal head portrait of matching degree, and by corresponding facial image with
The corresponding personal name of individual's head portrait is associated.It further,, will after overmatching in preferred embodiment of the invention
Facial image is with corresponding personal information (including user account, the Email on personal name, contact method, social networks
Deng) be associated.
In preferred embodiment of the invention, face images in traverse user photograph album, and ultimately form one first
Training sample.
In other words, in preferred embodiment of the invention, intelligent terminal by access mobile terminal in communication directory and
User's photograph album automatically forms the sample database of a recognition of face.
In preferred embodiment of the invention, as shown in figure 3, forming the method tool of the second training sample in above-mentioned steps 1
Body includes:
Step S11b, by being associated with the user account of at least one social network sites, the long-range user account that obtains is associated
Good friend's photograph album in facial image；
Step S12b, by being associated with the user account of at least one social network sites, the long-range user account that obtains is associated
Buddy list；
Step S13b is matched in buddy list using facial image, to find to match in buddy list
Number of people picture；
The corresponding personal name of the personal head portrait to match in facial image and buddy list is associated by step S14b；
Step S15b, circulation execute step S11b-S14b, to be formed according to the face images training in good friend's photograph album
Second training sample.
In preferred embodiment of the invention, above-mentioned steps S11b-15b is similar with step S11a-15a, and difference is: intelligence
Can terminal pass through buddy list in the user account on the social networks for using for reference user and good friend photograph album forms second
Training sample.In preferred embodiment of the invention, intelligent terminal is connected to social network sites by way of accessing internet, warp
Cross the buddy list and good friend's photograph album that corresponding user account is obtained after authorizing.
In preferred embodiment of the invention, intelligent terminal extracts the facial image in good friend's photograph album, and according to being extracted
Facial image matching buddy list in personal head portrait, by facial image it is corresponding with the personal head portrait to match individual surname
Name is associated.Further, by facial image personal information corresponding with the personal head portrait to match (including personal name, connection
It is mode, the user account on social networks, Email etc.) it is associated.
In preferred embodiment of the invention, facial image in good friend's photograph album is similarly energy in the photo of good friend's photograph album
The image of enough clear identification face shapes.
In preferred embodiment of the invention, intelligent terminal traverses all people's face image in good friend's photograph album and ultimately forms
One second training sample.
Step S2, in conjunction with the first training sample and/or the second training sample to form an initial training sample；
In a preferred embodiment of the invention, if only the first training sample or the second training sample, by phase
The training sample answered is set as initial training sample.
In another preferred embodiment of the invention, if being formed simultaneously the first training sample and the second training sample,
The two training samples are merged, to form initial training sample.
Step S3 acquires facial image, and carries out recognition and verification according to initial training sample, final to improve and form one
Identification data；
In preferred embodiment of the invention, as shown in figure 4, above-mentioned steps S3 is specifically included:
Step S31, one facial image of collection in worksite；
In preferred embodiment of the invention, so-called collection in worksite refers to scene to being associated with the one of initial training sample
A specific people carries out the acquisition of facial image, such as shooting, collecting etc., then using the facial image as input, according to just
It walks training sample and carries out facial image identification.
In a preferred embodiment of the invention, since the formation of above-mentioned training sample depends on the network of personal connections of user
Network (such as corresponding buddy list of user account of communication directory and social networks), therefore the object of above-mentioned collection in worksite can be with
It is defined as the people for including in the relational network of user.
In another preferred embodiment of the invention, the facial image of above-mentioned collection in worksite can equally be associated to one
The self study of initial training sample may be implemented in a stranger, that is, the facial image being not present in initial training sample in this way
Function, with exptended sample library.
Step S32 carries out recognition and verification to collected facial image according to initial training sample:
If can not identify, initial training sample is updated according to collected facial image；It is subsequently diverted to step S33；
If can identify, directly to step S33；
In preferred embodiment of the invention, so-called recognition and verification refers to the recognition accuracy of confirmation initial training sample.
An example is given below:
One facial image of collection in worksite is simultaneously input in intelligent terminal, and intelligent terminal is obtained according to initial training sample matches
To a corresponding result (such as corresponding personal name).Intelligent terminal can issue a prompt tone at this time, such as prompt makes
User " this time identifying whether correct? ".If user confirms, this time identify correct；Conversely, this time identifying mistake, need
It corrects.
Step S33, circulation executes step S31-S32, to improve initial training sample according to multiple collected facial images
This；
In preferred embodiment of the invention, identified as mentioned above it is possible, being executed according to multiple facial images circulation
The work of confirmation can be manually entered the personal information to match, or amendment sample if there is identification mistake using user
The modes such as the data in this library carry out initial training sample perfect.Above-mentioned steps S31-S32 is recycled, as far as possible compared with multiple faces
Image to carry out the perfect of recognition and verification and sample database to initial training sample.
Step S34 will be integrally formed identification data by perfect initial training sample, and be stored in intelligent terminal.
In preferred embodiment of the invention, (the first training sample, the second training sample are formed by three of the above method
This and improve training sample) training sample that acquires, by the image pattern to each personal name of correspondence sample into
Row mode identifies, screens duplicate after cross-matched as a result, simultaneously will finally merge by the initial training sample in evening,
To obtain identifying data (identification model and its related information formed after merging, such as personal information etc.) accordingly, and will
These identification data filings are into intelligent terminal.Specifically, it in preferred embodiment of the invention, is used due to initially relying on
The relational network of person establishes the training sample of corresponding recognition of face, therefore, finally that these identification data filings are whole to intelligence
In the relational network of the user saved in end.
In preferred embodiment of the invention, after forming identification data, intelligent terminal, which can acquire, to be needed to know others
Face, and identified according to identification data, export recognition result.In other words, a series of preparatory shape of above-mentioned training samples is completed
After operation, intelligent terminal can use finally formed identification data to identify face.
In preferred embodiment of the invention, above-mentioned intelligent terminal can be the intelligent terminal with robot appearance.It changes
Yan Zhi in preferred embodiment of the invention, above-mentioned facial image recognition method can be applied to that letter can be carried out with user
In the robot device for ceasing interaction.
The foregoing is merely preferred embodiments of the present invention, are not intended to limit embodiments of the present invention and protection model
It encloses, to those skilled in the art, should can appreciate that all with made by description of the invention and diagramatic content
Equivalent replacement and obviously change obtained scheme, should all be included within the scope of the present invention.
Claims (8)
1. a kind of facial image recognition method is suitable for intelligent terminal；It is characterised in that it includes a pre-training step, to obtain
Corresponding training sample；
The pre-training step specifically includes:
Step S1, the intelligent terminal remotely obtain the communication directory being stored in the mobile terminal of user and user's phase
Volume, is matched using the facial image in user's photograph album with the personal head portrait in the communication directory, described in establishing
Facial image and associated first training sample of the corresponding personal name of personal head portrait in the communication directory；With/
Or
The intelligent terminal remotely search and obtain the corresponding buddy list of the preset user account for being associated with social networks with
And good friend's photograph album, and matched according to the facial image in good friend's photograph album with the personal head portrait in the buddy list,
To establish the facial image and the corresponding personal name associated second of individual's head portrait in the buddy list
Training sample；
Step S2, in conjunction with first training sample and/or second training sample to form an initial training sample；
Step S3 acquires facial image, and carries out recognition and verification according to the initial training sample, final to improve and form one
Identification data；
After forming the identification data, the facial image that the intelligent terminal acquisition needs to identify, and according to the identification data
It is identified, exports recognition result.
2. facial image recognition method as described in claim 1, which is characterized in that in the step S1, form described first
The step of training sample, specifically includes:
Step S11a remotely obtains the facial image in user's photograph album；
Step S12a remotely obtains the communication directory；
Step S13a is matched using the facial image in the communication directory, to find phase in the communication directory
Matched individual's head portrait；
The facial image personal name corresponding with the personal head portrait to match is associated by step S14a；
Step S15a, circulation executes the step S11a-S14a, according to all facial images in user's photograph album
Training forms first training sample.
3. facial image recognition method as described in claim 1, which is characterized in that in the step S1, form described second
The step of training sample, specifically includes:
Step S11b remotely obtains the user account phase by being associated with the user account of at least one social network sites
The facial image in associated good friend's photograph album；
Step S12b remotely obtains the user account phase by being associated with the user account of at least one social network sites
Associated buddy list；
Step S13b is matched using the facial image in the buddy list, to find phase in the buddy list
Matched individual's head portrait；
Step S14b is related by the corresponding personal name of the personal head portrait to match in the facial image and the buddy list
Connection；
Step S15b, circulation executes the step S11b-S14b, according to all facial images in good friend's photograph album
Training forms second training sample.
4. facial image recognition method as described in claim 1, which is characterized in that in the step S1, the intelligent terminal
The mobile terminal is accessed by radio connection, and obtains the corresponding communication directory and institute out of described mobile terminal
State user's photograph album.
5. facial image recognition method as described in claim 1, which is characterized in that the step S3 is specifically included:
Step S31, one facial image of collection in worksite；
Step S32 carries out recognition and verification according to facial image of the initial training sample to the collection in worksite:
If can not identify, the initial training sample is updated according to the facial image of the collection in worksite；It is subsequently diverted to step
S33；
If can identify, directly to step S33；
Step S33, circulation execute the step S31-S32, described to be improved according to the facial image of collection in worksite described in multiple
Initial training sample；
Step S34 will be integrally formed the identification data by the perfect initial training sample, and be stored in the intelligence
In terminal.
6. facial image recognition method as claimed in claim 5, which is characterized in that in the step S32, the intelligent terminal
By issuing prompt tone for the corresponding user's confirmation of the facial image of collection in worksite.
7. facial image recognition method as described in claim 1, which is characterized in that the intelligent terminal be with robot outside
The intelligent terminal of sight.
8. facial image recognition method as described in claim 1, which is characterized in that the acquisition facial image in the step S3
It include: the face figure that acquisition is associated with the people in the initial training sample, the people in user's relational network or stranger
Picture.
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510381767.8A CN106326815B (en) | 2015-06-30 | 2015-06-30 | A kind of facial image recognition method |
| PCT/CN2016/086481 WO2017000807A1 (en) | 2015-06-30 | 2016-06-20 | Facial image recognition method |
| TW105120499A TWI579773B (en) | 2015-06-30 | 2016-06-29 | A face image recognition method |
| HK17105101.6A HK1231600A1 (en) | 2015-06-30 | 2017-05-19 | A face image recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510381767.8A CN106326815B (en) | 2015-06-30 | 2015-06-30 | A kind of facial image recognition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106326815A CN106326815A (en) | 2017-01-11 |
| CN106326815B true CN106326815B (en) | 2019-09-13 |
Family
ID=57607855
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510381767.8A Active CN106326815B (en) | 2015-06-30 | 2015-06-30 | A kind of facial image recognition method |
Country Status (4)
| Country | Link |
|---|---|
| CN (1) | CN106326815B (en) |
| HK (1) | HK1231600A1 (en) |
| TW (1) | TWI579773B (en) |
| WO (1) | WO2017000807A1 (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108985135A (en) * | 2017-06-02 | 2018-12-11 | 腾讯科技（深圳）有限公司 | A kind of human-face detector training method, device and electronic equipment |
| CN108241734A (en) * | 2017-12-01 | 2018-07-03 | 国政通科技股份有限公司 | A kind of information verification integral method and device based on photo library |
| CN108305146A (en) * | 2018-01-30 | 2018-07-20 | 杨太立 | A kind of hair style recommendation method and system based on image recognition |
| CN110309691B (en) * | 2018-03-27 | 2022-12-27 | 腾讯科技（深圳）有限公司 | Face recognition method, face recognition device, server and storage medium |
| CN108764149B (en) * | 2018-05-29 | 2022-02-18 | 北京中庆现代技术股份有限公司 | Training method for class student face model |
| CN109493073B (en) * | 2018-10-25 | 2021-07-16 | 创新先进技术有限公司 | Identity recognition method and device based on human face and electronic equipment |
| TWI759731B (en) * | 2020-04-27 | 2022-04-01 | 淡江大學 | Machine learning method |
| CN112364733B (en) * | 2020-10-30 | 2022-07-26 | 重庆电子工程职业学院 | Intelligent security face recognition system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7274822B2 (en) * | 2003-06-30 | 2007-09-25 | Microsoft Corporation | Face annotation for photo management |
| CN201774591U (en) * | 2010-08-12 | 2011-03-23 | 天津三星光电子有限公司 | Digital camera with address book and face recognition function |
| CN102867173A (en) * | 2012-08-28 | 2013-01-09 | 华南理工大学 | Human face recognition method and system thereof |
| CN103793697A (en) * | 2014-02-17 | 2014-05-14 | 北京旷视科技有限公司 | Identity labeling method of face images and face identity recognition method of face images |
| CN103970830A (en) * | 2014-03-31 | 2014-08-06 | 小米科技有限责任公司 | Information recommendation method and device |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101330636B1 (en) * | 2007-01-24 | 2013-11-18 | 삼성전자주식회사 | Face view determining apparatus and method and face detection apparatus and method employing the same |
| CN102637255A (en) * | 2011-02-12 | 2012-08-15 | 北京千橡网景科技发展有限公司 | Method and device for processing faces contained in images |
| TW201423454A (en) * | 2011-12-09 | 2014-06-16 | Primax Electronics Ltd | System of photo management |
| TW201348984A (en) * | 2012-05-18 | 2013-12-01 | Primax Electronics Ltd | Method for managing photo image and photo image managing system |
| CN102819726B (en) * | 2012-06-27 | 2016-08-24 | 宇龙计算机通信科技(深圳)有限公司 | photo processing system and method for mobile terminal |
| CN108345680A (en) * | 2012-06-29 | 2018-07-31 | 华为终端（东莞）有限公司 | Photograph album and the associated side of address list mutual information and terminal |
| CN102811286A (en) * | 2012-07-27 | 2012-12-05 | 广东欧珀移动通信有限公司 | Group creation method for address book |
| CN103399896B (en) * | 2013-07-19 | 2019-08-23 | 广州华多网络科技有限公司 | The method and system of incidence relation between identification user |
-
2015
- 2015-06-30 CN CN201510381767.8A patent/CN106326815B/en active Active
-
2016
- 2016-06-20 WO PCT/CN2016/086481 patent/WO2017000807A1/en active Application Filing
- 2016-06-29 TW TW105120499A patent/TWI579773B/en not_active IP Right Cessation
-
2017
- 2017-05-19 HK HK17105101.6A patent/HK1231600A1/en not_active IP Right Cessation
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7274822B2 (en) * | 2003-06-30 | 2007-09-25 | Microsoft Corporation | Face annotation for photo management |
| CN201774591U (en) * | 2010-08-12 | 2011-03-23 | 天津三星光电子有限公司 | Digital camera with address book and face recognition function |
| CN102867173A (en) * | 2012-08-28 | 2013-01-09 | 华南理工大学 | Human face recognition method and system thereof |
| CN103793697A (en) * | 2014-02-17 | 2014-05-14 | 北京旷视科技有限公司 | Identity labeling method of face images and face identity recognition method of face images |
| CN103970830A (en) * | 2014-03-31 | 2014-08-06 | 小米科技有限责任公司 | Information recommendation method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| HK1231600A1 (en) | 2017-12-22 |
| TWI579773B (en) | 2017-04-21 |
| CN106326815A (en) | 2017-01-11 |
| WO2017000807A1 (en) | 2017-01-05 |
| TW201701188A (en) | 2017-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106326815B (en) | A kind of facial image recognition method | |
| US10534959B2 (en) | Iris recognition device, manufacturing method therefor and application thereof | |
| WO2017167060A1 (en) | Information display method, device and system | |
| CN103310142B (en) | Based on the human-computer fusion safety certifying method of wearable device | |
| US20130129210A1 (en) | Recommendation system based on the recognition of a face and style, and method thereof | |
| WO2019000777A1 (en) | Internet-based face beautification system | |
| US20170169501A1 (en) | Method and system for evaluating fitness between wearer and eyeglasses | |
| KR102455966B1 (en) | Mediating Apparatus, Method and Computer Readable Recording Medium Thereof | |
| CN110399763B (en) | Face recognition method and system | |
| CN104994105B (en) | A kind of Android intelligent terminal safety certifying method | |
| CN104935440A (en) | Authentication method and authentication device for remote opening of financial account | |
| CN106027543A (en) | Identification method and apparatus based on weight calculation | |
| TW201608489A (en) | Smart glass and method for recognizing and prompting face using smart glass | |
| CN104378659A (en) | Personalization recommendation method based on smart television | |
| KR20120046653A (en) | System and method for recommending hair based on face and style recognition | |
| CN105095917A (en) | Image processing method, device and terminal | |
| CN105338183A (en) | People recognition prompting system and method based on mobile terminal | |
| CN110149618A (en) | Smart machine cut-in method, device, equipment and medium based on vocal print authorization | |
| CN103870735A (en) | Unlocking processing method and device | |
| Bansal et al. | Emotion recognition from facial expression based on bezier curve | |
| CN112242909B (en) | Method and device for generating management template, electronic equipment and storage medium | |
| CN109635138A (en) | A kind of social networks method for building up and system based on similar appearance | |
| CN115953168A (en) | Big data-based information technology consultation management system and method | |
| CN114722051A (en) | Updating method, device, equipment and medium of biological characteristic library | |
| CN110032955B (en) | Novel face recognition method based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1231600 Country of ref document: HK |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |