CN111161439A - Attendance application system based on face recognition technology - Google Patents
Attendance application system based on face recognition technology Download PDFInfo
- Publication number
- CN111161439A CN111161439A CN201811327631.9A CN201811327631A CN111161439A CN 111161439 A CN111161439 A CN 111161439A CN 201811327631 A CN201811327631 A CN 201811327631A CN 111161439 A CN111161439 A CN 111161439A
- Authority
- CN
- China
- Prior art keywords
- face
- personnel
- person
- attendance
- face recognition
- 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
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/10—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people together with the recording, indicating or registering of other data, e.g. of signs of identity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- 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
- G06V40/168—Feature extraction; Face representation
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Time Recorders, Dirve Recorders, Access Control (AREA)
- Collating Specific Patterns (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
An attendance application system based on a face recognition technology mainly comprises five parts, namely front-end acquisition equipment, a BS client, a CS client, face recognition service and a face bottom library, and is mainly used for carrying out attendance statistics on enterprise staff and carrying out alarm reminding on the entrance and exit of coming staff. When the staff signs, the face photos of the staff are collected through the camera, characteristic values are obtained from the collected photos through a face recognition algorithm and are analyzed and compared with characteristic values of the face photos of the staff pre-stored in the database, and the names of the staff are reported after the staff are successfully recognized, so that the attendance is successful; if the person is an external person, because the face picture is not stored in the database in advance, the face recognition algorithm can identify the unknown person and send out alarm sound to remind security personnel of removing the core to indicate the information of the person.
Description
Technical Field
The invention belongs to the technical field of face recognition, and particularly relates to an attendance application system based on a face recognition technology, which can be used in the fields of enterprise public safety management, video security management, building security management and the like.
Background
The office building becomes an important component of human life and workplace, and how to utilize scientific and technological means to realize safe modernization, management modernization and intelligent modernization of the office building is urgent. With the continuous excavation of artificial intelligence, the face recognition technology is also widely applied to various security occasions. The face recognition is a biological recognition technology for identity recognition based on face feature information of people, and the uniqueness and the good characteristic of being difficult to copy provide necessary premises for identity recognition.
Disclosure of Invention
The invention aims to apply a face recognition analysis technology to an attendance module, and provides an attendance application system based on the face recognition technology, which aims to improve the enterprise management efficiency and ensure the public safety of enterprises.
The technical scheme of the invention is as follows:
an attendance application system based on a face recognition technology mainly comprises five parts, namely front-end acquisition equipment, a BS client, a CS client, a face recognition service and a face bottom library, wherein the five parts comprise:
(1) front end collection equipment: the front-end acquisition equipment adopts a high-definition camera, mainly captures the human face in real time through real-time code stream, and provides a picture source to be identified for a subsequent process;
(2) the BS client: the BS client is mainly used for authority management of personnel, personnel information management, camera management, identification result query and attendance record query, java technology is adopted as a rear end, a web browser is adopted as a visual display interface, and a mysql database is used as a data storage mode;
(3) the CS client: the CS client is a middle linker of the BS and the face recognition service, winform is adopted as a visual display interface, the BS is communicated with the CS through an http protocol, and the CS is communicated with the face recognition service through an API (application programming interface) of dll; the CS client mainly realizes the functions of face registration starting, registration result storage, video identification and identification result information display and attendance rule calculation;
(4) face recognition service: the face recognition service mainly uses a large number of algorithms and models to detect, align, extract features, compare and analyze a video source and a picture, can convert the picture source into a face base and can also push comparison results to a CS (circuit switching) terminal for processing;
(5) a face base library: the face bottom library is an identified comparison data source, and if the similarity threshold value of the captured face and one data source in the face bottom library reaches above a certain value, the captured face is considered to be the face corresponding to the data source; the human face base library is actually a plurality of dat files which store human face characteristic information, and each person has a dat data source file.
According to the invention, through authority management, personnel types are distinguished, and the safety of an entrance is ensured; through displaying the video, checking the personnel coming in and going out, and ensuring the identification accuracy; the work flow is simplified and the like through attendance management and standardization of an attendance system. The system is mainly used for carrying out attendance statistics on enterprise staff and carrying out alarm reminding on coming and going of personnel. When the employee signs, the face photo of the employee needs to be acquired through the camera, a characteristic value is acquired from the acquired photo through a face recognition algorithm and is analyzed and compared with the characteristic value of the face photo of the employee prestored in the database, and the name of the employee is reported after the employee is successfully recognized, so that the attendance is successful; if the person is an external person, because the face picture is not stored in the database in advance, the face recognition algorithm can identify the unknown person and send out alarm sound to remind security personnel of removing the core to indicate the information of the person.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The system is a comprehensive application system and mainly comprises five parts, namely front-end acquisition equipment, a BS client, a CS client, face recognition service and a face base library. According to five major parts of the system, in specific implementation, three modules, namely front-end acquisition equipment, a workstation with a GPU display card, a BS application end and a CS application end (integrating a face recognition service dll) need to be deployed.
The front-end acquisition equipment is mainly used for installing a high-definition camera, and the installation position needs to be adjusted according to illumination, distance, an image identification range and the like. The workstation has hardware requirements, needs a GPU display card and larger memory and hard disk storage space, and provides an effective operating environment for face recognition service. The BS application end is mainly displayed in a web page form and supported by java service as a background. The CS application terminal is displayed by taking winform as a visual interface and can be executed only by integrating the face recognition service dll.
The operation flow of the invention is shown in figure 1.
(1) Front end collection equipment:
the front-end acquisition equipment adopts a high-definition camera, mainly captures the human face in real time through real-time code streams, and provides a picture source to be identified for the subsequent process. In order to calculate attendance, at least more than one camera installed in one attendance checking place is required.
If one camera is installed at the attendance checking place, the default is the type of the entrance and exit, namely, the people are captured by the camera for face capture and card capture during attendance checking;
if the number of the face capturing cards is two, one default is an entrance type and the other default is an exit type, the entrance camera captures the face to capture the card during working, and the exit camera captures the face to capture the card during working;
if the number of the cameras is three or more, the default is that the cameras at the entrance or the exit snap the face to punch the card during working, and the cameras at the exit or the exit snap the face to punch the card during working.
(2) The BS client:
the BS client is mainly used for authority management of personnel, personnel information management, camera management, identification result query, attendance record query and the like, java technology is adopted as the rear end, a web browser is adopted as a visual display interface, and a mysql database is used as a data storage mode.
The authority management of the personnel is mainly used for distinguishing the enterprise personnel from the external personnel, wherein the external personnel comprise other enterprise personnel, blacklist personnel and the like. If a person is captured by a foreign person, the face is highlighted when being displayed at the CS terminal, and an alarm sound is given out to prompt security personnel to check the information of the person.
The personnel information management is mainly to collect basic information and store information in a database for personnel in the enterprise, and to take provided personnel photos as a picture source for face registration to prepare for forming a face base; after the registration is completed, the registration result of the personnel can be checked.
The camera management mainly comprises the steps of adding, modifying, deleting and the like of a database on the front-section acquisition equipment, and the existence of a real-time code stream is ensured when the CS end starts an identification task.
The identification result query and the attendance record query are mainly used for querying personnel identification results existing in a database and attendance records of personnel of the whole enterprise, and the attendance records can also form a record table to be exported; moreover, the attendance checking personnel can also inquire the attendance checking information of the attendance checking personnel through the system, and the attendance checking system is more convenient and faster than the prior attendance checking through a paper edition or inquiry of personnel.
(3) The CS client:
the CS client is an intermediate link of the BS and the face recognition service, the winform is used as a visual display interface, the BS is communicated with the CS through an http protocol, and the CS is communicated with the face recognition service through an API (application programming interface) of dll. The CS client mainly realizes the functions of face registration starting, registration result storage, video identification and identification result information display, attendance rule calculation and the like.
The BS end sends an http protocol request for personnel registration to the CS end, and after receiving the request, the CS end transmits a personnel picture source address prepared in advance to face recognition service; detecting and extracting features by the face recognition service after a certain period of time, returning a registration result of each picture to the CS terminal, and generating a face base dat file of each person under an appointed path, so that a face base source is formed; after receiving the registration result, the CS terminal inserts the registration result of each picture into a corresponding database table, and generates a corresponding dat file as long as one picture is successfully registered according to an agreed rule 'one picture source', and the person is the face base record, namely the registration is successful; if all picture sources of the person fail to be registered, i.e. it is not possible to generate a corresponding dat file, the person does not exist in the face base, i.e. fails to be registered. After registration is completed, the face recognition service is notified to reload the face base library for the video resource that is performing recognition so that the person can be detected from the face base library.
When the CS terminal is started, the camera resource data of the local computer, which are required to execute the recognition task, are automatically read from the database and are transmitted to the face recognition service through the API for starting, the recognized video source is displayed in a corresponding window of the CS terminal interface after the starting is successful, and if a face exists, the recognized video source is displayed by a frame; if the number of the video sources to be started is larger than the current video window number, the CS end can automatically adjust the window layout to increase the window number to adapt to the video source number; when a person passes through the video source and is identified, the CS terminal can automatically display the information of the person and report the name of the person, and if the person is an external person, the CS terminal can display an unknown person and send an alarm sound; the CS end dynamically adjusts and displays the information of identified personnel, newly identified personnel information is displayed at the top, and the personnel information at the bottom can be extruded due to too many people; meanwhile, the CS terminal inserts the identification result into a database for the query of the BS terminal.
The attendance rules can be set by the BS terminal, the CS terminal is responsible for execution, and attendance information of the previous day is calculated and recorded after zero point every day and is inserted into the database for the BS terminal to inquire and use; in the face records captured by the video resources at the entrance and the exit, the earliest time is taken as the working card punching time of the day; in the face records captured by the video resources at the exit and the entrance, the latest time is used as the off-duty card-punching time of the day; if the on-duty card-punching time is later than the set on-duty time, the time is judged to be late; if the off-duty time is earlier than the set off-duty time, judging that the vehicle is early returned; if the off-duty card punching time is within the set off-duty time range, the off-duty card punching time is judged to be off-duty; during work, the person goes out for a certain time and does not return, and if the time exceeds the set work absence time range, the person is judged to be work absence.
(4) Face recognition service:
the face recognition service mainly uses a large number of algorithms and models to detect, align, extract features, compare and analyze a video source and a picture, can convert the picture source into a face base, and can also push comparison results to a CS (circuit switching) terminal for processing.
At present, an algorithm used by a face recognition service is a deep learning algorithm and is simply divided into three parts of detection, alignment and feature extraction, wherein the MTCNN algorithm is used for detection, the DAN algorithm is used for alignment, and a deep residual error network RES-NET model is used for feature extraction.
(5) A face base library:
and if the similarity threshold of the captured face and a data source in the face bottom library reaches above a certain value, the captured face is considered to be the face corresponding to the data source. The human face base library is actually a plurality of dat files which store human face characteristic information, and each person has a dat data source file.
Claims (5)
1. An attendance application system based on a face recognition technology mainly comprises five parts, namely front-end acquisition equipment, a BS client, a CS client, a face recognition service and a face bottom library, wherein the five parts comprise:
(1) front end collection equipment: the front-end acquisition equipment adopts a high-definition camera, mainly captures the human face in real time through real-time code stream, and provides a picture source to be identified for a subsequent process;
(2) the BS client: the BS client is mainly used for authority management of personnel, personnel information management, camera management, identification result query and attendance record query, java technology is adopted as a rear end, a web browser is adopted as a visual display interface, and a mysql database is used as a data storage mode;
(3) the CS client: the CS client is a middle linker of the BS and the face recognition service, winform is adopted as a visual display interface, the BS is communicated with the CS through an http protocol, and the CS is communicated with the face recognition service through an API (application programming interface) of dll; the CS client mainly realizes the functions of face registration starting, registration result storage, video identification and identification result information display and attendance rule calculation;
(4) face recognition service: the face recognition service mainly uses a large number of algorithms and models to detect, align, extract features, compare and analyze a video source and a picture, can convert the picture source into a face base and can also push comparison results to a CS (circuit switching) terminal for processing;
(5) a face base library: the face bottom library is an identified comparison data source, and if the similarity threshold value of the captured face and one data source in the face bottom library reaches above a certain value, the captured face is considered to be the face corresponding to the data source; the human face base library is actually a plurality of dat files which store human face characteristic information, and each person has a dat data source file.
2. The attendance application system based on the face recognition technology as claimed in claim 1, characterized in that: in order to calculate attendance, at least more than one camera installed in one attendance checking place is required;
if the number of the access points is one, the default is the type of the access point, namely, the people go to work and punch the card by using the camera to shoot the face of the person;
if the number of the face capturing cards is two, one default is an entrance type and the other default is an exit type, the face capturing card is captured by the entrance camera on duty, and the face capturing card is captured by the exit camera on duty;
if the number of the cameras is three or more, the face is captured by the entrance or the entrance camera to punch the card during working, and the face is captured by the exit or the entrance camera to punch the card during working.
3. The attendance application system based on the face recognition technology as claimed in claim 1, characterized in that:
the authority management of the personnel is mainly used for distinguishing the enterprise personnel from external personnel, wherein the external personnel comprise other enterprise personnel and blacklist personnel; if a person is captured by a foreign person, the person is highlighted when being displayed at the CS terminal, and an alarm sound is given out to prompt security personnel to check the information of the person;
the personnel information management is mainly to collect basic information and store an information database for personnel in the enterprise, and to take provided personnel photos as a picture source for face registration to prepare for forming a face base; after the registration is completed, the registration result of the personnel can be checked.
The camera management mainly comprises the steps of adding, modifying, deleting and the like of a database on the front-section acquisition equipment, and the existence of a real-time code stream when the CS end starts an identification task is ensured.
The identification result query and the attendance record query are mainly used for querying personnel identification results existing in a database and attendance records of personnel in the whole enterprise, and the attendance records can also form a record table to be exported; moreover, the attendance checking personnel can also inquire the attendance checking information of the attendance checking personnel through the system, and the attendance checking system is more convenient and faster than the prior attendance checking through a paper edition or inquiry of personnel.
4. The attendance application system based on the face recognition technology as claimed in claim 1, characterized in that: the BS end sends an http protocol request for personnel registration to the CS end, and after receiving the request, the CS end transmits a personnel picture source address prepared in advance to face recognition service; detecting and extracting features by the face recognition service after a certain period of time, returning a registration result of each picture to the CS terminal, and generating a face base dat file of each person under an appointed path, so that a face base source is formed; after receiving the registration result, the CS terminal inserts the registration result of each picture into a corresponding database table, and generates a corresponding dat file as long as one picture is successfully registered according to an agreed rule 'one picture source', and the person is the face base record, namely the registration is successful; if all picture sources of the person fail to be registered, i.e. it is not possible to generate a corresponding dat file, the person does not exist in the face base, i.e. fails to be registered. After registration is finished, the face recognition service is informed to reload a face base to the video resource which is executing recognition so as to detect the person from the face base;
when the CS terminal is started, the camera resource data of the local computer, which are required to execute the recognition task, are automatically read from the database and are transmitted to the face recognition service through the API for starting, the recognized video source is displayed in a corresponding window of the CS terminal interface after the starting is successful, and if a face exists, the recognized video source is displayed by a frame; if the number of the video sources to be started is larger than the current video window number, the CS end can automatically adjust the window layout to increase the window number to adapt to the video source number; when a person passes through the video source and is identified, the CS terminal can automatically display the information of the person and report the name of the person, and if the person is an external person, the CS terminal can display an unknown person and send an alarm sound; the CS end dynamically adjusts and displays the information of identified personnel, newly identified personnel information is displayed at the top, and the personnel information at the bottom can be extruded due to too many people; meanwhile, the CS terminal inserts the identification result into a database for the query of the BS terminal;
the attendance rules can be set by the BS terminal, the CS terminal is responsible for execution, and attendance information of the previous day is calculated and recorded after zero point every day and is inserted into the database for the BS terminal to inquire and use; in the face records captured by the video resources at the entrance and the exit, the earliest time is taken as the working card punching time of the day; in the face records captured by the video resources at the exit and the entrance, the latest time is used as the off-duty card-punching time of the day; if the on-duty card-punching time is later than the set on-duty time, the time is judged to be late; if the off-duty time is earlier than the set off-duty time, judging that the vehicle is early returned; if the off-duty card punching time is within the set off-duty time range, the off-duty card punching time is judged to be off-duty; during work, the person goes out for a certain time and does not return, and if the time exceeds the set work absence time range, the person is judged to be work absence.
5. The attendance application system based on the face recognition technology as claimed in claim 1, characterized in that: the algorithm used by the face recognition service is a deep learning algorithm and is simply divided into three parts of detection, alignment and feature extraction, wherein the MTCNN algorithm is used for detection, the DAN algorithm is used for alignment, and a deep residual error network RES-NET model is used for feature extraction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811327631.9A CN111161439A (en) | 2018-11-08 | 2018-11-08 | Attendance application system based on face recognition technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811327631.9A CN111161439A (en) | 2018-11-08 | 2018-11-08 | Attendance application system based on face recognition technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111161439A true CN111161439A (en) | 2020-05-15 |
Family
ID=70555233
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811327631.9A Pending CN111161439A (en) | 2018-11-08 | 2018-11-08 | Attendance application system based on face recognition technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111161439A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114040163A (en) * | 2021-11-08 | 2022-02-11 | 成都泰盟软件有限公司 | Large-screen display method and system for real-time monitoring of personnel information in laboratory |
-
2018
- 2018-11-08 CN CN201811327631.9A patent/CN111161439A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114040163A (en) * | 2021-11-08 | 2022-02-11 | 成都泰盟软件有限公司 | Large-screen display method and system for real-time monitoring of personnel information in laboratory |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10812761B2 (en) | Complex hardware-based system for video surveillance tracking | |
| CN108052925B (en) | Intelligent management method for community personnel files | |
| WO2019134346A1 (en) | Face recognition method, application server, and computer-readable storage medium | |
| CN106780250B (en) | Intelligent community security event processing method and system based on Internet of things technology | |
| US20210357624A1 (en) | Information processing method and device, and storage medium | |
| CN107920223B (en) | Object behavior detection method and device | |
| CN110290346B (en) | Bidding video acquisition method based on intelligent video analysis | |
| CN215182149U (en) | Infrared temperature measurement system of checking card based on face identification | |
| CN112330834A (en) | Attendance checking method, equipment and storage medium for actively selecting attendance checking event | |
| CN110705472A (en) | Off-duty method and system based on video image recognition | |
| CN111161439A (en) | Attendance application system based on face recognition technology | |
| CN112507314B (en) | Client identity verification method, device, electronic equipment and storage medium | |
| CN111126167B (en) | Method and system for quickly identifying serial activities of multiple specific personnel | |
| CN112686180A (en) | Method for calculating number of personnel in closed space | |
| CN112418063A (en) | Face recognition method and device, electronic equipment and storage medium | |
| CN112218044A (en) | Analysis system and analysis method with security inspection artificial intelligence recognition function | |
| CN112669478A (en) | Attendance system based on face recognition | |
| CN114973135A (en) | Head-shoulder-based sequential video sleep post identification method and system and electronic equipment | |
| CN211087319U (en) | Intelligent card punching machine | |
| WO2021114985A1 (en) | Companionship object identification method and apparatus, server and system | |
| WO2021135933A1 (en) | Target recognition method and device, storage medium and electronic device | |
| CN115810228A (en) | Face recognition access control management method and device, electronic equipment and storage medium | |
| CN113158842A (en) | Identification method, system, device and medium | |
| AU2020202221A1 (en) | Privacy preserving visitor recognition and movement pattern analysis based on computer vision | |
| CN117953557A (en) | Face recognition verification application system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200515 |
|
| WD01 | Invention patent application deemed withdrawn after publication |