CN111950364B - System and method for identifying library-separating face of tens of millions of libraries - Google Patents
System and method for identifying library-separating face of tens of millions of libraries Download PDFInfo
- Publication number
- CN111950364B CN111950364B CN202010647085.8A CN202010647085A CN111950364B CN 111950364 B CN111950364 B CN 111950364B CN 202010647085 A CN202010647085 A CN 202010647085A CN 111950364 B CN111950364 B CN 111950364B
- Authority
- CN
- China
- Prior art keywords
- library
- face
- face picture
- module
- stranger
- 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
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000008520 organization Effects 0.000 claims abstract description 8
- 230000007246 mechanism Effects 0.000 claims abstract description 6
- 238000012937 correction Methods 0.000 claims description 13
- 238000007405 data analysis Methods 0.000 claims description 8
- 230000005012 migration Effects 0.000 claims description 6
- 238000013508 migration Methods 0.000 claims description 6
- 230000030279 gene silencing Effects 0.000 claims description 5
- 238000011161 development Methods 0.000 abstract description 7
- 230000008569 process Effects 0.000 description 6
- 238000000926 separation method Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006872 improvement Effects 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
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/51—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
- G06V40/173—Classification, e.g. identification face re-identification, e.g. recognising unknown faces across different face tracks
-
- 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
- G06V40/50—Maintenance of biometric data or enrolment thereof
Abstract
The invention discloses a system and a method for identifying the face of a sub-library of a tens of millions of libraries, wherein the method comprises the following steps: receiving a face picture acquired by image acquisition equipment, and acquiring information of the image acquisition equipment, an organization mechanism and an area to which the image acquisition equipment belongs; determining a database corresponding to an attribution area of the image acquisition equipment; identifying face pictures in the acquaintance library, if not, identifying in the corresponding city library, and if not, identifying in the silent library; if face pictures are identified in the acquaintance library, the local market library or the silence library, natural person information corresponding to the face pictures is obtained; if the customer corresponding to the face picture is not recognized in the silent library, judging that the customer corresponding to the face picture is a stranger, and sending the recognition information corresponding to the face picture to a stranger base library. By the technical scheme, the development cost and the hardware cost are reduced while the function operation of the tens of millions of magnitude database is ensured, and the new and increased capability of the face database in transverse expansion is improved.
Description
Technical Field
The invention relates to the technical field of databases, in particular to a multi-tens of millions of base database face recognition system and a multi-tens of millions of base database face recognition method.
Background
In the process of product landing, the intelligent business hall project has the library building requirement of a face base by the AI-based face recognition technology, and the face base data is important, but the existing product has less face base with the body quantity of more than 10W level. When the face database reaches more than 10W, the problems of low identification accuracy and insufficient performance exist, so that the database separation strategy becomes a special strategy to solve.
Disclosure of Invention
Aiming at the problems, the invention provides a multi-tens of thousands-of-class database sub-library face recognition system and method, which are characterized in that the face recognition process is combined with the hall shop attribute of business hall image acquisition equipment to perform data routing through a distributed processing sub-library strategy, tens of thousands of class data are distributed in a plurality of multi-tens of thousands or millions of class database to be processed, the function operation of the tens of thousands of class database is ensured, the development cost and hardware cost are reduced, the development efficiency is improved, the new and increased capability of the face database is transversely expanded, and the high availability of the face recognition system is improved.
In order to achieve the above object, the present invention provides a multi-ten-millions-of-base-based face recognition system, comprising: the system comprises an image acquisition module, a data analysis module, a acquaintance library module, a ground city library module, a silencing library module, a stranger library module and an image identification module; the image acquisition module is used for acquiring face pictures; the data analysis module is used for acquiring equipment information of the image acquisition module, an organization mechanism where the equipment information is located and a region where the equipment information belongs to, and determining a database corresponding to the region where the equipment information belongs to; the acquaintance library module, the city library module, the silencing library module and the stranger library module are respectively used for storing face pictures and visit record information of clients in a separate mode; the image recognition module is used for carrying out step-by-step recognition comparison on the collected face picture and the face pictures of the clients in the acquaintance library module, the ground city library module and the silence library module, acquiring natural person information corresponding to the face picture when the recognition is successful, judging that the client corresponding to the face picture is a stranger when the recognition is unsuccessful, and sending the recognition information to the stranger base.
In the above technical solution, preferably, the database-division face recognition system of the millions of base databases further includes a data updating module and a data pushing module, after obtaining the natural person information corresponding to the face picture, the data updating module updates and queries a natural person table, obtains the last time the natural person was acquired, and judges whether the face picture acquisition is a new visit according to preset logic and the time difference between the two acquired face pictures; if the face picture is acquired as a new visit, generating a new visit record, and storing the acquired face picture after matching with the visit record; if the face picture acquisition is not a new visit, inquiring a visit record, and matching the acquired face picture with the latest visit record and warehousing; and updating the natural person table according to the face picture in the warehouse, identifying the natural person table and pushing the natural person table to a designated party through the data pushing module.
In the above technical solution, preferably, the database-division face recognition system of the millions of databases further includes a data correction module, where the data correction module obtains information of the record table or the natural person table, analyzes the identified face picture in combination with the big data module, queries whether the face picture exists in the corresponding database, obtains a home area corresponding to the database if the face picture exists, determines whether the database is matched, and performs database data migration if the database is not matched; if the data in the stranger bottom library does not exist, inquiring the stranger bottom library, and migrating the data in the stranger bottom library to a local market library in a corresponding attribution area.
In the above technical solution, preferably, the acquaintance library module is a customer database in which the number of times of collecting the faces in a preset time reaches a preset number, the silence library module is a customer database in which the faces are not collected in the preset time, and the city library module is a customer database corresponding to an attribution area of an image collecting device for collecting the faces.
The invention also provides a library-division face recognition method of the millions of libraries, which is applied to the library-division face recognition system of the tens of millions of libraries according to any one of the technical schemes, and comprises the following steps: receiving a face picture acquired by image acquisition equipment, and acquiring information of the image acquisition equipment, an organization mechanism and a region to which the image acquisition equipment belongs; determining a database corresponding to the attribution area of the image acquisition equipment according to the information of the image acquisition equipment; identifying the face picture in a acquaintance library, if the face picture is not identified in the acquaintance library, identifying the face picture in a city library corresponding to the image acquisition equipment, and if the face picture is not identified in the city library, identifying the face picture in a silent library which is not acquired by the image acquisition equipment within a preset time; if the face picture is identified in the acquaintance library, the city library or the silence library, natural person information corresponding to the face picture is obtained; if the customer corresponding to the face picture is not recognized in the silence library, judging that the customer corresponding to the face picture is a stranger, and sending the recognition information corresponding to the face picture to a stranger base library.
In the above technical solution, preferably, the method for identifying the sub-library face of the millions of base libraries further includes: after natural person information corresponding to the face picture is obtained, updating and inquiring a natural person table, obtaining the last time the natural person is acquired, and judging whether the face picture acquisition is a new visit according to preset logic and the time difference of the two acquired face pictures; if the face picture is acquired as a new visit, generating a new visit record, and storing the acquired face picture after matching with the visit record; if the face picture acquisition is not a new visit, inquiring a visit record, and matching the acquired face picture with the latest visit record and warehousing; and updating the natural person table according to the face picture in the warehouse, identifying the natural person table and pushing the natural person table to a designated party.
In the above technical solution, preferably, the determining that the client corresponding to the face picture is a stranger and sending the identification information corresponding to the face picture to the stranger base specifically includes: after judging that the corresponding customer of the face picture is a stranger, re-identifying the face picture, if the identification is successful, warehousing the face picture to a stranger table, and if the identification is not successful, warehousing the face picture to a stranger bottom library, and then warehousing the face picture to the stranger table; and updating the information of the new strangers in the warehouse to a record list according to the stranger list, identifying the record list and pushing the record list to a designated party.
In the above technical solution, preferably, the method for identifying the sub-library face of the millions of base libraries further includes: acquiring information of the record table or the natural person table, and analyzing the identified face picture; inquiring whether the face picture exists in the corresponding database, if so, acquiring a attribution area corresponding to the database, judging whether the database is matched, and if not, performing database data migration; if the face picture does not exist in the corresponding database, inquiring a stranger database, and transferring the data in the stranger database to a local market database of a corresponding attribution area.
In the above technical solution, preferably, the large data module is combined to analyze and query the identified face picture, so as to migrate the database to implement data correction.
In the above technical solution, preferably, the customer data of the face collected in the silent library is migrated to a local market library in the area of the image collecting device, the customer data of the face collected in the local market library with the number of times reaching the preset number in the preset time is migrated to the acquaintance library, and the customer data of the face collected in the local market library with the number of times not reaching the preset number in the preset time is migrated to the silent library.
Compared with the prior art, the invention has the beneficial effects that: through a distributed processing database separation strategy, the face recognition process is combined with the hall store attribute of the business hall image acquisition equipment to carry out data routing, tens of millions of magnitude data are distributed in a plurality of hundreds of thousands or millions of magnitude database to be processed, the development cost and the hardware cost are reduced while the function operation of the tens of millions of magnitude database is ensured, the development efficiency is improved, the transverse expansion new increasing capacity of the face database is improved, and the high availability of the face recognition system is improved.
Drawings
Fig. 1 is a schematic block diagram of a library-division face recognition system of a tens of millions of libraries according to an embodiment of the present invention;
fig. 2 is a flow chart of a method for identifying a split library of a millions of libraries according to an embodiment of the present invention;
fig. 3 is a flow chart of a data correction method according to an embodiment of the invention.
In the figure, the correspondence between each component and the reference numeral is:
11. the system comprises an image acquisition module, a data analysis module, a acquaintance library module, a ground city library module, a silence library module, a stranger library module, an image identification module, a data updating module, a data pushing module and a data correction module.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention is described in further detail below with reference to the attached drawing figures:
as shown in fig. 1, a multi-ten-millions-level base-based face recognition system according to the present invention includes: the system comprises an image acquisition module 11, a data analysis module 12, a acquaintance library module 13, a city library module 14, a silencing library module 15, a stranger library module 16 and an image identification module 17; the image acquisition module 11 is used for acquiring face pictures; the data analysis module 12 is used for acquiring equipment information, an organization structure and a region of the image acquisition module 11, and determining a database corresponding to the region; the acquaintance library module 13, the city library module 14, the silence library module 15 and the stranger library module 16 are respectively used for storing face pictures and visit record information of the clients in a separate manner; the image recognition module 17 is configured to perform step-by-step recognition and comparison on the collected face picture and the face pictures of the clients in the acquaintance library module 13, the city library module 14 and the silence library module 15, obtain natural person information corresponding to the face picture when the recognition is successful, determine that the client corresponding to the face picture is a stranger when the recognition is unsuccessful, and send the recognition information to the stranger library.
In the embodiment, the data routing is performed by combining the face recognition process with the hall property of the business hall image acquisition equipment through the distributed processing database separation strategy, tens of millions of data are distributed in a plurality of hundreds of thousands or millions of databases for processing, the development cost and the hardware cost are reduced while the function operation of the tens of millions of databases is ensured, the development efficiency is improved, the new and increased capacity of the face database is improved in the transverse expansion, and the high availability of the face recognition system is improved.
Preferably, the acquaintance library module 13 is a customer database in which the number of faces collected in a preset time reaches a preset number, the silence library module 15 is a customer database in which faces are not collected in a preset time, and the city library module 14 is a customer database corresponding to the attribution area of the image collecting device for collecting faces.
Specifically, in this embodiment, taking the face recognition system of the intelligent business hall as an example, the image acquisition module 11 is an image acquisition device (preferably a camera) in the business hall, and the acquaintance library module 13, the city library module 14, the silence library module 15 and the stranger library module 16 are different databases respectively used for storing face pictures and visit record information of the visiting clients who visit the intelligent business hall. The data analysis module 12 determines a database corresponding to the region of the image acquisition device by analyzing the organization code, the region, the device IP, the device ID, and the like of the region to which the image acquisition device belongs, so as to support a face database of a larger magnitude by a database separation strategy of the region to which the image acquisition device belongs.
The image recognition module 17 is used for comparing and recognizing face pictures captured by the image capturing device in the business hall with databases of different levels respectively, and judging whether the databases comprise face information matched with the face pictures. The sequence of face picture comparison and identification is carried out from the database of the high-frequency visiting business hall to the database of the low-frequency visiting business hall, each database is classified according to the visiting frequency and the belonging area, the database separation strategy is designed in a nested manner with the face identification flow, and the functions of the database of tens of millions or higher are realized in a distributed manner through the smaller database of hundreds of thousands or millions.
In the above embodiment, preferably, the database-division face recognition system of the millions of base further includes a data updating module 18 and a data pushing module 19, after obtaining the natural person information corresponding to the face picture, the data updating module 18 updates and queries the natural person table, obtains the last time the natural person was collected, and determines whether the face picture collection is a new visit according to preset logic and the time difference between the two collected face pictures; if the face picture is acquired as a new visit, generating a new visit record, and storing the acquired face picture after matching with the visit record; if the face picture acquisition is not a new visit, inquiring a visit record, and matching the acquired face picture with the latest visit record and warehousing; and updating the natural person table according to the face picture in the warehouse, identifying the natural person table and pushing the natural person table to a designated party through a data pushing module 19.
The data updating module 18 updates the visit record and the strange visitor table in the face recognition system according to the comparison and recognition result through the comparison and recognition between the image recognition module 17 and the acquaintance library module 13, the city library module 14 and the silence library module 15, stores the corresponding face picture into the corresponding database, and pushes the face picture and the table information to a preset appointed party, such as a data correction module through the data pushing module 19.
In the above embodiment, preferably, the database-division face recognition system of the millions of databases further includes a data correction module 20, the data correction module 20 obtains information of a record table or a natural person table, analyzes the identified face picture in combination with the big data module, queries whether the face picture exists in the corresponding database, if so, obtains a attribution area corresponding to the database, determines whether the database is matched, and if not, performs database data migration; if the data in the stranger bottom library does not exist, inquiring the stranger bottom library, and migrating the data in the stranger bottom library to a local market library in a corresponding attribution area.
As shown in fig. 2, the present invention further provides a method for identifying a sub-library of a multi-ten-million-level library, which is applied to the sub-library identification system of a multi-ten-million-level library according to any one of the above embodiments, and includes: receiving a face picture acquired by image acquisition equipment, and acquiring information of the image acquisition equipment, an organization mechanism and an area to which the image acquisition equipment belongs; determining a database corresponding to the attribution area of the image acquisition equipment according to the information of the image acquisition equipment; identifying the face picture in the acquaintance library, if the face picture is not identified in the acquaintance library, identifying the face picture in a local market library corresponding to the image acquisition equipment, and if the face picture is not identified in the local market library, identifying the face picture in a silent library which is not acquired by the image acquisition equipment within a preset time; if face pictures are identified in the acquaintance library, the local market library or the silence library, natural person information corresponding to the face pictures is obtained; if the customer corresponding to the face picture is not recognized in the silent library, judging that the customer corresponding to the face picture is a stranger, and sending the recognition information corresponding to the face picture to a stranger base library.
In the embodiment, the face pictures acquired by the image acquisition equipment in the business hall are respectively compared and identified with the databases of different levels, and whether the databases contain the face information matched with the face pictures is judged. The sequence of face picture comparison and identification is carried out from the database of the high-frequency visiting business hall to the database of the low-frequency visiting business hall, each database is classified according to the visiting frequency and the belonging area, the database separation strategy is designed in a nested manner with the face identification flow, and the functions of the database of tens of millions or higher orders are realized in a distributed manner through the smaller database of hundreds of thousands or millions.
In the above embodiment, preferably, the method for identifying the sub-library face of the millions of base libraries further includes: after natural person information corresponding to the face picture is obtained, updating and inquiring a natural person table, obtaining the last time the natural person is acquired, and judging whether the face picture acquisition is a new visit or not according to preset logic and the time difference of the two acquired face pictures; if the face picture is acquired as a new visit, generating a new visit record, and storing the acquired face picture after matching with the visit record; if the face picture acquisition is not a new visit, inquiring a visit record, and matching the acquired face picture with the latest visit record and warehousing; and updating the natural person table according to the face picture in the warehouse, identifying the natural person table and pushing the natural person table to the appointed party. Through the comparison and identification between the face pictures and the acquaintance library module 13, the city library module 14 and the silence library module 15, the visit records and strange visitor tables in the face recognition system are updated according to the comparison and identification results, the corresponding face pictures are put in the corresponding database, and the face pictures and table information are pushed to preset appointed parties.
In the foregoing embodiment, preferably, the determining that the client corresponding to the face picture is a stranger and sending the identification information corresponding to the face picture to the stranger base specifically includes: after judging that the corresponding customer of the face picture is a stranger, re-identifying the face picture, if the identification is successful, warehousing the face picture to a stranger table, if the identification is not successful, warehousing the face picture to a stranger bottom library, and then warehousing the face picture to the stranger table; and updating the information of the new strangers in the warehouse to the record list according to the stranger list, identifying the record list and pushing the record list to the appointed party.
As shown in fig. 3, in the above embodiment, preferably, the method for identifying the sub-library face of the multi-ten-millions of libraries further includes: acquiring information of a record table or a natural human table, and analyzing the identified human face picture by combining a big data module; inquiring whether the face picture exists in the corresponding database by combining the big data module, if so, acquiring a attribution area corresponding to the database, judging whether the database is matched, and if not, performing database data migration; if the face picture does not exist in the corresponding database, inquiring the stranger database, and transferring the data in the stranger database to a local market database of the corresponding attribution area.
In the above embodiment, preferably, in the data correction process, the customer data of the face collected in the silent library is migrated to the local market library in the image collecting device area, the customer data of the face collected in the local market library with the number of times reaching the preset number in the preset time is migrated to the acquaintance library, and the customer data of the face collected in the local market library with the number of times not reaching the preset number in the preset time is migrated to the silent library.
The data in the data base are shifted through the background, and as each data base is classified according to the visiting frequency and the belonging area, the shifting process shifts the data to the data base corresponding to the visiting frequency according to the updating of the visiting frequency information in the record table or the natural person table, and data correction is realized, so that the new visiting customer can be identified more quickly and optimally.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. A multi-level library-based face recognition system, comprising: the system comprises an image acquisition module, a data analysis module, a acquaintance library module, a ground market library module, a silencing library module, a stranger library module, an image identification module, a data updating module and a data pushing module;
the image acquisition module is used for acquiring face pictures;
the data analysis module is used for acquiring equipment information of the image acquisition module, an organization mechanism where the equipment information is located and a region where the equipment information belongs to, and determining a database corresponding to the region where the equipment information belongs to;
the acquaintance library module, the city library module, the silencing library module and the stranger library module are respectively used for storing face pictures and visit record information of clients in a separate mode;
the image recognition module is used for carrying out step-by-step recognition comparison on the acquired face picture and the face pictures of the clients in the acquaintance library module, the ground city library module and the silence library module, acquiring natural person information corresponding to the face picture when the recognition is successful, judging that the client corresponding to the face picture is a stranger when the recognition is unsuccessful, and sending the recognition information to a stranger base;
after natural person information corresponding to the face picture is obtained, the data updating module updates and inquires a natural person table, obtains the last time the natural person is acquired, and judges whether the face picture acquisition is a new visit according to preset logic and the time difference of the two acquired face pictures;
if the face picture is acquired as a new visit, generating a new visit record, and storing the acquired face picture after matching with the visit record;
if the face picture acquisition is not a new visit, inquiring a visit record, and matching the acquired face picture with the latest visit record and warehousing;
and updating the natural person table according to the face picture in the warehouse, identifying the natural person table and pushing the natural person table to a designated party through the data pushing module.
2. The system for identifying the sub-library face of the millions of sub-libraries according to claim 1, further comprising a data correction module, wherein the data correction module acquires information of a record table or the natural person table, analyzes the identified face picture by combining a big data module, inquires whether the face picture exists in the corresponding sub-library, acquires a attribution area corresponding to the sub-library if the face picture exists, judges whether the sub-library is matched, and performs sub-library data migration if the sub-library is not matched; if the data in the stranger bottom library does not exist, inquiring the stranger bottom library, and migrating the data in the stranger bottom library to a local market library in a corresponding attribution area.
3. The system for classifying and identifying the base of millions of people according to claim 1, wherein the acquaintance base module is a customer database of which the number of times of collecting the faces reaches a preset number in a preset time, the silence base module is a customer database of which the faces are not collected in the preset time, and the city base module is a customer database corresponding to the attribution area of the image collecting equipment for collecting the faces.
4. A method for identifying the face of a sub-library of a multi-million-level library, which is applied to the sub-library face identification system of a multi-million-level library according to any one of claims 1 to 3, and is characterized by comprising:
receiving a face picture acquired by image acquisition equipment, and acquiring information of the image acquisition equipment, an organization mechanism and a region to which the image acquisition equipment belongs;
determining a database corresponding to the attribution area of the image acquisition equipment according to the information of the image acquisition equipment;
identifying the face picture in a acquaintance library, if the face picture is not identified in the acquaintance library, identifying the face picture in a city library corresponding to the image acquisition equipment, and if the face picture is not identified in the city library, identifying the face picture in a silent library which is not acquired by the image acquisition equipment within a preset time;
if the face picture is identified in the acquaintance library, the city library or the silence library, natural person information corresponding to the face picture is obtained;
if the customer corresponding to the face picture is not recognized in the silence library, judging that the customer corresponding to the face picture is a stranger, and sending recognition information corresponding to the face picture to a stranger bottom library;
after natural person information corresponding to the face picture is obtained, updating and inquiring a natural person table, obtaining the last time the natural person is acquired, and judging whether the face picture acquisition is a new visit according to preset logic and the time difference of the two acquired face pictures;
if the face picture is acquired as a new visit, generating a new visit record, and storing the acquired face picture after matching with the visit record;
if the face picture acquisition is not a new visit, inquiring a visit record, and matching the acquired face picture with the latest visit record and warehousing;
and updating the natural person table according to the face picture in the warehouse, identifying the natural person table and pushing the natural person table to a designated party.
5. The method for identifying a sub-library of a multi-tens of thousands of sub-libraries according to claim 4, wherein said determining that the customer corresponding to the face picture is a stranger and transmitting identification information corresponding to the face picture to the stranger sub-library specifically comprises:
after judging that the corresponding customer of the face picture is a stranger, re-identifying the face picture, if the identification is successful, warehousing the face picture to a stranger table, and if the identification is not successful, warehousing the face picture to a stranger bottom library, and then warehousing the face picture to the stranger table;
and updating the information of the new strangers in the warehouse to a record list according to the stranger list, identifying the record list and pushing the record list to a designated party.
6. The method for identifying the sub-library face of the millions of sub-libraries according to claim 5, further comprising:
acquiring information of the record table or the natural person table, and analyzing the identified face picture;
inquiring whether the face picture exists in the corresponding database, if so, acquiring a attribution area corresponding to the database, judging whether the database is matched, and if not, performing database data migration;
if the face picture does not exist in the corresponding database, inquiring a stranger database, and transferring the data in the stranger database to a local market database of a corresponding attribution area.
7. The method for identifying the sub-library face of the millions of sub-libraries according to claim 6, wherein the large data module is combined to analyze and inquire the identified face pictures so as to migrate the sub-libraries to realize data correction.
8. The method for face recognition by a sub-library of a multi-million-level library according to claim 7, wherein the customer data of the face collected in the silent library is migrated to a local market library of the area of the image collection device, the customer data of the local market library, the number of times of the face collected in a preset time reaches a preset number, is migrated to the acquaintance library, and the customer data of the local market library, the number of times of the face collected in a preset time does not reach a preset number, is migrated to the silent library.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010647085.8A CN111950364B (en) | 2020-07-07 | 2020-07-07 | System and method for identifying library-separating face of tens of millions of libraries |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010647085.8A CN111950364B (en) | 2020-07-07 | 2020-07-07 | System and method for identifying library-separating face of tens of millions of libraries |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111950364A CN111950364A (en) | 2020-11-17 |
| CN111950364B true CN111950364B (en) | 2024-03-22 |
Family
ID=73340362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010647085.8A Active CN111950364B (en) | 2020-07-07 | 2020-07-07 | System and method for identifying library-separating face of tens of millions of libraries |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111950364B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113190707B (en) * | 2021-05-24 | 2023-04-07 | 浙江大华技术股份有限公司 | Face library management system, method and device, storage equipment and storage medium |
| CN113590609A (en) * | 2021-06-22 | 2021-11-02 | 北京旷视科技有限公司 | Database partitioning method and device, storage medium and electronic equipment |
| CN114429663B (en) * | 2022-01-28 | 2023-10-20 | 北京百度网讯科技有限公司 | Updating method of face base, face recognition method, device and system |
| CN115187915A (en) * | 2022-09-07 | 2022-10-14 | 苏州万店掌网络科技有限公司 | Passenger flow analysis method, device, equipment and medium |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105574498A (en) * | 2015-12-15 | 2016-05-11 | 重庆凯泽科技有限公司 | Face recognition system and recognition method based on customs security check |
| CN106469296A (en) * | 2016-08-30 | 2017-03-01 | 北京旷视科技有限公司 | Face identification method, device and gate control system |
| CN107067504A (en) * | 2016-12-25 | 2017-08-18 | 北京中海投资管理有限公司 | A kind of recognition of face safety-protection system and a suspect's detection and method for early warning |
| CN107480626A (en) * | 2017-08-09 | 2017-12-15 | 广州云从信息科技有限公司 | A kind of method that census using recognition of face monitors |
| CN107886079A (en) * | 2017-11-22 | 2018-04-06 | 北京旷视科技有限公司 | Object identifying method, apparatus and system |
| CN109214290A (en) * | 2018-08-03 | 2019-01-15 | 深圳市前海圆舟网络科技股份有限公司 | A kind of shops's client management method and device based on recognition of face |
| CN109800691A (en) * | 2019-01-07 | 2019-05-24 | 深圳英飞拓科技股份有限公司 | Demographics method and system based on face recognition technology |
| TWM584937U (en) * | 2019-03-12 | 2019-10-11 | 中光電智能雲服股份有限公司 | Customer identification device |
| CN110334570A (en) * | 2019-03-30 | 2019-10-15 | 深圳市晓舟科技有限公司 | The automatic banking process of recognition of face, device, equipment and storage medium |
| CN110458091A (en) * | 2019-08-08 | 2019-11-15 | 北京阿拉丁智慧科技有限公司 | Recognition of face 1 based on position screening is than N algorithm optimization method |
| CN110473320A (en) * | 2019-08-02 | 2019-11-19 | 浙江天地人科技有限公司 | A kind of judgment method of rental housing personnel management system, permanent personnel and its corresponding door |
| CN110532416A (en) * | 2019-08-30 | 2019-12-03 | 成都智元汇信息技术股份有限公司 | A method of improving recognition of face efficiency and precision |
| CN111079557A (en) * | 2019-11-25 | 2020-04-28 | 山大地纬软件股份有限公司 | Face recognition-based automatic distribution method and system for electric power business hall service terminals |
| CN111126119A (en) * | 2018-11-01 | 2020-05-08 | 百度在线网络技术（北京）有限公司 | Method and device for counting user behaviors arriving at store based on face recognition |
| CN111368622A (en) * | 2019-10-18 | 2020-07-03 | 杭州海康威视系统技术有限公司 | Personnel identification method and device, and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9245276B2 (en) * | 2012-12-12 | 2016-01-26 | Verint Systems Ltd. | Time-in-store estimation using facial recognition |
-
2020
- 2020-07-07 CN CN202010647085.8A patent/CN111950364B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105574498A (en) * | 2015-12-15 | 2016-05-11 | 重庆凯泽科技有限公司 | Face recognition system and recognition method based on customs security check |
| CN106469296A (en) * | 2016-08-30 | 2017-03-01 | 北京旷视科技有限公司 | Face identification method, device and gate control system |
| CN107067504A (en) * | 2016-12-25 | 2017-08-18 | 北京中海投资管理有限公司 | A kind of recognition of face safety-protection system and a suspect's detection and method for early warning |
| CN107480626A (en) * | 2017-08-09 | 2017-12-15 | 广州云从信息科技有限公司 | A kind of method that census using recognition of face monitors |
| CN107886079A (en) * | 2017-11-22 | 2018-04-06 | 北京旷视科技有限公司 | Object identifying method, apparatus and system |
| CN109214290A (en) * | 2018-08-03 | 2019-01-15 | 深圳市前海圆舟网络科技股份有限公司 | A kind of shops's client management method and device based on recognition of face |
| CN111126119A (en) * | 2018-11-01 | 2020-05-08 | 百度在线网络技术（北京）有限公司 | Method and device for counting user behaviors arriving at store based on face recognition |
| CN109800691A (en) * | 2019-01-07 | 2019-05-24 | 深圳英飞拓科技股份有限公司 | Demographics method and system based on face recognition technology |
| TWM584937U (en) * | 2019-03-12 | 2019-10-11 | 中光電智能雲服股份有限公司 | Customer identification device |
| CN110334570A (en) * | 2019-03-30 | 2019-10-15 | 深圳市晓舟科技有限公司 | The automatic banking process of recognition of face, device, equipment and storage medium |
| CN110473320A (en) * | 2019-08-02 | 2019-11-19 | 浙江天地人科技有限公司 | A kind of judgment method of rental housing personnel management system, permanent personnel and its corresponding door |
| CN110458091A (en) * | 2019-08-08 | 2019-11-15 | 北京阿拉丁智慧科技有限公司 | Recognition of face 1 based on position screening is than N algorithm optimization method |
| CN110532416A (en) * | 2019-08-30 | 2019-12-03 | 成都智元汇信息技术股份有限公司 | A method of improving recognition of face efficiency and precision |
| CN111368622A (en) * | 2019-10-18 | 2020-07-03 | 杭州海康威视系统技术有限公司 | Personnel identification method and device, and storage medium |
| CN111079557A (en) * | 2019-11-25 | 2020-04-28 | 山大地纬软件股份有限公司 | Face recognition-based automatic distribution method and system for electric power business hall service terminals |
Non-Patent Citations (2)
| Title |
|---|
| 基于公安大数据的人像比对系统设计;于晓昀 等;《中国安防》;74-78 * |
| 面向人工智能的智慧营业厅系统研究与应用;赵东明 等;《科技成果》;1-4 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111950364A (en) | 2020-11-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111950364B (en) | System and method for identifying library-separating face of tens of millions of libraries | |
| US9535761B2 (en) | Tracking large numbers of moving objects in an event processing system | |
| CN111459985B (en) | Identification information processing method and device | |
| US20150234891A1 (en) | Method and system for providing code scanning result information | |
| CN113392646A (en) | Data relay system, construction method and device | |
| CN1489738A (en) | Storing data based on proximity | |
| CN109858354B (en) | Face identity library, track table establishment method and face track query method and system | |
| CN111209776A (en) | Method, device, processing server, storage medium and system for identifying pedestrians | |
| CN105518644A (en) | Method for processing and displaying real-time social data on map | |
| CN104486585A (en) | Method and system for managing urban mass surveillance video based on GIS | |
| CN114078277A (en) | One-person-one-file face clustering method and device, computer equipment and storage medium | |
| CN111241350B (en) | Graph data query method, device, computer equipment and storage medium | |
| CN110825893A (en) | Target searching method, device, system and storage medium | |
| CN106844553B (en) | Data detection and expansion method and device based on sample data | |
| CN114547386A (en) | Positioning method and device based on Wi-Fi signal and electronic equipment | |
| CN113065016A (en) | Offline store information processing method, device, equipment and system | |
| EP2778966A2 (en) | Systems and methods for point of interest data ingestion | |
| CN110705297A (en) | Enterprise name-identifying method, system, medium and equipment | |
| CN112100452B (en) | Method, apparatus, device and computer readable storage medium for data processing | |
| CN111107493B (en) | Method and system for predicting position of mobile user | |
| CN111400339B (en) | Retrieval method and system for reverse analysis of product database identifier | |
| CN112487082B (en) | Biological feature recognition method and related equipment | |
| CN114781517A (en) | Risk identification method and device and terminal equipment | |
| CN113065893A (en) | Client information identification method, device, equipment and storage medium | |
| Shahid et al. | Images based indoor positioning using AI and crowdsourcing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |