CN109815775A - A kind of face identification method and system based on face character - Google Patents
A kind of face identification method and system based on face character Download PDFInfo
- Publication number
- CN109815775A CN109815775A CN201711169706.0A CN201711169706A CN109815775A CN 109815775 A CN109815775 A CN 109815775A CN 201711169706 A CN201711169706 A CN 201711169706A CN 109815775 A CN109815775 A CN 109815775A
- Authority
- CN
- China
- Prior art keywords
- face
- facial image
- character
- image
- identification
- 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
Abstract
The invention discloses a kind of face identification methods based on face character, comprising: obtains facial image to be identified；When detecting that the facial image is clear, the face character in the facial image is identified；In retrieval face database with word bank corresponding to the face character that identifies, to carry out recognition of face to the facial image；The face database by storing there are multiple word banks of the face sample image of different faces attribute to form.In addition, the invention discloses a kind of face identification systems based on face character.The present invention can effectively improve the efficiency of recognition of face.
Description
Technical field
The present invention relates to technical field of image processing more particularly to a kind of face identification method based on face character and it is
System.
Background technique
As the research of field of face identification continues to develop, face recognition accuracy rate is higher and higher.In particular with calculating
The continuous promotion of machine hardware, and the acquisition of large-scale face database are more and more convenient, the recognition of face side based on deep learning
Method increasingly useization so that recognition of face using more and more extensive.Mobile payment, access control system, employee are registered system, peace
Anti- system, VIP client management system etc. are all embedded in face recognition algorithms.
Although recognition of face has been widely used, common face identification system is all Custom Prosthesis, Neng Goushi
Other face limited amount, such as the at most thousands of people of access control system of a cell, such system can only become small-sized
Face identification system.Due to the face negligible amounts in face database, speed is influenced little.But for the face of million ranks
Database, the face comparison of Yao Jinhang 1:N, then take a long time, for the occasions such as public security department suspect retrieval, fast lock
Determine suspect's identity to be very important, and entire face database is retrieved, it is very time-consuming.
Summary of the invention
The present invention aiming at the problems existing in the prior art, provide a kind of face identification method based on face character and
System can effectively improve the efficiency of recognition of face.
The technical solution that the present invention is proposed with regard to above-mentioned technical problem is as follows:
On the one hand, the present invention provides a kind of face identification method based on face character, comprising:
Obtain facial image to be identified；
When detecting that the facial image is clear, the face character in the facial image is identified；
In retrieval face database with word bank corresponding to the face character that identifies, to carry out face knowledge to the facial image
Not；The face database by storing there are multiple word banks of the face sample image of different faces attribute to form.
Further, the face character includes gender and age；
Before the acquisition facial image to be identified, further includes:
Obtain face sample image；
Gender identification is carried out to the face sample image；
Age identification is carried out to the face sample image；
Face database is divided into multiple word banks, the face sample image with identical gender and age is made to be stored in a son
In library.
Further, described when detecting that the facial image is clear, identify the face character in the facial image,
It specifically includes:
It is detected using clarity of the Laplace method to the facial image；
When detecting that the clarity is greater than preset threshold, determines that the facial image is clear, identify the face figure
Face character as in.
Further, the method also includes:
When detecting that the clarity is less than preset threshold, determines that the facial image is unintelligible, retrieve entire face
Library, to carry out recognition of face to the facial image.
Further, in the retrieval face database with word bank corresponding to the face character that identifies, to the face
Image carries out recognition of face, specifically includes:
In retrieval face database with word bank corresponding to the face character that identifies；
Calculate separately the similarity of each face sample image in the facial image and the word bank；
Calculated highest similarity is compared with default similar threshold value；
If the highest similarity is greater than default similar threshold value, for described in most by the recognition of face in the facial image
Face corresponding to the face sample image of high similarity；
It is stranger by the recognition of face in the facial image if the highest similarity is less than default similar threshold value
Face.
On the other hand, the present invention provides a kind of face identification system based on face character, can be realized above-mentioned based on people
All processes of the face identification method of face attribute, the system comprises:
Image collection module, for obtaining facial image to be identified；
Identification module, for identifying the face character in the facial image when detecting that the facial image is clear；
And
First retrieval module, for retrieve in face database with word bank corresponding to the face character that identifies, to described
Facial image carries out recognition of face；The face database has multiple word banks of the face sample image of different faces attribute by storing
Composition.
Further, the face character includes gender and age；
The system also includes:
Sample image obtains module, for obtaining face sample image；
Gender identification module, for carrying out gender identification to the face sample image；
Age identification module, for carrying out age identification to the face sample image；And
Division module makes the face sample graph with identical gender and age for face database to be divided into multiple word banks
As being stored in a word bank.
Further, the identification module specifically includes:
Detection unit, for being detected using Laplace method to the clarity of the facial image；
Recognition unit is known for determining that the facial image is clear when detecting that the clarity is greater than preset threshold
Face character in the not described facial image.
Further, the system also includes:
Second retrieval module, for determining the facial image not when detecting that the clarity is less than preset threshold
Clearly, entire face database is retrieved, to carry out recognition of face to the facial image.
Further, first retrieval module specifically includes:
Retrieval unit, for retrieve in face database with word bank corresponding to the face character that identifies；
Similarity calculated, for calculating separately each face sample image in the facial image and the word bank
Similarity；
Comparing unit, for calculated highest similarity to be compared with default similar threshold value；
First face identification unit is used for when the highest similarity is greater than default similar threshold value, by the face figure
Recognition of face as in is face corresponding to the face sample image of the highest similarity；And
Second face identification unit is used for when the highest similarity is less than default similar threshold value, by the face figure
Recognition of face as in is strange face.
Technical solution provided in an embodiment of the present invention has the benefit that
First the face character of facial image to be identified is identified, then to corresponding with the face character in face database
Word bank retrieved the range of search that facial image is greatly reduced to identify facial image, improve recall precision, meanwhile, examine
Consider since face character identifies existing error, the identification of face character is only carried out to clearly facial image, effectively
Improve error hiding problem caused by the error due to existing for image definition and face character identification, reduces misclassification rate and people
Face reject rate.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing.
Fig. 1 is the flow diagram of one embodiment of the face identification method provided by the invention based on face character；
Fig. 2 is the schematic layout pattern of face database in the face identification method provided by the invention based on face character；
Fig. 3 is the process signal of another embodiment of the face identification method provided by the invention based on face character
Figure；
Fig. 4 is one embodiment structural schematic diagram of the face identification system provided by the invention based on face character.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention
Formula is described in further detail.
The embodiment of the invention provides a kind of face identification methods based on face character, referring to Fig. 1, the method packet
It includes:
S1, facial image to be identified is obtained；
S2, when detecting that the facial image is clear, identify the face character in the facial image；
In S3, retrieval face database with word bank corresponding to the face character that identifies, to carry out people to the facial image
Face identification；The face database by storing there are multiple word banks of the face sample image of different faces attribute to form.
It should be noted that in step sl, the acquisition of facial image can be obtained by directly reading image file,
It can be obtained by camera.If the face in image accounts for smaller, method for detecting human face can be used by the human face region in image
It detected, and stored after being cut.
Further, the face character includes gender and age；
Before the acquisition facial image to be identified, further includes:
Obtain face sample image；
Gender identification is carried out to the face sample image；
Age identification is carried out to the face sample image；
Face database is divided into multiple word banks, the face sample image with identical gender and age is made to be stored in a son
In library.
It should be noted that need to be laid out again to face database before being identified to facial image.First obtain face
Sample image, everyone is correspondingly provided with a unique file folder, everyone face sample image is stored in its unique file folder
In.In turn, the face sample image obtained using deep neural network or other any machine learning method traversals, to face
Sample image carries out gender identification, wherein the output neuron number of deep neural network is 2, respectively represents male and female
Probability, the gender recognition result using the high gender of probability as the face sample image, end of identification identifies according to gender
As a result gender label is added to the face sample image.Equally, using deep neural network or other any machine learning methods
The face sample image obtained is traversed, to carry out age identification to face sample image, wherein the output mind of deep neural network
It is preferably 8 through first number, respectively represents 0-2 years old, 4-6 years old, 8-13 years old, 15-20 years old, 25-32 years old, 38-43 years old, 48-53 years old, 60
Year old or more probability, the age recognition result using the high age bracket of probability as the face sample image, end of identification, according to
Age recognition result adds age label to the face sample image.Wherein, due to consideration that age identification has certain mistake
Difference, therefore divided using section discrete age bracket.
According to gender and age label, face database is laid out again.Face database is first divided into two big son according to gender
Each big word bank is divided into 8 word banks according still further to the age, so that each word bank is corresponding with a gender and age label, together by library
When will have the face sample image of identical gender and age label be stored in the word bank with each word bank, as shown in Figure 2.
Wherein, since face sample image is deposited in everyone unique file folder, everyone unique file folder is deposited in
In corresponding word bank.
Further, in step s 2, described when detecting that the facial image is clear, it identifies in the facial image
Face character, specifically include:
It is detected using clarity of the Laplace method to the facial image；
When detecting that the clarity is greater than preset threshold, determines that the facial image is clear, identify the face figure
Face character as in.
Further, the method also includes:
When detecting that the clarity is less than preset threshold, determines that the facial image is unintelligible, retrieve entire face
Library, to carry out recognition of face to the facial image.
It should be noted that after obtaining facial image to be identified, first with Laplace method to facial image
Clarity is assessed, if clarity is high, using convolutional neural networks or other any machine learning methods respectively to face
The gender of image and age are identified, and then carry out face retrieval in the word bank corresponding to the gender and age, with identification
Facial image；If clarity is low, face retrieval is directly carried out in entire face database, to identify facial image.
Further, in step s3, in the retrieval face database with word bank corresponding to the face character that identifies, with
Recognition of face is carried out to the facial image, is specifically included:
In retrieval face database with word bank corresponding to the face character that identifies；
Calculate separately the similarity of each face sample image in the facial image and the word bank；
Calculated highest similarity is compared with default similar threshold value；
If the highest similarity is greater than default similar threshold value, for described in most by the recognition of face in the facial image
Face corresponding to the face sample image of high similarity；
It is stranger by the recognition of face in the facial image if the highest similarity is less than default similar threshold value
Face.
It should be noted that, first according to face sample image training deep neural network, being obtained before retrieving face in word bank
The feature vector of each face sample image is taken accordingly to be stored in word bank.When retrieval, knowledge is treated first with face alignment method
Other facial image is zoomed in and out and is rotated, and makes the face sample image of the face and training deep neural network in facial image
In face alignment keep deep neural network defeated and then using the facial image after alignment as the input of deep neural network
A feature vector (row vector) out, by this feature vector respectively with the feature vector of each of word bank face sample image into
Row comparison, calculates similarity, and wherein the methods of the optional Euclidean distance of similarity, cosine similarity or mahalanobis distance are counted
It calculates.In turn, the maximum similarity in comparing result is compared with default similar threshold value, maximum similarity is similar higher than presetting
Threshold value then determines face corresponding to face sample image of the face in the facial image for highest similarity, otherwise determines
Face in the facial image is strange face.In addition, when the clarity of facial image to be identified is low, in entire face database
In retrieved and identified, it is identical that method for distinguishing is retrieved in word bank and known to the method used, herein no longer in detail
It repeats.
It is the process of another embodiment of the face identification method provided by the invention based on face character referring to Fig. 3
Schematic diagram, which comprises
S301, facial image is obtained.
S302, judge whether picture quality is high；If so, S303 is thened follow the steps, if it is not, thening follow the steps S306.
S303, gender classification.
S304, the identification of face age.
S305, face retrieval is carried out for corresponding gender and the sub- face database at age.
S306, face retrieval is carried out for entire face database.
The embodiment of the present invention first identifies the face character of facial image to be identified, then in face database with the people
The corresponding word bank of face attribute is retrieved the range of search that facial image is greatly reduced to identify facial image, improves retrieval
Efficiency, simultaneously, it is contemplated that since face character identifies existing error, face character only is carried out to clearly facial image
Identification, be effectively improved since image definition and face character identify error hiding problem caused by existing error, drop
Low misclassification rate and face reject rate.
The embodiment of the invention provides a kind of face identification systems based on face character, can be realized above-mentioned based on face
All processes of the face identification method of attribute, referring to fig. 4, the system comprises:
Image collection module 1, for obtaining facial image to be identified；
Identification module 2, for when detecting that the facial image is clear, identifying the face category in the facial image
Property；And
First retrieval module 3, for retrieve in face database with word bank corresponding to the face character that identifies, to described
Facial image carries out recognition of face；The face database has multiple word banks of the face sample image of different faces attribute by storing
Composition.
Further, the face character includes gender and age；
The system also includes:
Sample image obtains module, for obtaining face sample image；
Gender identification module, for carrying out gender identification to the face sample image；
Age identification module, for carrying out age identification to the face sample image；And
Division module makes the face sample graph with identical gender and age for face database to be divided into multiple word banks
As being stored in a word bank.
Further, the identification module specifically includes:
Detection unit, for being detected using Laplace method to the clarity of the facial image；
Recognition unit is known for determining that the facial image is clear when detecting that the clarity is greater than preset threshold
Face character in the not described facial image.
Further, the system also includes:
Second retrieval module, for determining the facial image not when detecting that the clarity is less than preset threshold
Clearly, entire face database is retrieved, to carry out recognition of face to the facial image.
Further, first retrieval module specifically includes:
Retrieval unit, for retrieve in face database with word bank corresponding to the face character that identifies；
Similarity calculated, for calculating separately each face sample image in the facial image and the word bank
Similarity；
Comparing unit, for calculated highest similarity to be compared with default similar threshold value；
First face identification unit is used for when the highest similarity is greater than default similar threshold value, by the face figure
Recognition of face as in is face corresponding to the face sample image of the highest similarity；And
Second face identification unit is used for when the highest similarity is less than default similar threshold value, by the face figure
Recognition of face as in is strange face.
The embodiment of the present invention first identifies the face character of facial image to be identified, then in face database with the people
The corresponding word bank of face attribute is retrieved the range of search that facial image is greatly reduced to identify facial image, improves retrieval
Efficiency, simultaneously, it is contemplated that since face character identifies existing error, face character only is carried out to clearly facial image
Identification, be effectively improved since image definition and face character identify error hiding problem caused by existing error, drop
Low misclassification rate and face reject rate.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of face identification method based on face character characterized by comprising
Obtain facial image to be identified；
When detecting that the facial image is clear, the face character in the facial image is identified；
In retrieval face database with word bank corresponding to the face character that identifies, to carry out recognition of face to the facial image；
The face database by storing there are multiple word banks of the face sample image of different faces attribute to form.
2. as described in claim 1 based on the face identification method of face character, which is characterized in that the face character includes
Gender and age；
Before the acquisition facial image to be identified, further includes:
Obtain face sample image；
Gender identification is carried out to the face sample image；
Age identification is carried out to the face sample image；
Face database is divided into multiple word banks, the face sample image with identical gender and age is made to be stored in a word bank
In.
3. as described in claim 1 based on the face identification method of face character, which is characterized in that it is described detect it is described
When facial image is clear, identifies the face character in the facial image, specifically includes:
It is detected using clarity of the Laplace method to the facial image；
When detecting that the clarity is greater than preset threshold, determines that the facial image is clear, identify in the facial image
Face character.
4. as claimed in claim 3 based on the face identification method of face character, which is characterized in that the method also includes:
When detecting that the clarity is less than preset threshold, determines that the facial image is unintelligible, retrieve entire face database, with
Recognition of face is carried out to the facial image.
5. as described in claim 1 based on the face identification method of face character, which is characterized in that in the retrieval face database
It is specifically included with word bank corresponding to the face character that identifies with carrying out recognition of face to the facial image:
In retrieval face database with word bank corresponding to the face character that identifies；
Calculate separately the similarity of each face sample image in the facial image and the word bank；
Calculated highest similarity is compared with default similar threshold value；
It is the highest phase by the recognition of face in the facial image if the highest similarity is greater than default similar threshold value
The face like corresponding to the face sample image of degree；
It is strange face by the recognition of face in the facial image if the highest similarity is less than default similar threshold value.
6. a kind of face identification system based on face character characterized by comprising
Image collection module, for obtaining facial image to be identified；
Identification module, for identifying the face character in the facial image when detecting that the facial image is clear；With
And
First retrieval module, for retrieve in face database with word bank corresponding to the face character that identifies, to the face
Image carries out recognition of face；The face database has multiple word bank groups of the face sample image of different faces attribute by storing
At.
7. as claimed in claim 6 based on the face identification system of face character, which is characterized in that the face character includes
Gender and age；
The system also includes:
Sample image obtains module, for obtaining face sample image；
Gender identification module, for carrying out gender identification to the face sample image；
Age identification module, for carrying out age identification to the face sample image；And
Division module deposits the face sample image with identical gender and age for face database to be divided into multiple word banks
Storage is in a word bank.
8. as claimed in claim 6 based on the face identification system of face character, which is characterized in that the identification module is specific
Include:
Detection unit, for being detected using Laplace method to the clarity of the facial image；
Recognition unit identifies institute for determining that the facial image is clear when detecting that the clarity is greater than preset threshold
State the face character in facial image.
9. as claimed in claim 8 based on the face identification system of face character, which is characterized in that the system also includes:
Second retrieval module, for determining that the facial image is unintelligible when detecting that the clarity is less than preset threshold,
Entire face database is retrieved, to carry out recognition of face to the facial image.
10. as claimed in claim 6 based on the face identification system of face character, which is characterized in that the first retrieval mould
Block specifically includes:
Retrieval unit, for retrieve in face database with word bank corresponding to the face character that identifies；
Similarity calculated, it is similar to face sample image each in the word bank for calculating separately the facial image
Degree；
Comparing unit, for calculated highest similarity to be compared with default similar threshold value；
First face identification unit is used for when the highest similarity is greater than default similar threshold value, will be in the facial image
Recognition of face be the highest similarity face sample image corresponding to face；And
Second face identification unit is used for when the highest similarity is less than default similar threshold value, will be in the facial image
Recognition of face be strange face.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711169706.0A CN109815775A (en) | 2017-11-22 | 2017-11-22 | A kind of face identification method and system based on face character |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711169706.0A CN109815775A (en) | 2017-11-22 | 2017-11-22 | A kind of face identification method and system based on face character |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109815775A true CN109815775A (en) | 2019-05-28 |
Family
ID=66600877
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711169706.0A Pending CN109815775A (en) | 2017-11-22 | 2017-11-22 | A kind of face identification method and system based on face character |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109815775A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110738777A (en) * | 2019-09-27 | 2020-01-31 | 恒大智慧科技有限公司 | Property management method and system, and storage medium |
| CN110826525A (en) * | 2019-11-18 | 2020-02-21 | 天津高创安邦技术有限公司 | Face recognition method and system |
| CN111259813A (en) * | 2020-01-17 | 2020-06-09 | 实地地产集团有限公司 | Face detection tracking method and device, computer equipment and storage medium |
| CN111310725A (en) * | 2020-03-13 | 2020-06-19 | 恒睿(重庆)人工智能技术研究院有限公司 | Object identification method, system, machine readable medium and device |
| CN111708906A (en) * | 2020-06-16 | 2020-09-25 | 重庆锐云科技有限公司 | Visiting retrieval method, device and equipment based on face recognition and storage medium |
| CN112052733A (en) * | 2020-07-31 | 2020-12-08 | 中国建设银行股份有限公司 | Database construction method, face recognition device and electronic equipment |
| CN112084904A (en) * | 2020-08-26 | 2020-12-15 | 武汉普利商用机器有限公司 | Face searching method, device and storage medium |
| CN112733717A (en) * | 2021-01-11 | 2021-04-30 | 广州市金其利信息科技有限公司 | Method for optimizing face recognition based on face attributes |
| CN113409055A (en) * | 2021-06-30 | 2021-09-17 | 深圳市商汤科技有限公司 | Payment method, system, electronic device and storage medium |
| CN113469698A (en) * | 2021-06-30 | 2021-10-01 | 深圳市商汤科技有限公司 | Registration method, system, electronic device and storage medium |
| CN113822189A (en) * | 2021-09-15 | 2021-12-21 | 珠海格力电器股份有限公司 | Face recognition method and device, computer readable storage medium and processor |
| WO2022179046A1 (en) * | 2021-02-26 | 2022-09-01 | 深圳壹账通智能科技有限公司 | Facial recognition method and apparatus, computer device, and storage medium |
-
2017
- 2017-11-22 CN CN201711169706.0A patent/CN109815775A/en active Pending
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110738777A (en) * | 2019-09-27 | 2020-01-31 | 恒大智慧科技有限公司 | Property management method and system, and storage medium |
| CN110826525A (en) * | 2019-11-18 | 2020-02-21 | 天津高创安邦技术有限公司 | Face recognition method and system |
| CN110826525B (en) * | 2019-11-18 | 2023-05-26 | 天津高创安邦技术有限公司 | Face recognition method and system |
| CN111259813A (en) * | 2020-01-17 | 2020-06-09 | 实地地产集团有限公司 | Face detection tracking method and device, computer equipment and storage medium |
| CN111259813B (en) * | 2020-01-17 | 2024-01-12 | 珠海市横琴盈实科技研发有限公司 | Face detection tracking method, device, computer equipment and storage medium |
| CN111310725A (en) * | 2020-03-13 | 2020-06-19 | 恒睿(重庆)人工智能技术研究院有限公司 | Object identification method, system, machine readable medium and device |
| CN111708906A (en) * | 2020-06-16 | 2020-09-25 | 重庆锐云科技有限公司 | Visiting retrieval method, device and equipment based on face recognition and storage medium |
| CN111708906B (en) * | 2020-06-16 | 2022-12-27 | 重庆锐云科技有限公司 | Visiting retrieval method, device and equipment based on face recognition and storage medium |
| CN112052733A (en) * | 2020-07-31 | 2020-12-08 | 中国建设银行股份有限公司 | Database construction method, face recognition device and electronic equipment |
| CN112084904A (en) * | 2020-08-26 | 2020-12-15 | 武汉普利商用机器有限公司 | Face searching method, device and storage medium |
| CN112733717A (en) * | 2021-01-11 | 2021-04-30 | 广州市金其利信息科技有限公司 | Method for optimizing face recognition based on face attributes |
| CN112733717B (en) * | 2021-01-11 | 2021-11-23 | 广州市金其利信息科技有限公司 | Method for optimizing face recognition based on face attributes |
| WO2022179046A1 (en) * | 2021-02-26 | 2022-09-01 | 深圳壹账通智能科技有限公司 | Facial recognition method and apparatus, computer device, and storage medium |
| CN113469698A (en) * | 2021-06-30 | 2021-10-01 | 深圳市商汤科技有限公司 | Registration method, system, electronic device and storage medium |
| CN113409055A (en) * | 2021-06-30 | 2021-09-17 | 深圳市商汤科技有限公司 | Payment method, system, electronic device and storage medium |
| CN113822189A (en) * | 2021-09-15 | 2021-12-21 | 珠海格力电器股份有限公司 | Face recognition method and device, computer readable storage medium and processor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109815775A (en) | A kind of face identification method and system based on face character | |
| CN109344787B (en) | Specific target tracking method based on face recognition and pedestrian re-recognition | |
| CN102201061B (en) | Intelligent safety monitoring system and method based on multilevel filtering face recognition | |
| CN107392222B (en) | Face clustering method and device and storage medium | |
| EP2113846B1 (en) | Behavior history searching device and behavior history searching method | |
| CN109190514B (en) | Face attribute recognition method and system based on bidirectional long-short term memory network | |
| US8064653B2 (en) | Method and system of person identification by facial image | |
| CN105825176A (en) | Identification method based on multi-mode non-contact identity characteristics | |
| US8953852B2 (en) | Method for face recognition | |
| CN107122439A (en) | A kind of video segment querying method and device | |
| Yen et al. | Facial feature extraction using genetic algorithm | |
| CN111968152A (en) | Dynamic identity recognition method and device | |
| CN110941993A (en) | Dynamic personnel classification and storage method based on face recognition | |
| Manikandan et al. | A neural network aided attuned scheme for gun detection in video surveillance images | |
| Sandhya et al. | Deep Learning Based Face Detection and Identification of Criminal Suspects. | |
| Parekh et al. | Review of Parameters of Fingerprint Classification Methods Based on Algorithmic Flow | |
| Irbaz et al. | Real-time face recognition system for remote employee tracking | |
| Karakuş et al. | Real-Time Detection and Identification of Suspects in Forensic Imagery Using Advanced YOLOv8 Object Recognition Models. | |
| Devi et al. | Deep learn helmets-enhancing security at ATMs | |
| Sivasangari et al. | Facial recognition system using decision tree algorithm | |
| Wu et al. | A new approach for classification of fingerprint image quality | |
| CN117113321B (en) | Image searching method and system for searching face by face | |
| Mohammad et al. | Cross Input Neighbourhood Difference for Re-identification of Occupational Humans | |
| Desai et al. | Person re-identification via deep metric learning | |
| Natsheh et al. | Authentication System by Facial Recognition with Principal Component Analysis and Deep Neural Networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190528 |