WO2019228091A1 - 考勤管理的方法和考勤管理设备 - Google Patents

考勤管理的方法和考勤管理设备 Download PDF

Info

Publication number
WO2019228091A1
WO2019228091A1 PCT/CN2019/083089 CN2019083089W WO2019228091A1 WO 2019228091 A1 WO2019228091 A1 WO 2019228091A1 CN 2019083089 W CN2019083089 W CN 2019083089W WO 2019228091 A1 WO2019228091 A1 WO 2019228091A1
Authority
WO
WIPO (PCT)
Prior art keywords
authenticated user
attendance
portrait image
voiceprint
database
Prior art date
Application number
PCT/CN2019/083089
Other languages
English (en)
French (fr)
Inventor
李婉瑜
陈展
Original Assignee
杭州海康威视数字技术股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 杭州海康威视数字技术股份有限公司 filed Critical 杭州海康威视数字技术股份有限公司
Publication of WO2019228091A1 publication Critical patent/WO2019228091A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/30Authentication, i.e. establishing the identity or authorisation of security principals
    • G06F21/31User authentication
    • G06F21/32User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/10Office automation; Time management
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/10Office automation; Time management
    • G06Q10/105Human resources
    • GPHYSICS
    • G07CHECKING-DEVICES
    • G07CTIME 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
    • G07C1/00Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
    • G07C1/10Registering, 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
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L17/00Speaker identification or verification techniques
    • G10L17/06Decision making techniques; Pattern matching strategies
    • G10L17/08Use of distortion metrics or a particular distance between probe pattern and reference templates
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L17/00Speaker identification or verification techniques
    • G10L17/22Interactive procedures; Man-machine interfaces

Definitions

  • This application relates to the field of electronic technology, and more particularly to a method and device for attendance management.
  • a method for attendance management includes:
  • the attendance record corresponding to the first authenticated user includes current time information and access behavior information of the first authenticated user currently corresponding attendance location, and the access behavior information includes leaving the attendance location or entering the attendance place;
  • an alarm process is performed.
  • the method further includes:
  • the first portrait image is captured by the image shooting device
  • a second authenticated user corresponding to the second portrait image is determined, and the second authenticated user is added to the first database
  • the corresponding attendance record wherein the attendance record corresponding to the second authenticated user includes current time information and the behavior information of the second authenticated user currently corresponding to the attendance place.
  • the method further includes:
  • a voiceprint matching recognition model of the second authenticated user is trained.
  • the method further includes:
  • the method includes:
  • the method further includes:
  • a communication identifier corresponding to a preset alarm object is obtained Information, sending alarm information to a terminal corresponding to the communication identification information.
  • an attendance management device includes:
  • An audio collection device configured to collect the first vocal audio
  • a voiceprint feature extraction module configured to acquire a first voiceprint feature of a first human voice audio collected by the audio collection device
  • An input module configured to input the first voiceprint feature into a pre-trained voiceprint matching recognition model of each authenticated user to obtain a voiceprint matching degree between the first voiceprint feature and each authenticated user;
  • a first determining module configured to determine a first authenticated user corresponding to the target voiceprint matching degree when there is a target voiceprint matching degree greater than a preset threshold in the obtained voiceprint matching degree, and add in the first database with The attendance record corresponding to the first authenticated user, wherein the attendance record corresponding to the first authenticated user includes current time information and the entry and exit behavior information of the first authentication user currently corresponding to the attendance place, the entry and exit behavior information This includes leaving the attendance site or entering the attendance site;
  • An alarm module is configured to perform alarm processing when it is determined that there is no authenticated user in the attendance place within a preset time period based on the attendance record in the first database.
  • the attendance management device further includes:
  • An image capturing device configured to capture a first portrait image when there is no target voiceprint matching degree greater than a preset threshold in the obtained voiceprint matching degree
  • a first search module configured to search, in a second database storing a portrait image of an authenticated user, whether there is a second portrait image matching the first portrait image;
  • a second determining module configured to determine a second authenticated user corresponding to the second portrait image when a second portrait image matching the first portrait image is found, and add the second authenticated user to the first database
  • the attendance record corresponding to the second authenticated user wherein the attendance record corresponding to the second authenticated user includes current time information and the behavior information of the second authenticated user currently corresponding to the attendance place.
  • the attendance management device further includes:
  • a training module is configured to train a voiceprint matching recognition model of the second authenticated user based on the first voiceprint feature.
  • the image capturing device is further configured to capture a third portrait image through an image capturing attendance management device;
  • the attendance management device further includes:
  • a second search module configured to search a fourth portrait image of the first authenticated user in a second database storing a portrait image of the authenticated user
  • a detection module configured to detect the sharpness of the third portrait image and the fourth portrait image separately
  • a replacement module configured to replace the fourth portrait image with the third portrait image in the second database when the definition of the third portrait image is greater than that of the fourth portrait image.
  • the alarm module includes:
  • An obtaining unit configured to obtain communication identification information corresponding to a preset alarm object when it is determined that there is no authenticated user in the attendance place within a preset time period based on the attendance record in the first database;
  • the sending unit is configured to send alarm information to a terminal corresponding to the communication identification information.
  • a third determining module configured to determine a leaveable time period other than the preset time period
  • a sending module configured to obtain a preset alarm when it is determined that the length of time for which there is no authenticated user in the time and attendance place is greater than a preset maximum time based on the attendance record in the first database during the leaveable period
  • the communication identification information corresponding to the object sends alarm information to the terminal corresponding to the communication identification information.
  • a terminal includes a processor and a memory.
  • the memory stores at least one instruction, at least one program, code set, or instruction set.
  • the at least one instruction, The at least one piece of program, the code set, or the instruction set is loaded and executed by the processor to implement the above-mentioned method of attendance management.
  • a computer-readable storage medium stores at least one instruction, at least one piece of program, code set, or instruction set, and the at least one instruction, the at least one piece
  • the program, the code set or the instruction set is loaded and executed by a processor to implement the above-mentioned method of attendance management.
  • the first voiceprint feature of the first human voice audio collected by the audio collection device is obtained; the first voiceprint feature is input into the voiceprint matching recognition model of each authenticated user trained in advance to obtain the first The degree of voiceprint matching with each authenticated user; if there is a target voiceprint matching degree greater than a preset threshold in the obtained voiceprint matching degree, the first authenticated user corresponding to the target voiceprint matching degree is determined.
  • Attendance records corresponding to the first authenticated user are added to a database, wherein the attendance record corresponding to the first authenticated user includes current time information and information on the behavior of the first certified user currently corresponding to attendance, and the information on the behavior includes departure from attendance Place or enter attendance place; when it is determined that there is no authenticated user in the attendance place within the preset time period based on the attendance record in the first database, alarm processing is performed.
  • This application provides an effective mechanism to determine whether there are certified users in the attendance place within a preset period of time, thereby reducing the probability of an accident due to no one in the attendance place.
  • Fig. 1 is a schematic flowchart of a method for attendance management according to an exemplary embodiment
  • Fig. 2 is a schematic flowchart of a method for attendance management according to an exemplary embodiment
  • Fig. 3 is a schematic structural diagram of an attendance management device according to an exemplary embodiment
  • Fig. 4 is a schematic structural diagram of a network-side device according to an exemplary embodiment.
  • the embodiment of the present application provides a method for attendance management, which can be implemented by a network-side device.
  • the network-side device may be a single server, or several different servers.
  • the network-side device may include components such as a transceiver, a processor, and a memory.
  • the transceiver can be used for data transmission with the terminal, or when the network-side device includes several different servers, one of the servers can perform data transmission with the other server through the transceiver.
  • the transceiver may include a Bluetooth component, a WiFi (Wireless-Fidelity) component, an antenna, a matching circuit, a modem, and the like.
  • the processor may be a CPU (Central Processing Unit), etc., and may be used to perform alarm processing when it is determined that there is no authenticated user in the attendance place within a preset time period based on the attendance record in the first database, etc. deal with.
  • Memory which can be RAM (Random Access Memory), Flash (Flash), etc., can be used to store received data, data required for processing, data generated during processing, etc., such as pre-trained Voiceprint matching recognition model for each authenticated user, etc.
  • An exemplary embodiment of the present application provides a method for attendance management. As shown in FIG. 1, a processing flow of the method may include the following steps:
  • Step S110 Obtain a first voiceprint feature of the first human voice audio collected by the audio collection device.
  • the network-side device may receive the first human voice audio collected by an audio collection device such as a microphone, and extract a first voiceprint feature of the first human voice audio.
  • the audio collection device can be installed in the network side equipment, or it can be installed independently around the access control equipment. If the audio collection device is installed in the network-side device, the network-side device can be set near the access control device, so that the network-side device can collect the first human voice through the installed audio collection device. If the network-side device includes several different servers, one of the servers can be set as a server dedicated to processing voiceprints, which can be referred to as a voiceprint recognition server.
  • step S120 the first voiceprint feature is input into a voiceprint matching recognition model of each authenticated user trained in advance to obtain a voiceprint matching degree between the first voiceprint feature and each authenticated user.
  • the network-side device may input the first voiceprint feature into a pre-trained voiceprint matching recognition model of each authenticated user to obtain a voiceprint matching degree between the first voiceprint feature and each authenticated user.
  • the voiceprint recognition server may input the first voiceprint feature into a voiceprint matching recognition model of each authenticated user trained in advance to obtain the voiceprint matching degree between the first voiceprint feature and each authenticated user.
  • the voiceprint matching degree indicates the similarity between the first voiceprint feature and the voiceprint feature of the authenticated user corresponding to the input voiceprint matching recognition model. When the similarity between them is higher, the user who proves to have the first voiceprint feature and the above-mentioned authenticated user are more likely to be the same person.
  • the similarity between them exceeds a preset threshold, it is proved that the first voiceprint feature is very similar to the voiceprint feature of the above-mentioned authenticated user, and it can be determined that the user having the first voiceprint feature and the above-mentioned authenticated user are the same person. In this way, the identity of the user having the first voiceprint characteristic can be identified.
  • the voiceprint matching recognition model of each authenticated user is pre-trained. Before training, a voiceprint database can be established to store a large number of human voices of different authenticated users in the voiceprint database. After that, a voiceprint matching recognition model can be initially established, and the voiceprint features of human voice audio of different authenticated users can be extracted. Based on the extracted voiceprint features, the initially established voiceprint matching recognition model is trained to obtain a voiceprint matching recognition model for each authenticated user. It should be noted that, for each authenticated user, a corresponding voiceprint matching recognition model can be established respectively.
  • the voiceprint matching recognition model of each authenticated user After training the voiceprint matching recognition model of each authenticated user, when the first voiceprint feature is input into the voiceprint matching recognition model of each authenticated user, the voiceprint matching recognition model of each authenticated user can output the first A voiceprint feature matches the voiceprint of each authenticated user.
  • the matching degree can be a percentage. For example, there are 100 voiceprint matching recognition models. The first voiceprint feature is input into the 100 voiceprint matching recognition models, and 99 of the models have a matching degree of 3% -10. %, The matching degree of 1 model output is 95%. The authenticated user corresponding to the voiceprint matching recognition model with an output matching degree of 95% is likely to belong to the same user as the user having the first voiceprint feature.
  • Step S130 if there is a target voiceprint matching degree greater than a preset threshold in the obtained voiceprint matching degree, determine a first authenticated user corresponding to the target voiceprint matching degree, and add a corresponding corresponding to the first authenticated user to the first database.
  • Attendance Record if there is a target voiceprint matching degree greater than a preset threshold in the obtained voiceprint matching degree, determine a first authenticated user corresponding to the target voiceprint matching degree, and add a corresponding corresponding to the first authenticated user to the first database.
  • the attendance record corresponding to the first authenticated user includes current time information and entry and exit behavior information of the first authenticated user corresponding to the attendance place, and the exit and entry behavior information includes leaving the attendance place or entering the attendance place.
  • the preset threshold is set to 90%
  • 99 of the models have a matching degree of 3% -10%
  • 1 The matching degree of the model output is 95%
  • the matching degree of 95% is greater than the preset threshold, and it is confirmed that there is a target voiceprint matching degree greater than the preset threshold in the obtained voiceprint matching degree.
  • each authenticated user can correspond to one voiceprint matching recognition model, when the output matching degree of one voiceprint matching recognition model among all the voiceprint matching recognition models is greater than a preset threshold, it can be determined that the output matching degree is greater than a preset
  • the authentication user corresponding to the threshold model is the first authentication user. At this time, it may be determined that the vocalizer of the first human voice audio collected by the audio collection device is the first authenticated user. Attendance records corresponding to the first authenticated user may be added to the first database.
  • the voiceprint recognition server recognizes that the vocalist of the first human voice is the first authenticated user, the identification of the first authenticated user, current time information, and access The behavior information is sent to the attendance server together, and the attendance server adds an attendance record corresponding to the first authenticated user in the first database.
  • Two microphones can be installed on the inside and outside of the access control device. When the first vocal audio is collected through the inside microphone, it can be determined that the first authenticated user has left the attendance site, and when the first microphone is collected through the outside microphone, When the human voice is audio, it can be determined that the first authenticated user is entering the attendance place.
  • the current attendance record corresponding to the first authenticated user may be obtained. If the current attendance record is to leave the attendance place, the current attendance record is updated to enter the attendance place. If the current attendance record is to enter the attendance place, the current attendance record is updated to leave the attendance place.
  • the voiceprint recognition server can also send an unlocking instruction to the electric lock control terminal to control the electric lock to perform the unlocking operation, thereby opening the access control device.
  • the voiceprint recognition server may also send an unlocking instruction to the remote monitoring server, and the remote monitoring server forwards the unlocking instruction to the electric lock control terminal to control the electric lock to perform the unlocking operation, thereby opening the access control device.
  • the first portrait image is captured by the image capturing device; in the second database storing the portrait image of the authenticated user To find whether there is a second portrait image matching the first portrait image; if a second portrait image matching the first portrait image is found, determine a second authenticated user corresponding to the second portrait image, in Attendance records corresponding to the second authenticated user are added to the first database, where the attendance record corresponding to the second authenticated user includes current time information and information on the behavior of the second authenticated user's current attendance location.
  • the first portrait image can be captured by an image capturing device such as a camera.
  • a face image of a user standing near the access control device may exist in the first portrait image. Face images can be identified to determine if the user standing at the access control device is an authenticated user.
  • the first portrait image can be captured by the image capturing device.
  • the second database storing the portrait image of the authenticated user, it is found whether there is a second portrait image matching the first portrait image.
  • a second authenticated user corresponding to the second portrait image is determined, and an attendance record corresponding to the second authenticated user is added to the first database.
  • the above operations may be performed on a network-side device or a face recognition server in the network-side device.
  • a voiceprint matching recognition model of a second authenticated user is trained based on the first voiceprint feature.
  • the voiceprint matching recognition model of the second authenticated user may be updated. .
  • the voiceprint matching recognition model of the second authenticated user may be trained based on the newly extracted first voiceprint features.
  • the method provided in this embodiment further includes: taking a third portrait image by using an image capturing device; and storing a second database of portrait images of the authenticated user To find the fourth portrait image of the first authenticated user; detect the sharpness of the third portrait image and the fourth portrait image respectively; if the third portrait image is sharper than the fourth portrait image, the second database is , The fourth portrait image is replaced with the third portrait image.
  • the third portrait image may be captured by the image capturing device.
  • the third portrait image includes a face image of the first authenticated user.
  • a fourth portrait image of the first authenticated user may be found in the second database, the third portrait image and the fourth portrait image may be compared, and an image with higher quality may be selected and stored in the second database.
  • the sharpness of the third portrait image and the fourth portrait image can be detected. If the sharpness of the third portrait image is greater than the sharpness of the fourth portrait image, then in the second database, the fourth portrait image is replaced with the third portrait image. image.
  • step S140 when it is determined that there is no authenticated user in the attendance place within the preset time period based on the attendance record in the first database, an alarm process is performed.
  • the network-side device may perform alarm processing when it is determined that there is no authenticated user in the attendance place within a preset time period based on the attendance record in the first database. Or, if the network-side device includes several different servers, set up a separate time attendance server to perform the above operations.
  • step S140 may include: when it is determined based on the attendance record in the first database that there is no authenticated user in the attendance place within a preset time period, obtaining communication identification information corresponding to the preset alarm object; The corresponding terminal sends an alarm message.
  • a leaveable time period other than the preset time period such as between 12 noon and 1 o'clock.
  • the leaveable time period is reached, based on the attendance records in the first database, it can be determined that the length of time for which there are no authenticated users in the attendance place, and the time starts when all authenticated users leave the time and attendance place. If the time exceeds the preset maximum time, there is no authentication
  • the user enters the attendance place he can obtain the preset communication identification information corresponding to the alarm object, and send the alarm information to the terminal corresponding to the communication identification information. If there is an authenticated user entering the attendance site when the timing does not exceed the preset maximum time, you can stop and clear the timing.
  • an on-duty person can be arranged to be on duty for two days and two nights at a time, and then other personnel will replace the on-duty person who has been on duty for two days and two nights.
  • This cycle achieves the purpose of having on-duty personnel on duty at all times in the high-voltage substation.
  • the communication identification information corresponding to the preset alarm object can be obtained, such as the phone number of the manager who should be on duty at present Number to send alarm information to the terminal corresponding to the communication identification information.
  • the manager of the currently on duty staff can receive the alarm message and learn that the currently on duty staff is not on duty.
  • the manager of the person currently on duty may dispatch the person currently on duty to the post immediately, or send other personnel to the post immediately.
  • the authenticated user who entered the attendance site entered the attendance site for the first time that day.
  • the network-side device includes several servers, as shown in Figure 2, when a user comes near the access control device, words such as "open the door” and "Suzhou substation” can be said. Then, the audio collection device near the access control device can collect the first human voice and send the first human voice to the voiceprint recognition server.
  • the voiceprint recognition server can identify the first human voice and determine whether the user is authenticated. User, and which authenticated user the user is.
  • the voiceprint recognition server fails to recognize the audio, the voiceprint recognition server sends an image capture instruction to the image capture device near the access control device.
  • the image capture device can capture the first portrait image and send the first portrait image to the face recognition server. .
  • the face recognition server can recognize the first portrait image, determine whether the user is an authenticated user, and which authenticated user the user is. After the voiceprint recognition server or the face recognition server has passed the recognition, the identification of the first authenticated user, current time information, and access behavior information, etc. may be sent to the time and attendance server, and an unlocking instruction may also be sent to the remote monitoring server. After the remote monitoring server receives the unlocking instruction, it can forward the unlocking instruction to the electric lock control end. After receiving the unlock command, the electric lock control terminal performs the unlock operation.
  • the first voiceprint feature of the first human voice audio collected by the audio collection device is obtained; the first voiceprint feature is input into the voiceprint matching recognition model of each authenticated user trained in advance to obtain The degree of voiceprint matching between the first voiceprint feature and each authenticated user; if there is a target voiceprint matching degree greater than a preset threshold in the obtained voiceprint matching degree, determining a first authenticated user corresponding to the target voiceprint matching degree, Attendance records corresponding to the first authenticated user are added to the first database, where the attendance record corresponding to the first authenticated user includes current time information and the entry and exit behavior information of the first certified user's current attendance location, the entry and exit behavior information includes Leaving the attendance site or entering the attendance site; when it is determined that there is no authenticated user in the attendance site within a preset time period based on the attendance record in the first database, an alarm process is performed.
  • the attendance management device includes:
  • An audio collection device 310 configured to collect first vocal audio
  • a voiceprint feature extraction module 320 configured to acquire a first voiceprint feature of a first human voice audio collected by the audio collection device
  • An input module 330 is configured to input the first voiceprint feature into a pre-trained voiceprint matching recognition model of each authenticated user to obtain a voiceprint matching degree between the first voiceprint feature and each authenticated user;
  • the first determining module 340 is configured to determine a first authenticated user corresponding to the target voiceprint matching degree when there is a target voiceprint matching degree that is greater than a preset threshold in the obtained voiceprint matching degree, and add it to the first database.
  • Attendance record corresponding to the first authenticated user wherein the attendance record corresponding to the first authenticated user includes current time information and access behavior information of the first certified user's current attendance location, the access behavior The information includes leaving the attendance site or entering the attendance site;
  • the alarm module 350 is configured to perform alarm processing when it is determined that there is no authenticated user in the attendance place within a preset time period based on the attendance record in the first database.
  • the attendance management device further includes:
  • An image capturing device configured to capture a first portrait image when there is no target voiceprint matching degree greater than a preset threshold in the obtained voiceprint matching degree
  • a first search module configured to search, in a second database storing a portrait image of an authenticated user, whether there is a second portrait image matching the first portrait image;
  • a second determining module configured to determine a second authenticated user corresponding to the second portrait image when a second portrait image matching the first portrait image is found, and add the second authenticated user to the first database
  • the attendance record corresponding to the second authenticated user wherein the attendance record corresponding to the second authenticated user includes current time information and the behavior information of the second authenticated user currently corresponding to the attendance place.
  • the attendance management device further includes:
  • a training module is configured to train a voiceprint matching recognition model of the second authenticated user based on the first voiceprint feature.
  • the image capturing device is further configured to capture a third portrait image through an image capturing attendance management device;
  • the attendance management device further includes:
  • a second search module configured to search a fourth portrait image of the first authenticated user in a second database storing a portrait image of the authenticated user
  • a detection module configured to detect the sharpness of the third portrait image and the fourth portrait image separately
  • a replacement module configured to replace the fourth portrait image with the third portrait image in the second database when the definition of the third portrait image is greater than that of the fourth portrait image.
  • the alarm module includes:
  • An obtaining unit configured to obtain communication identification information corresponding to a preset alarm object when it is determined that there is no authenticated user in the attendance place within a preset time period based on the attendance record in the first database;
  • the sending unit is configured to send alarm information to a terminal corresponding to the communication identification information.
  • the attendance management device further includes:
  • a third determining module configured to determine a leaveable time period other than the preset time period
  • a sending module configured to obtain a preset alarm when it is determined that the length of time for which there is no authenticated user in the time and attendance place is greater than a preset maximum time based on the attendance record in the first database during the leaveable period of time
  • the communication identification information corresponding to the object sends alarm information to the terminal corresponding to the communication identification information.
  • This application provides an effective mechanism to determine whether there are certified users in the attendance place within a preset period of time, thereby reducing the probability of an accident due to no one in the attendance place.
  • the attendance management device only uses the above-mentioned division of function modules as an example to describe the time and attendance management.
  • the above functions may be allocated by different function modules as required. That is, the internal structure of the network-side device is divided into different functional modules, or multiple servers are set up to complete all or part of the functions described above.
  • the device for attendance management provided in the foregoing embodiment belongs to the same concept as the method embodiment for attendance management, and the specific implementation process thereof is described in the method embodiment, which will not be repeated here.
  • FIG. 4 is a schematic structural diagram of a network-side device 1900 according to an exemplary embodiment of the present invention.
  • the network-side device 1900 may have a relatively large difference due to different configurations or performance, and may include one or more processors (central processing units) 1910 and one or more memories 1920.
  • the memory 1920 stores at least one instruction, and the at least one instruction is loaded and executed by the processor 1910 to implement the method for attendance management described in the foregoing embodiment.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Multimedia (AREA)
  • Health & Medical Sciences (AREA)
  • Acoustics & Sound (AREA)
  • Human Computer Interaction (AREA)
  • Theoretical Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Strategic Management (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Tourism & Hospitality (AREA)
  • Computer Security & Cryptography (AREA)
  • Quality & Reliability (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Economics (AREA)
  • Data Mining & Analysis (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种考勤管理的方法和考勤管理设备,属于电子技术领域,该方法包括:获取音频采集装置采集的第一人声音频的第一声纹特征(S110),将第一声纹特征,输入预先训练的每个认证用户的声纹匹配识别模型中,得到第一声纹特征与每个认证用户的声纹匹配度(S120);如果得到的声纹匹配度中存在大于预设阈值的目标声纹匹配度,则确定目标声纹匹配度对应的第一认证用户,在第一数据库中添加与第一认证用户对应的考勤记录;当基于第一数据库中的考勤记录确定在预设时间段内考勤场所中不存在认证用户时,进行报警处理(S140);其可以降低因为考勤场所中无人在岗而出现事故的概率。

Description

考勤管理的方法和考勤管理设备
本申请要求于2018年06月01日提交的申请号为201810556426.3、发明名称为“考勤管理的方法和考勤管理设备”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请是关于电子技术领域,尤其是关于一种考勤管理的方法和考勤管理设备。
背景技术
在某些重点安防场所中,经常需要值班人员24小时进行值守。如果没有值班人员值班,则当有紧急情况发生时,容易出现事故。
在实现本申请的过程中,发明人发现至少存在以下问题:
没有有效的机制判断重点安防场所中是否一直都有值班人员在值班,容易提高事故出现的频率。
发明内容
为了克服相关技术中存在的问题,本申请提供了以下技术方案:
根据本申请实施例的第一方面,提供一种考勤管理的方法,所述方法包括:
获取音频采集装置采集的第一人声音频的第一声纹特征;
将所述第一声纹特征,输入预先训练的每个认证用户的声纹匹配识别模型中,得到所述第一声纹特征与每个认证用户的声纹匹配度;
如果得到的声纹匹配度中存在大于预设阈值的目标声纹匹配度,则确定所述目标声纹匹配度对应的第一认证用户,在第一数据库中添加与所述第一认证用户对应的考勤记录,其中,与所述第一认证用户对应的考勤记录包括当前的时间信息和所述第一认证用户当前对应考勤场所的出入行为信息,所述出入行为信息包括离开考勤场所或进入考勤场所;
当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中 不存在认证用户时,进行报警处理。
可选地,所述方法还包括:
如果得到的声纹匹配度中不存在大于预设阈值的目标声纹匹配度,则通过图像拍摄装置拍摄第一人像图像;
在存储有认证用户的人像图像的第二数据库中,查找是否有与所述第一人像图像相匹配的第二人像图像;
如果查找到与所述第一人像图像相匹配的第二人像图像,则确定与所述第二人像图像对应的第二认证用户,在所述第一数据库中添加与所述第二认证用户对应的考勤记录,其中,与所述第二认证用户对应的考勤记录包括当前的时间信息和所述第二认证用户当前对应考勤场所的出入行为信息。
可选地,所述方法还包括:
基于所述第一声纹特征,训练所述第二认证用户的声纹匹配识别模型。
可选地,在确定所述目标声纹匹配度对应的第一认证用户之后,所述方法还包括:
通过图像拍摄装置拍摄第三人像图像;
在存储有认证用户的人像图像的第二数据库中,查找所述第一认证用户的第四人像图像;
分别检测所述第三人像图像和所述第四人像图像的清晰度;
如果所述第三人像图像的清晰度大于所述第四人像图像的清晰度,则在所述第二数据库中,用所述第三人像图像替换所述第四人像图像。
可选地,所述当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中不存在认证用户时,进行报警处理,包括:
当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中不存在认证用户时,获取预设的报警对象对应的通信标识信息;
向所述通信标识信息对应的终端发送报警信息。
可选地,所述方法还包括:
确定除所述预设时间段之外的可离岗时间段;
如果在所述可离岗时间段内,基于所述第一数据库中的考勤记录确定所述考勤场所中不存在认证用户的时长大于预设最大时长,则获取预设的报警对象对应的通信标识信息,向所述通信标识信息对应的终端发送报警信息。
根据本申请实施例的第二方面,提供一种考勤管理设备,所述考勤管理设备包括:
音频采集装置,用于采集第一人声音频;
声纹特征提取模块,用于获取所述音频采集装置采集的第一人声音频的第一声纹特征;
输入模块,用于将所述第一声纹特征,输入预先训练的每个认证用户的声纹匹配识别模型中,得到所述第一声纹特征与每个认证用户的声纹匹配度;
第一确定模块,用于当得到的声纹匹配度中存在大于预设阈值的目标声纹匹配度时,确定所述目标声纹匹配度对应的第一认证用户,在第一数据库中添加与所述第一认证用户对应的考勤记录,其中,与所述第一认证用户对应的考勤记录包括当前的时间信息和所述第一认证用户当前对应考勤场所的出入行为信息,所述出入行为信息包括离开考勤场所或进入考勤场所;
报警模块,用于当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中不存在认证用户时,进行报警处理。
可选地,所述考勤管理设备还包括:
图像拍摄装置,用于当得到的声纹匹配度中不存在大于预设阈值的目标声纹匹配度时,拍摄第一人像图像;
第一查找模块,用于在存储有认证用户的人像图像的第二数据库中,查找是否有与所述第一人像图像相匹配的第二人像图像;
第二确定模块,用于当查找到与所述第一人像图像相匹配的第二人像图像时,确定与所述第二人像图像对应的第二认证用户,在所述第一数据库中添加与所述第二认证用户对应的考勤记录,其中,与所述第二认证用户对应的考勤记录包括当前的时间信息和所述第二认证用户当前对应考勤场所的出入行为信息。
可选地,所述考勤管理设备还包括:
训练模块,用于基于所述第一声纹特征,训练所述第二认证用户的声纹匹配识别模型。
可选地,所述图像拍摄装置,还用于通过图像拍摄考勤管理设备拍摄第三人像图像;
所述考勤管理设备还包括:
第二查找模块,用于在存储有认证用户的人像图像的第二数据库中,查找所述第一认证用户的第四人像图像;
检测模块,用于分别检测所述第三人像图像和所述第四人像图像的清晰度;
替换模块,用于当所述第三人像图像的清晰度大于所述第四人像图像的清晰度时,在所述第二数据库中,用所述第三人像图像替换所述第四人像图像。
可选地,所述报警模块包括:
获取单元,用于当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中不存在认证用户时,获取预设的报警对象对应的通信标识信息;
发送单元,用于向所述通信标识信息对应的终端发送报警信息。
可选地,所述考勤管理设备还包括:
第三确定模块,用于确定除所述预设时间段之外的可离岗时间段;
发送模块,用于当在所述可离岗时间段内,基于所述第一数据库中的考勤记录确定所述考勤场所中不存在认证用户的时长大于预设最大时长时,获取预设的报警对象对应的通信标识信息,向所述通信标识信息对应的终端发送报警信息。
根据本申请实施例的第三方面,提供一种终端,所述终端包括处理器和存储器,所述存储器中存储有至少一条指令、至少一段程序、代码集或指令集,所述至少一条指令、所述至少一段程序、所述代码集或指令集由所述处理器加载并执行以实现上述考勤管理的方法。
根据本申请实施例的第四方面,提供一种计算机可读存储介质,所述存储介质中存储有至少一条指令、至少一段程序、代码集或指令集,所述至少一条指令、所述至少一段程序、所述代码集或指令集由处理器加载并执行以实现上述考勤管理的方法。
本申请的实施例提供的技术方案可以包括以下有益效果:
本申请实施例中,获取音频采集装置采集的第一人声音频的第一声纹特征;将第一声纹特征,输入预先训练的每个认证用户的声纹匹配识别模型中,得到第一声纹特征与每个认证用户的声纹匹配度;如果得到的声纹匹配度中存在大于预设阈值的目标声纹匹配度,则确定目标声纹匹配度对应的第一认证用户,在第一数据库中添加与第一认证用户对应的考勤记录,其中,与第一认证用户对应的考勤记录包括当前的时间信息和第一认证用户当前对应考勤场所的出入 行为信息,出入行为信息包括离开考勤场所或进入考勤场所;当基于第一数据库中的考勤记录确定在预设时间段内考勤场所中不存在认证用户时,进行报警处理。本申请提供了一种有效机制来判断考勤场所中是否在预设时间段内都有认证用户在岗,进而可以降低因为考勤场所中无人在岗而出现事故的概率。
应当理解的是,以上的一般描述和后文的细节描述仅是示例性和解释性的,并不能限制本申请。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分,示出了符合本申请的实施例,并与说明书一起用于解释本申请的原理。在附图中:
图1是根据一示例性实施例示出的一种考勤管理的方法的流程图示意图;
图2是根据一示例性实施例示出的一种考勤管理的方法的流程图示意图;
图3是根据一示例性实施例示出的一种考勤管理设备的结构示意图;
图4是根据一示例性实施例示出的一种网络侧设备的结构示意图。
通过上述附图,已示出本申请明确的实施例,后文中将有更详细的描述。这些附图和文字描述并不是为了通过任何方式限制本申请构思的范围,而是通过参考特定实施例为本领域技术人员说明本申请的概念。
具体实施方式
这里将详细地对示例性实施例进行说明,其示例表示在附图中。下面的描述涉及附图时,除非另有表示,不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本申请相一致的所有实施方式。相反,它们仅是与如所附权利要求书中所详述的、本申请的一些方面相一致的装置和方法的例子。
本申请实施例提供了一种考勤管理的方法,该方法可以由网络侧设备实现。其中,网络侧设备可以是单独的一台服务器,或者是几台不同的服务器。
网络侧设备可以包括收发器、处理器、存储器等部件。收发器,可以用于与终端进行数据传输,或者在网络侧设备包括几台不同的服务器的情况下,其中某一台服务器可以通过收发器与其他服务器进行数据传输。收发器可以包括蓝牙部件、WiFi(Wireless-Fidelity,无线高保真技术)部件、天线、匹配电路、 调制解调器等。处理器,可以为CPU(Central Processing Unit,中央处理单元)等,可以用于当基于第一数据库中的考勤记录确定在预设时间段内考勤场所中不存在认证用户时,进行报警处理,等处理。存储器,可以为RAM(Random Access Memory,随机存取存储器),Flash(闪存)等,可以用于存储接收到的数据、处理过程所需的数据、处理过程中生成的数据等,如预先训练的每个认证用户的声纹匹配识别模型等。
本申请一示例性实施例提供了一种考勤管理的方法,如图1所示,该方法的处理流程可以包括如下的步骤:
步骤S110,获取音频采集装置采集的第一人声音频的第一声纹特征。
在实施中,网络侧设备可以接收音频采集装置如麦克风采集的第一人声音频,并提取第一人声音频的第一声纹特征。其中,音频采集装置可以安装在网络侧设备中,也可以独立安装在门禁设备周围。如果音频采集装置安装在网络侧设备中,网络侧设备可以设置在门禁设备附近,这样网络侧设备可以通过安装的音频采集装置采集第一人声音频。如果网络侧设备包括几台不同的服务器,其中一台服务器可以设置为专门处理声纹的服务器,可以将其称作声纹识别服务器。
步骤S120,将第一声纹特征,输入预先训练的每个认证用户的声纹匹配识别模型中,得到第一声纹特征与每个认证用户的声纹匹配度。
在实施中,网络侧设备可以将第一声纹特征,输入预先训练的每个认证用户的声纹匹配识别模型中,得到第一声纹特征与每个认证用户的声纹匹配度。或者,声纹识别服务器可以将第一声纹特征,输入预先训练的每个认证用户的声纹匹配识别模型中,得到第一声纹特征与每个认证用户的声纹匹配度。声纹匹配度表示第一声纹特征和输入的声纹匹配识别模型对应的认证用户的声纹特征之间的相似度。当它们之间的相似度越高时,证明拥有第一声纹特征的用户和上述认证用户越可能是同一人。当它们之间的相似度超过某一预设阈值时,证明第一声纹特征和上述认证用户的声纹特征极为相似,可以确定拥有第一声纹特征的用户和上述认证用户就是同一人,这样就可以识别拥有第一声纹特征的用户的身份。
每个认证用户的声纹匹配识别模型是预先训练的。在训练之前,可以先建 立声纹数据库,在声纹数据库中存储大量的不同认证用户的人声音频。之后,可以初步建立声纹匹配识别模型,并提取不同认证用户的人声音频的声纹特征。基于提取的声纹特征,对初步建立的声纹匹配识别模型进行训练,得到每个认证用户的声纹匹配识别模型。需要说明的是,对于每个认证用户,可以分别建立一个与其对应的声纹匹配识别模型。
在训练完毕每个认证用户的声纹匹配识别模型之后,将第一声纹特征输入到每个认证用户的声纹匹配识别模型中时,每个认证用户的声纹匹配识别模型中可以输出第一声纹特征与每个认证用户的声纹匹配度。该匹配度可以是百分比,例如,共有100个声纹匹配识别模型,将第一声纹特征输入到这100个声纹匹配识别模型中,其中有99个模型输出的匹配度是3%-10%,有1个模型输出的匹配度是95%。输出匹配度是95%的声纹匹配识别模型对应的认证用户很可能和拥有第一声纹特征的用户属于同一用户。
步骤S130,如果得到的声纹匹配度中存在大于预设阈值的目标声纹匹配度,则确定目标声纹匹配度对应的第一认证用户,在第一数据库中添加与第一认证用户对应的考勤记录。
其中,与第一认证用户对应的考勤记录包括当前的时间信息和第一认证用户当前对应考勤场所的出入行为信息,出入行为信息包括离开考勤场所或进入考勤场所。
在实施中,假如设置预设阈值为90%,在将第一声纹特征输入到100个声纹匹配识别模型时,其中有99个模型输出的匹配度是3%-10%,有1个模型输出的匹配度是95%,则95%的匹配度大于预设阈值,确认得到的声纹匹配度中存在大于预设阈值的目标声纹匹配度。
由于每个认证用户都可以对应一个的声纹匹配识别模型,因此当所有声纹匹配识别模型中的一个声纹匹配识别模型输出的匹配度大于预设阈值时,可以确定输出匹配度大于预设阈值的模型对应的认证用户为第一认证用户。此时,可以确定音频采集装置采集到的第一人声音频的发声者为第一认证用户。可以在第一数据库中添加与第一认证用户对应的考勤记录。或者,当网络侧设备包括几台不同的服务器时,在声纹识别服务器识别出第一人声音频的发声者为第一认证用户时,将第一认证用户的标识、当前的时间信息、出入行为信息一并发送至考勤服务器,由考勤服务器在第一数据库中添加与第一认证用户对应的 考勤记录。其中,可以在门禁设备的内侧和外侧分别安装两个麦克风,当通过内侧的麦克风采集到第一人声音频时,可以确定第一认证用户是离开考勤场所,当通过外侧的麦克风采集到第一人声音频时,可以确定第一认证用户是进入考勤场所。
或者,在识别出第一认证用户之后,可以获取第一认证用户对应的当前的考勤记录,如果当前的考勤记录为离开考勤场所,则将当前的考勤记录更新为进入考勤场所。如果当前的考勤记录为进入考勤场所,则将当前的考勤记录更新为离开考勤场所。
此外,如果在门禁设备中设置了电锁,声纹识别服务器还可以向电锁控制端发送开锁指令,以控制电锁进行开锁操作,从而打开门禁设备。再或者,声纹识别服务器还可以向远程监控服务器发送开锁指令,远程监控服务器向电锁控制端转发开锁指令,以控制电锁进行开锁操作,从而打开门禁设备。
可选地,如果得到的声纹匹配度中不存在大于预设阈值的目标声纹匹配度,则通过图像拍摄装置拍摄第一人像图像;在存储有认证用户的人像图像的第二数据库中,查找是否有与第一人像图像相匹配的第二人像图像;如果查找到与第一人像图像相匹配的第二人像图像,则确定与第二人像图像对应的第二认证用户,在第一数据库中添加与第二认证用户对应的考勤记录,其中,与第二认证用户对应的考勤记录包括当前的时间信息和第二认证用户当前对应考勤场所的出入行为信息。
在实施中,当所有声纹匹配识别模型输出的匹配度都不大于预设阈值时,可以认为声纹识别失败,此时可以通过图像拍摄装置如摄像头,拍摄第一人像图像。第一人像图像中可以存在站在门禁设备附近的用户的人脸图像。可以对人脸图像进行识别,确定站在门禁设备附件的用户是不是认证用户。具体地,可以通过图像拍摄装置拍摄第一人像图像,在存储有认证用户的人像图像的第二数据库中,查找是否有与第一人像图像相匹配的第二人像图像,如果查找到与第一人像图像相匹配的第二人像图像,则确定与第二人像图像对应的第二认证用户,在第一数据库中添加与第二认证用户对应的考勤记录。上述操作可以在网络侧设备执行,也可以在网络侧设备中的人脸识别服务器中执行。
可选地,基于第一声纹特征,训练第二认证用户的声纹匹配识别模型。
在实施中,可以当声纹识别失败,但是人脸识别成功时,可以认为第二认 证用户的声纹匹配识别模型的识别结果不准确,此时可以更新第二认证用户的声纹匹配识别模型。具体地,可以基于新提取的第一声纹特征,训练第二认证用户的声纹匹配识别模型。
可选地,在确定目标声纹匹配度对应的第一认证用户之后,本实施例提供的方法还包括:通过图像拍摄装置拍摄第三人像图像;在存储有认证用户的人像图像的第二数据库中,查找第一认证用户的第四人像图像;分别检测第三人像图像和第四人像图像的清晰度;如果第三人像图像的清晰度大于第四人像图像的清晰度,则在第二数据库中,用第三人像图像替换第四人像图像。
在实施中,由于当声纹识别失败时,可以采用人脸识别进行验证。例如,当用户感冒、前一天熬夜等情况发生时,有可能声纹匹配识别模型的识别结果不准确。此时,可以采用人脸识别进行验证。为了使得人脸识别的结识别果更准确,可以选一些质量较高的图像存入第二数据库。具体地,在声纹识别成功之后,可以通过图像拍摄装置拍摄第三人像图像。第三人像图像中包括第一认证用户的人脸图像。可以在第二数据库中,查找第一认证用户的第四人像图像,比对第三人像图像和第四人像图像,选择质量较高的图像存储在第二数据库中。例如,可以检测第三人像图像和第四人像图像的清晰度,如果第三人像图像的清晰度大于第四人像图像的清晰度,则在第二数据库中,用第三人像图像替换第四人像图像。
步骤S140,当基于第一数据库中的考勤记录确定在预设时间段内考勤场所中不存在认证用户时,进行报警处理。
在实施中,网络侧设备可以在基于第一数据库中的考勤记录确定在预设时间段内考勤场所中不存在认证用户时,进行报警处理。或者,在网络侧设备包括几台不同的服务器的情况下,设置单独的考勤服务器执行上述操作。
可选地,步骤S140可以包括:当基于第一数据库中的考勤记录确定在预设时间段内考勤场所中不存在认证用户时,获取预设的报警对象对应的通信标识信息;向通信标识信息对应的终端发送报警信息。
在实施中,对于某些高危考勤场所,例如高压变电站,每时每刻都需要值班人员在高压变电站中进行值守,不能出现无人值守的情况。除非是在中午12点到1点之间,可以有半小时没有值班人员进行值守,留半小时给值班人员外出用餐。可以确定除预设时间段之外的可离岗时间段;如果在可离岗时间段内, 基于第一数据库中的考勤记录确定考勤场所中不存在认证用户的时长大于预设最大时长,则获取预设的报警对象对应的通信标识信息,向通信标识信息对应的终端发送报警信息。
在实施中,可以设置除预设时间段之外的可离岗时间段,如中午12点到1点之间。到达可离岗时间段时,可以基于第一数据库中的考勤记录确定考勤场所不存在认证用户的时长,在所有认证用户都离开考勤场所时开始计时,如果计时超过预设最大时长时还没有认证用户进入考勤场所,则可以获取预设的报警对象对应的通信标识信息,向通信标识信息对应的终端发送报警信息。如果计时未超过预设最大时长时就有认证用户进入考勤场所,则可以停止并清空计时。
除此以外,在预设时间段(即可离岗时间段之外的时间段)内不能出现无人值守的情况。对于到底是哪个值班人员在值守不做要求,但是必须有人在值守,否则就要进行报警处理。
一般可以安排一个值班人员一次性值班两天两夜,之后会有其他人员去接替已值班两天两夜的值班人员。如此循环,达到每时每刻都有值班人员在高压变电站中进行值守的目的。
当基于第一数据库中的考勤记录确定在预设时间段内考勤场所中不存在认证用户时,可以获取预设的报警对象对应的通信标识信息,如获取当前应该值班的值班人员的经理的电话号码,向通信标识信息对应的终端发送报警信息。这样,当前应该值班的值班人员的经理可以接收到报警信息,获知当前应该值班的值班人员没有到岗。当前应该值班的值班人员的经理可以派遣当前应该值班的值班人员立即前往岗位,或者派遣其他人员立即前往岗位。
除此以外,还可以基于第一数据库中的考勤记录,确定进入考勤场所的认证用户是不是当天第一次进入考勤场所。或者,基于第一数据库中的每个认证用户的一整个月的考勤记录,确定每个认证用户是不是当月每天都能按时到岗,是不是有提前离岗的行为等,进一步来确定当月的考勤绩效。
总之,如果网络侧设备包括几台服务器时,如图2所示,当有用户来到门禁设备附近时,可以说出“开门”、“宿州变电站”等词语。接着,门禁设备附近的音频采集装置可以采集到第一人声音频,将第一人声音频发送到声纹识别服务器,声纹识别服务器可以对第一人声音频进行识别,判断用户是不是认证 用户,且用户是哪名认证用户。当声纹识别服务器识别音频失败时,声纹识别服务器向门禁设备附近的图像拍摄装置发送图像拍摄指令,图像拍摄装置可以拍摄第一人像图像,将第一人像图像发送给人脸识别服务器。人脸识别服务器可以对第一人像图像进行识别,判断用户是不是认证用户,且用户是哪名认证用户。在声纹识别服务器或者人脸识别服务器识别通过之后,可以向考勤服务器发送第一认证用户的标识、当前的时间信息、出入行为信息等,同时还可以向远程监控服务器发送开锁指令。当远程监控服务器接收到开锁指令之后,可以将开锁指令转发至电锁控制端。电锁控制端在接收到开锁指令之后,进行开锁操作。
通过本实施例提供的方法,获取音频采集装置采集的第一人声音频的第一声纹特征;将第一声纹特征,输入预先训练的每个认证用户的声纹匹配识别模型中,得到第一声纹特征与每个认证用户的声纹匹配度;如果得到的声纹匹配度中存在大于预设阈值的目标声纹匹配度,则确定目标声纹匹配度对应的第一认证用户,在第一数据库中添加与第一认证用户对应的考勤记录,其中,与第一认证用户对应的考勤记录包括当前的时间信息和第一认证用户当前对应考勤场所的出入行为信息,出入行为信息包括离开考勤场所或进入考勤场所;当基于第一数据库中的考勤记录确定在预设时间段内考勤场所中不存在认证用户时,进行报警处理。提供了一种有效机制来判断考勤场所中是否在预设时间段内都有认证用户在岗,进而可以降低因为考勤场所中无人在岗而出现事故的概率。
本申请又一示例性实施例提供了一种考勤管理设备,如图3所示,该考勤管理设备包括:
音频采集装置310,用于采集第一人声音频;
声纹特征提取模块320,用于获取所述音频采集装置采集的第一人声音频的第一声纹特征;
输入模块330,用于将所述第一声纹特征,输入预先训练的每个认证用户的声纹匹配识别模型中,得到所述第一声纹特征与每个认证用户的声纹匹配度;
第一确定模块340,用于当得到的声纹匹配度中存在大于预设阈值的目标声纹匹配度时,确定所述目标声纹匹配度对应的第一认证用户,在第一数据库中 添加与所述第一认证用户对应的考勤记录,其中,与所述第一认证用户对应的考勤记录包括当前的时间信息和所述第一认证用户当前对应考勤场所的出入行为信息,所述出入行为信息包括离开考勤场所或进入考勤场所;
报警模块350,用于当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中不存在认证用户时,进行报警处理。
可选地,所述考勤管理设备还包括:
图像拍摄装置,用于当得到的声纹匹配度中不存在大于预设阈值的目标声纹匹配度时,拍摄第一人像图像;
第一查找模块,用于在存储有认证用户的人像图像的第二数据库中,查找是否有与所述第一人像图像相匹配的第二人像图像;
第二确定模块,用于当查找到与所述第一人像图像相匹配的第二人像图像时,确定与所述第二人像图像对应的第二认证用户,在所述第一数据库中添加与所述第二认证用户对应的考勤记录,其中,与所述第二认证用户对应的考勤记录包括当前的时间信息和所述第二认证用户当前对应考勤场所的出入行为信息。
可选地,所述考勤管理设备还包括:
训练模块,用于基于所述第一声纹特征,训练所述第二认证用户的声纹匹配识别模型。
可选地,所述图像拍摄装置,还用于通过图像拍摄考勤管理设备拍摄第三人像图像;
所述考勤管理设备还包括:
第二查找模块,用于在存储有认证用户的人像图像的第二数据库中,查找所述第一认证用户的第四人像图像;
检测模块,用于分别检测所述第三人像图像和所述第四人像图像的清晰度;
替换模块,用于当所述第三人像图像的清晰度大于所述第四人像图像的清晰度时,在所述第二数据库中,用所述第三人像图像替换所述第四人像图像。
可选地,所述报警模块包括:
获取单元,用于当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中不存在认证用户时,获取预设的报警对象对应的通信标识信息;
发送单元,用于向所述通信标识信息对应的终端发送报警信息。
可选地,所述考勤管理设备还包括:
第三确定模块,用于确定除所述预设时间段之外的可离岗时间段;
发送模块,用于当在所述可离岗时间段内,基于所述第一数据库中的考勤记录确定所述考勤场所中不存在认证用户的时长大于预设最大时长时,获取预设的报警对象对应的通信标识信息,向所述通信标识信息对应的终端发送报警信息。
关于上述实施例中的装置,其中各个模块执行操作的具体方式已经在有关该方法的实施例中进行了详细描述,此处将不做详细阐述说明。
本申请提供了一种有效机制来判断考勤场所中是否在预设时间段内都有认证用户在岗,进而可以降低因为考勤场所中无人在岗而出现事故的概率。
需要说明的是:上述实施例提供的考勤管理的装置在考勤管理时,仅以上述各功能模块的划分进行举例说明,实际应用中,可以根据需要而将上述功能分配由不同的功能模块完成,即将网络侧设备的内部结构划分成不同的功能模块,或者设置多台服务器,以完成以上描述的全部或者部分功能。另外,上述实施例提供的考勤管理的装置与考勤管理的方法实施例属于同一构思,其具体实现过程详见方法实施例,这里不再赘述。
图4示出了本发明一个示例性实施例提供的网络侧设备1900的结构示意图。该网络侧设备1900可因配置或性能不同而产生比较大的差异,可以包括一个或一个以上处理器(central processing units,CPU)1910和一个或一个以上的存储器1920。其中,所述存储器1920中存储有至少一条指令,所述至少一条指令由所述处理器1910加载并执行以实现上述实施例所述的考勤管理的方法。
本领域技术人员在考虑说明书及实践这里公开的公开后,将容易想到本申请的其它实施方案。本申请旨在涵盖本申请的任何变型、用途或者适应性变化,这些变型、用途或者适应性变化遵循本申请的一般性原理并包括本申请未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的,本申请的真正范围和精神由权利要求指出。
应当理解的是,本申请并不局限于上面已经描述并在附图中示出的精确结 构,并且可以在不脱离其范围进行各种修改和改变。本申请的范围仅由所附的权利要求来限制。

Claims (12)

  1. 一种考勤管理的方法,其特征在于,所述方法包括:
    获取音频采集装置采集的第一人声音频的第一声纹特征;
    将所述第一声纹特征,输入预先训练的每个认证用户的声纹匹配识别模型中,得到所述第一声纹特征与每个认证用户的声纹匹配度;
    如果得到的声纹匹配度中存在大于预设阈值的目标声纹匹配度,则确定所述目标声纹匹配度对应的第一认证用户,在第一数据库中添加与所述第一认证用户对应的考勤记录,其中,与所述第一认证用户对应的考勤记录包括当前的时间信息和所述第一认证用户当前对应考勤场所的出入行为信息,所述出入行为信息包括离开考勤场所或进入考勤场所;
    当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中不存在认证用户时,进行报警处理。
  2. 根据权利要求1所述的方法,其特征在于,所述方法还包括:
    如果得到的声纹匹配度中不存在大于预设阈值的目标声纹匹配度,则通过图像拍摄装置拍摄第一人像图像;
    在存储有认证用户的人像图像的第二数据库中,查找是否有与所述第一人像图像相匹配的第二人像图像;
    如果查找到与所述第一人像图像相匹配的第二人像图像,则确定与所述第二人像图像对应的第二认证用户,在所述第一数据库中添加与所述第二认证用户对应的考勤记录,其中,与所述第二认证用户对应的考勤记录包括当前的时间信息和所述第二认证用户当前对应考勤场所的出入行为信息。
  3. 根据权利要求2所述的方法,其特征在于,所述方法还包括:
    基于所述第一声纹特征,训练所述第二认证用户的声纹匹配识别模型。
  4. 根据权利要求1所述的方法,其特征在于,在确定所述目标声纹匹配度对应的第一认证用户之后,所述方法还包括:
    通过图像拍摄装置拍摄第三人像图像;
    在存储有认证用户的人像图像的第二数据库中,查找所述第一认证用户的第四人像图像;
    分别检测所述第三人像图像和所述第四人像图像的清晰度;
    如果所述第三人像图像的清晰度大于所述第四人像图像的清晰度,则在所述第二数据库中,用所述第三人像图像替换所述第四人像图像。
  5. 根据权利要求1所述的方法,其特征在于,所述当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中不存在认证用户时,进行报警处理,包括:
    当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中不存在认证用户时,获取预设的报警对象对应的通信标识信息;
    向所述通信标识信息对应的终端发送报警信息。
  6. 根据权利要求5所述的方法,其特征在于,所述方法还包括:
    确定除所述预设时间段之外的可离岗时间段;
    如果在所述可离岗时间段内,基于所述第一数据库中的考勤记录确定所述考勤场所中不存在认证用户的时长大于预设最大时长,则获取预设的报警对象对应的通信标识信息,向所述通信标识信息对应的终端发送报警信息。
  7. 一种考勤管理设备,其特征在于,所述考勤管理设备包括:
    音频采集装置,用于采集第一人声音频;
    声纹特征提取模块,用于获取所述音频采集装置采集的第一人声音频的第一声纹特征;
    输入模块,用于将所述第一声纹特征,输入预先训练的每个认证用户的声纹匹配识别模型中,得到所述第一声纹特征与每个认证用户的声纹匹配度;
    第一确定模块,用于当得到的声纹匹配度中存在大于预设阈值的目标声纹匹配度时,确定所述目标声纹匹配度对应的第一认证用户,在第一数据库中添加与所述第一认证用户对应的考勤记录,其中,与所述第一认证用户对应的考勤记录包括当前的时间信息和所述第一认证用户当前对应考勤场所的出入行为信息,所述出入行为信息包括离开考勤场所或进入考勤场所;
    报警模块,用于当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中不存在认证用户时,进行报警处理。
  8. 根据权利要求7所述的考勤管理设备,其特征在于,所述考勤管理设备还包括:
    图像拍摄装置,用于当得到的声纹匹配度中不存在大于预设阈值的目标声 纹匹配度时,拍摄第一人像图像;
    第一查找模块,用于在存储有认证用户的人像图像的第二数据库中,查找是否有与所述第一人像图像相匹配的第二人像图像;
    第二确定模块,用于当查找到与所述第一人像图像相匹配的第二人像图像时,确定与所述第二人像图像对应的第二认证用户,在所述第一数据库中添加与所述第二认证用户对应的考勤记录,其中,与所述第二认证用户对应的考勤记录包括当前的时间信息和所述第二认证用户当前对应考勤场所的出入行为信息。
  9. 根据权利要求8所述的考勤管理设备,其特征在于,所述考勤管理设备还包括:
    训练模块,用于基于所述第一声纹特征,训练所述第二认证用户的声纹匹配识别模型。
  10. 根据权利要求7所述的考勤管理设备,其特征在于,所述图像拍摄装置,还用于通过图像拍摄考勤管理设备拍摄第三人像图像;
    所述考勤管理设备还包括:
    第二查找模块,用于在存储有认证用户的人像图像的第二数据库中,查找所述第一认证用户的第四人像图像;
    检测模块,用于分别检测所述第三人像图像和所述第四人像图像的清晰度;
    替换模块,用于当所述第三人像图像的清晰度大于所述第四人像图像的清晰度时,在所述第二数据库中,用所述第三人像图像替换所述第四人像图像。
  11. 根据权利要求7所述的考勤管理设备,其特征在于,所述报警模块包括:
    获取单元,用于当基于所述第一数据库中的考勤记录确定在预设时间段内所述考勤场所中不存在认证用户时,获取预设的报警对象对应的通信标识信息;
    发送单元,用于向所述通信标识信息对应的终端发送报警信息。
  12. 根据权利要求11所述的考勤管理设备,其特征在于,所述考勤管理设备还包括:
    第三确定模块,用于确定除所述预设时间段之外的可离岗时间段;
    发送模块,用于当在所述可离岗时间段内,基于所述第一数据库中的考勤记录确定所述考勤场所中不存在认证用户的时长大于预设最大时长时,获取预 设的报警对象对应的通信标识信息,向所述通信标识信息对应的终端发送报警信息。
PCT/CN2019/083089 2018-06-01 2019-04-17 考勤管理的方法和考勤管理设备 WO2019228091A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810556426.3A CN110555918B (zh) 2018-06-01 2018-06-01 考勤管理的方法和考勤管理设备
CN201810556426.3 2018-06-01

Publications (1)

Publication Number Publication Date
WO2019228091A1 true WO2019228091A1 (zh) 2019-12-05

Family

ID=68696835

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/083089 WO2019228091A1 (zh) 2018-06-01 2019-04-17 考勤管理的方法和考勤管理设备

Country Status (2)

Country Link
CN (1) CN110555918B (zh)
WO (1) WO2019228091A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111192374A (zh) * 2020-01-02 2020-05-22 湖南映客互娱网络信息有限公司 一种实时在线监控方法与系统
CN112052731A (zh) * 2020-07-30 2020-12-08 广州市标准化研究院 一种智能人像识别打卡考勤系统及方法
CN114399850A (zh) * 2021-12-27 2022-04-26 四川领歌智谷科技有限公司 基于语音及位置识别的考勤方法、电子学生证及存储介质
CN114462720A (zh) * 2022-04-08 2022-05-10 山东云小兵信息技术有限公司 智能考勤管理方法、装置、设备及存储介质
CN117412180A (zh) * 2023-12-15 2024-01-16 杭州三信网络技术有限公司 一种基于多相机联动目标监测的焊机和目标监测方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112600846B (zh) * 2020-12-15 2023-01-10 中标慧安信息技术股份有限公司 基于物联网的考勤管理方法和系统

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102034288A (zh) * 2010-12-09 2011-04-27 江南大学 基于多生物特征识别的智能门禁系统
US9152837B2 (en) * 2007-06-11 2015-10-06 Jeffrey A. Matos Apparatus and method for verifying the identity of an author and a person receiving information
CN205028367U (zh) * 2015-09-30 2016-02-10 恒东信息科技无锡有限公司 声纹、身份证信息感知通道
CN106251419A (zh) * 2016-08-24 2016-12-21 天津中利网络科技有限公司 一种新型智能化物联网考勤机
CN205920510U (zh) * 2016-06-28 2017-02-01 安徽联商信息科技有限公司 一种人脸声纹考勤系统
CN106790054A (zh) * 2016-12-20 2017-05-31 四川长虹电器股份有限公司 基于人脸识别和声纹识别的交互式认证系统及方法
CN107993300A (zh) * 2017-10-23 2018-05-04 林楚莲 一种基于智能终端的考勤方法及装置

Family Cites Families (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1936967A (zh) * 2005-09-20 2007-03-28 吴田平 声纹考勤机
JP2008065572A (ja) * 2006-09-07 2008-03-21 Konica Minolta Business Technologies Inc 生体認証システムに用いる生体情報の更新方法および生体認証システム
CN101231691A (zh) * 2007-12-24 2008-07-30 浙江金指码科技有限公司 通过指纹数据数量调整的学习型智能指纹识别比对方法
JP5218991B2 (ja) * 2009-12-08 2013-06-26 株式会社日立製作所 複数種類のテンプレートを用いた生体認証システム及び生体認証方法
CN201654950U (zh) * 2010-04-29 2010-11-24 李长觉 值勤监督报警器
CN102339483A (zh) * 2010-07-14 2012-02-01 新谊整合科技股份有限公司 整合门禁与信息设备的保全系统与方法
CN103065377A (zh) * 2011-10-21 2013-04-24 深圳市新文鼎信息技术有限公司 基于有源rfid的出入管理系统
CN103632667B (zh) * 2013-11-25 2017-08-04 华为技术有限公司 声学模型优化方法、装置及语音唤醒方法、装置和终端
CN103605971B (zh) * 2013-12-04 2017-05-24 深圳市捷顺科技实业股份有限公司 一种捕获人脸图像的方法及装置
CN103871118A (zh) * 2014-03-05 2014-06-18 浙江硕坤智能科技有限公司 基于有源2.4g rfid的考勤管理系统及方法
CN204117204U (zh) * 2014-07-18 2015-01-21 南京交通职业技术学院 指纹识别的rfid考勤管理装置
CN107209848B (zh) * 2014-12-01 2020-10-13 厦门熵基科技有限公司 用于基于多模式生物识别信息的个人识别的系统和方法
CN104680375A (zh) * 2015-02-28 2015-06-03 优化科技(苏州)有限公司 电子支付真人活体身份验证系统
CN204598183U (zh) * 2015-05-11 2015-08-26 保定和易法电气科技有限公司 值班人员在岗监控系统
CN204650614U (zh) * 2015-05-21 2015-09-16 张亮 一种执勤监督装置
CN105184884A (zh) * 2015-09-28 2015-12-23 席广涵 一种指纹考勤装置
CN105578392A (zh) * 2015-12-21 2016-05-11 国网辽宁省电力有限公司鞍山供电公司 人员离岗智能监测装置、方法及系统
CN105631430A (zh) * 2015-12-30 2016-06-01 浙江宇视科技有限公司 一种人脸图像的匹配方法和装置
CN205405625U (zh) * 2016-03-08 2016-07-27 徐志秀 一种建筑从业人员到岗监管终端
CN106022290B (zh) * 2016-05-30 2018-03-27 广东欧珀移动通信有限公司 一种指纹模板更新方法及终端设备
CN106156583A (zh) * 2016-06-03 2016-11-23 深圳市金立通信设备有限公司 一种语音解锁的方法及终端
CN106127152B (zh) * 2016-06-21 2019-07-19 Oppo广东移动通信有限公司 一种指纹模板更新方法及终端设备
CN106204393A (zh) * 2016-07-13 2016-12-07 合肥指南针电子科技有限责任公司 一种服务大厅智能监控系统
CN107644189B (zh) * 2016-07-20 2021-02-02 腾讯科技(深圳)有限公司 一种生物特征识别的方法及装置
CN107808423A (zh) * 2016-09-09 2018-03-16 广州零号软件科技有限公司 一种基于服务机器人的员工考勤签到系统与实现方法
CN106406798B (zh) * 2016-09-30 2023-08-04 曲阜师范大学 远程交互显示实时推送方法及系统
CN106600739B (zh) * 2016-12-08 2022-12-16 江苏展邦智能科技有限公司 一种监测巡检人员的方法及系统
CN106534206A (zh) * 2016-12-28 2017-03-22 汉王科技股份有限公司 一种基于云平台的身份认证方法、装置以及系统
CN106934870A (zh) * 2017-02-17 2017-07-07 安徽金猫数字科技有限公司 一种语音考勤系统
CN106951864A (zh) * 2017-03-20 2017-07-14 广东小天才科技有限公司 一种指纹采集方法及装置
CN106951762A (zh) * 2017-03-29 2017-07-14 中控智慧科技股份有限公司 一种身份识别方法及系统
CN107316070A (zh) * 2017-06-05 2017-11-03 安徽福讯信息技术有限公司 一种基于rfid的人员与重要设备监控系统
CN107909665A (zh) * 2017-11-06 2018-04-13 北京上古视觉科技有限公司 一种学生签到系统以及学生签到方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9152837B2 (en) * 2007-06-11 2015-10-06 Jeffrey A. Matos Apparatus and method for verifying the identity of an author and a person receiving information
CN102034288A (zh) * 2010-12-09 2011-04-27 江南大学 基于多生物特征识别的智能门禁系统
CN205028367U (zh) * 2015-09-30 2016-02-10 恒东信息科技无锡有限公司 声纹、身份证信息感知通道
CN205920510U (zh) * 2016-06-28 2017-02-01 安徽联商信息科技有限公司 一种人脸声纹考勤系统
CN106251419A (zh) * 2016-08-24 2016-12-21 天津中利网络科技有限公司 一种新型智能化物联网考勤机
CN106790054A (zh) * 2016-12-20 2017-05-31 四川长虹电器股份有限公司 基于人脸识别和声纹识别的交互式认证系统及方法
CN107993300A (zh) * 2017-10-23 2018-05-04 林楚莲 一种基于智能终端的考勤方法及装置

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111192374A (zh) * 2020-01-02 2020-05-22 湖南映客互娱网络信息有限公司 一种实时在线监控方法与系统
CN112052731A (zh) * 2020-07-30 2020-12-08 广州市标准化研究院 一种智能人像识别打卡考勤系统及方法
CN112052731B (zh) * 2020-07-30 2024-03-29 广州市标准化研究院 一种智能人像识别打卡考勤系统及方法
CN114399850A (zh) * 2021-12-27 2022-04-26 四川领歌智谷科技有限公司 基于语音及位置识别的考勤方法、电子学生证及存储介质
CN114462720A (zh) * 2022-04-08 2022-05-10 山东云小兵信息技术有限公司 智能考勤管理方法、装置、设备及存储介质
CN117412180A (zh) * 2023-12-15 2024-01-16 杭州三信网络技术有限公司 一种基于多相机联动目标监测的焊机和目标监测方法
CN117412180B (zh) * 2023-12-15 2024-03-15 杭州三信网络技术有限公司 一种基于多相机联动目标监测的焊机和目标监测方法

Also Published As

Publication number Publication date
CN110555918B (zh) 2022-04-26
CN110555918A (zh) 2019-12-10

Similar Documents

Publication Publication Date Title
WO2019228091A1 (zh) 考勤管理的方法和考勤管理设备
US6975346B2 (en) Method for suspect identification using scanning of surveillance media
CN206557873U (zh) 一种基于人脸识别的智能安保防进入系统
CN106204815B (zh) 一种基于人脸检测和识别的门禁系统
CN104966344B (zh) 一种基于视频的安防门禁系统
WO2017036049A1 (zh) 一种用户身份识别的方法、装置、服务器以及系统
CN206515931U (zh) 一种人脸识别系统
US20130216107A1 (en) Method of surveillance by face recognition
US20190197848A1 (en) Networked premises security
CN106951846A (zh) 一种人脸3d模型录入和识别方法和装置
CN207895501U (zh) 用于监狱的自动点名装置
KR102237086B1 (ko) 안면인식 기능에 기초하여 5g 이동통신망의 이용이 가능한 통신 단말기를 통해 영상감시가 가능한 로비폰 제어 장치 및 방법
CN103246868A (zh) 一种实现快速报警且快速确警的方法
CN112734988A (zh) 一种智能门锁的胁迫开锁的判断与报警系统及其方法
CN110096974A (zh) 一种基于人脸识别的景点游客入园核验的方法及系统
CN108846912A (zh) 考勤方法、终端及服务器
CN109784231A (zh) 安防信息管理方法、装置及存储介质
KR20160072386A (ko) 특징점을 기반으로 한 얼굴 인식을 이용한 홈네트워크 시스템 및 그 제어 방법
CN111429638B (zh) 一种基于语音识别和人脸识别的门禁控制方法
CN113034768A (zh) 一种智能门禁装置及其工作方法
WO2015096429A1 (zh) 通话声音识别方法及装置
CN108573554A (zh) 一种基于人脸锁的自动识别报警方法及系统
CN108764153A (zh) 人脸识别方法、装置、系统和存储介质
CN114898443A (zh) 一种人脸数据获取方法及装置
EP3828837A1 (en) Frictionless security processing

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19812210

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19812210

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 22/06/2021)