CN110738770A - Face recognition forbidden processing method, gate, control end and system - Google Patents
Face recognition forbidden processing method, gate, control end and system Download PDFInfo
- Publication number
- CN110738770A CN110738770A CN201910910550.XA CN201910910550A CN110738770A CN 110738770 A CN110738770 A CN 110738770A CN 201910910550 A CN201910910550 A CN 201910910550A CN 110738770 A CN110738770 A CN 110738770A
- Authority
- CN
- China
- Prior art keywords
- gate
- face
- sub
- library
- comparison result
- 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
- 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)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The application relates to human face recognition forbidden processing methods, gates, a control end, a system, computer equipment and a storage medium, wherein the method comprises the steps that a gate detects a human face and calculates a human face characteristic value, the human face characteristic value is sent to a second gate, the human face characteristic value is compared with a sub-human face library stored in the gate at the gate , the second gate compares the human face characteristic value with a second sub-human face library stored in the second gate and feeds back the comparison result to a gate , the two gates calculate in parallel, the calculation efficiency is improved, the utilization rate of a gate CPU is also improved, meanwhile, the sending of the human face characteristic value and the feedback of the result are both local communication, the communication efficiency is high, and therefore the efficiency of the forbidden gate human face recognition process is improved.
Description
Technical Field
The application relates to the technical field of artificial intelligence, in particular to a forbidden processing method, a gate, a control end, a system, computer equipment and a storage medium for face recognition with types of load balancing.
Background
With the development of the face recognition technology, the technology is gradually applied to an entrance banned security management system, by shooting a face image of a person entering an entrance area of a gate, extracting facial feature information of the person from the face image, and comparing the facial feature information with a face in a remote server or a face library local to the gate, banned security management system finally makes a judgment whether to open .
However, in the face recognition method applied to the banned security management system in the related art, a plurality of channels with gates are arranged at an entrance and an exit, each gate in the gates is independent arithmetic units, although each gate can support large-capacity face library comparison through an intelligent server connected with a cloud platform, face recognition is performed only by using the cloud platform, the gate opening is slow due to the time of communication between the gate and the cloud platform, and in the method, periods of time exist when a single gate catches a face opening from times to times after face identification, a CPU of the gate is idle in the period of time, and the CPU arithmetic resources of the gate are wasted, so that the process of the face recognition opening is not efficient.
Aiming at the problem that the face recognition method of the forbidden system in the related technology is not high in efficiency, an effective solution is not provided at present.
Disclosure of Invention
Based on this, it is necessary to provide load balancing face recognition methods, systems, apparatuses, computer devices and storage media for solving the above technical problems.
To achieve the above object, according to aspects of the present invention, there are provided face recognition disabling methods, including:
the gate obtains a face feature value, and sends the face feature value to a second gate, wherein the gate stores a sub-face library of a face library, and the second gate stores a second sub-face library of the face library;
the gate compares the face feature value with the sub-face library to obtain a comparison result, and the gate receives a second comparison result of the second gate, wherein the second comparison result is obtained by the second gate comparing the face feature value with the second sub-face library;
the gate is switched according to the comparison result and the second comparison result.
In embodiments, the comparing the face feature values with the sub-face library by the gate to obtain a comparison result comprises:
the gate acquires a th load state of the gate and a third load state of a third gate, and determines a th processing gate according to a load balancing algorithm, wherein the th gate and the third gate both belong to a th gate group, and the third gate stores the th sub-face library;
the gate obtains th comparison result, wherein the th processing gate compares the face feature value with the th sub-face library to obtain th comparison result.
In embodiments, the comparison result obtained by the second gate comparing the face feature value with the second sub-face library includes:
the gate acquires a second load state of the second gate and a fourth load state of a fourth gate, and determines a second processing gate according to a load balancing algorithm, wherein the second gate and the fourth gate both belong to a second gate group, the fourth gate stores the second child face library,
and the second comparison result is obtained by comparing the face characteristic value with the second sub-face library by the second processing gate.
According to another aspects of the invention, there are provided face recognition processing-forbidden methods, the method comprising:
the control end receives a face feature value sent by an th gate, the control end sends the face feature value to a second gate, the th gate stores a th sub-face library of a face library, the th gate performs face comparison on the face feature value and the th sub-face library to obtain a th comparison result, the second gate stores a second sub-face library of the face library, the second gate performs face comparison on the face feature value and the second sub-face library to obtain a second comparison result, and the second gate sends the second comparison result to the th gate.
In embodiments, the gate comparing the face feature values with the sub-face library to obtain a comparison result includes:
the control end obtains a load state of the gate and a third load state of a third gate, and determines a processing gate according to a load balancing algorithm, wherein the gate and the third gate both belong to a gate group, the third gate stores the sub-face library, the control end sends the face feature value to the processing gate, and the processing gate performs face comparison on the face feature value and the sub-face library to obtain an comparison result.
In embodiments, the performing, by the second gate, a face comparison between the face feature value and the second sub-face library to obtain a second comparison result includes:
the control end obtains a second load state of the second gate and a fourth load state of a fourth gate, and determines a second processing gate according to a load balancing algorithm, wherein the second gate and the fourth gate both belong to a second gate set, the fourth gate stores the second sub-face library, the control end sends the face feature value to the second processing gate, and the second processing gate compares the face feature value with the second sub-face library to obtain a second comparison result.
According to another aspects of the invention, there are also provided gate machines comprising:
a capturing module for obtaining the face characteristic value,
the receiving and sending module is used for sending the face characteristic value to a second gate, wherein the th gate stores a th sub-face library of a face library, and the second gate stores a second sub-face library of the face library;
a comparison module for comparing the face feature value with the th sub-face library to obtain th comparison result,
a collecting module, configured to receive, by the th gate, a second comparison result of the second gate, where the second comparison result is a result obtained by the second gate comparing the face feature value with the second sub-face library;
and the gate interceptor performs switching according to the th comparison result and the second comparison result.
In embodiments, the gate further comprises:
an load module, configured to obtain a load state of the gate and a third load state of a third gate, and determine a processing gate according to a load balancing algorithm, where the gate and the third gate both belong to a gate group, and the third gate stores the sub-face library;
the gathering module is further configured to obtain th comparison results, wherein the th processing gate compares the face feature values with the th sub-face library to obtain th comparison results.
In embodiments, the gate further comprises,
And the second load module is used for acquiring a second load state of a second gate and a fourth load state of a fourth gate, and determining a second processing gate according to a load balancing algorithm, wherein the second gate and the fourth gate both belong to a second gate set, the fourth gate stores the second sub-face library, and the second comparison result is obtained by comparing the face feature value with the second sub-face library by the second processing gate.
According to another aspects of the present invention, there are also provided control terminals, comprising:
a forwarding module, configured to receive a face feature value sent by an th gate, and send the face feature value to a second gate, where the th gate stores a th sub-face library of a face library, the th gate performs face comparison on the face feature value and the th sub-face library to obtain a th comparison result, the second gate stores a second sub-face library of the face library, the second gate performs face comparison on the face feature value and the second sub-face library to obtain a second comparison result, and the second gate sends the second comparison result to the th gate.
In embodiments, the control terminal further includes:
an selecting module, configured to obtain a th load state of the gate and a third load state of a third gate, and determine a th processing gate according to a load balancing algorithm, where the th gate and the third gate both belong to a th gate group, and the third gate stores the th sub-face library;
the forwarding module is further configured to send the face feature value to the th processing gate, where the th processing gate performs face comparison on the face feature value and the th sub-face library to obtain a th comparison result.
In embodiments, the control terminal further includes:
the second selection module is used for acquiring a second load state of the second gate and a fourth load state of a fourth gate, and determining a second processing gate according to a load balancing algorithm, wherein the second gate and the fourth gate both belong to a second gate set, and the fourth gate stores the second sub-face library;
the forwarding module is further configured to send the face feature value to the second processing gate, where the second processing gate compares the face feature value with the second sub-face library to obtain a second comparison result.
According to another aspects of the invention, there is also provided a face recognition ban system, the system comprising a gate and a second gate:
the gate obtains a face feature value, and sends the face feature value to a second gate, wherein the gate stores a sub-face library of a face library, and the second gate stores a second sub-face library of the face library;
the gate compares the face feature value with the sub-face library to obtain a comparison result, the second gate compares the face feature value with the second sub-face library to obtain a second comparison result, and the second gate sends the second comparison result to the gate;
the gate is switched according to the comparison result and the second comparison result.
According to another aspects of the invention, there is also provided a face recognition banning system, which comprises a control terminal, a gate, a second gate:
, acquiring a face characteristic value and sending the face characteristic value to a control end;
the control end receives the face characteristic value sent by the th gate, the control end sends the face characteristic value to a second gate, the th gate stores a th sub-face library of a face library, and the second gate stores a second sub-face library of the face library;
the gate compares the face feature value with the sub-face library to obtain a comparison result, the second gate compares the face feature value with the second sub-face library to obtain a second comparison result, the second gate sends the second comparison result to the gate, and the gate performs switching according to the comparison result and the second comparison result.
According to another aspects of the invention, there are also provided computer devices including a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the face recognition disabling method when executing the computer program.
According to another aspects of the invention, there are also provided computer-readable storage media having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-described method of face recognition suppression.
According to the invention, the gate detects a human face and calculates a human face characteristic value, the human face characteristic value is sent to the second gate, the gate compares the human face characteristic value with the sub-human face library stored in the gate, the second gate compares the human face characteristic value with the second sub-human face library stored in the second gate and feeds back the comparison result to the gate, the two gates calculate in parallel, the calculation efficiency is improved, the utilization rate of a gate CPU is also improved, meanwhile, the sending of the human face characteristic value and the feedback of the result are local communication, the communication efficiency is high, and the efficiency of the gate-forbidden human face recognition process is improved.
Drawings
FIG. 1 is a schematic diagram of an application scenario of face recognition processing-forbidden methods according to an embodiment of the present invention;
FIG. 2 is a flow chart of a processing method for face recognition inhibition according to an embodiment of the invention;
FIG. 3 is a schematic diagram of an application scenario of forbidden methods for recognizing in accordance with another embodiments of the present invention;
FIG. 4 is a schematic diagram of a second exemplary application scenario of forbidden face recognition processing methods according to another embodiments of the present invention;
FIG. 5 is a schematic diagram of a third exemplary application scenario of forbidden face recognition processing methods according to another embodiments of the present invention;
fig. 6 is a schematic diagram of an application scenario of face recognition forbidden processing methods according to another embodiments of the present invention;
FIG. 7 is a flow chart of a face recognition disable processing method in accordance with another embodiments of the present invention;
fig. 8 is a schematic diagram of an application scenario of face recognition forbidden processing methods according to another embodiments of the present invention;
fig. 9 is a schematic diagram six of an application scenario of face recognition forbidden processing methods according to another embodiments of the present invention;
fig. 10 is a diagram illustrating an application scenario seven of forbidden face recognition processing methods according to another embodiments of the present invention;
FIG. 11 is a block diagram of a face recognition gate inhibition machine according to embodiments of the present invention;
FIG. 12 is a block diagram of face recognition gate inhibitors in accordance with another embodiments of the present invention;
FIG. 13 is a schematic diagram of a face recognition gate inhibition machine according to another embodiments of the present invention;
FIG. 14 is a block diagram of a control-disabled terminal of face recognitions in embodiments according to the invention;
FIG. 15 is a block diagram of face recognition disable control terminals in accordance with another embodiments of the present invention;
FIG. 16 is a schematic diagram of a second exemplary configuration of a forbidden control terminal of face recognition in another embodiments of the present invention;
fig. 17 is a schematic diagram eight of an application scenario of face recognition forbidden processing methods according to another embodiments of the present invention;
fig. 18 is a schematic diagram of face recognition disable processing methods in another embodiments according to the invention.
Detailed Description
For purposes of making the present application, its objects, aspects and advantages more apparent, the present application is described in further detail with reference to the drawings and the examples.
The forbidden processing method for face recognition can be applied to a forbidden system for face recognition.
In embodiments, fig. 1 is an application scenario schematic diagram of face recognition forbidding processing methods in the embodiment of the present invention, fig. 2 is a flowchart of face recognition forbidding processing methods in the embodiment of the present invention, and as shown in fig. 2, face recognition forbidding processing methods are provided, which are described by taking an example that the method is applied to a gate in fig. 1, and include the following steps:
step S210, the th gate 11 obtains a face feature value, and sends the face feature value to the second gate 12, the th gate 11 stores the th sub-face library of the face library, and the second gate 12 stores the second sub-face library of the face library;
in step S210, the face recognition disabled device includes gateway 11 and a second gateway 12, and th gateway 11 and the second gateway 12 serve the same face library, in an actual application scenario, typically th gateway 11 and the second gateway 12 are disabled gateways of the same place, such as a cell or an office building, th gateway 11 stores th sub-face library of the face library, the second gateway 12 stores a second sub-face library of the face library, th sub-face library and the second sub-face library are different, especially when the face library capacity exceeds the storage capacity of gate 11, the face library can be split into two small sub-face libraries, which are stored in gate 11 and the second gateway 12, and th gateway 11 and the second gateway 12 are connected through a network, when a person passes through , a face image feature value is extracted from 4611, and a face feature value is extracted and sent to 4612 after the face feature value is detected.
Step S220, the gate 11 compares the face feature value with the sub-face library to obtain a comparison result, the gate 11 receives a second comparison result of the second gate 12, wherein the second comparison result is a result of the second gate 12 comparing the face feature value with the second sub-face library;
in step S220, the gate 11 compares the face feature value with the sub-face library to obtain a comparison result, and the second gate 12 compares the face feature value with the second sub-face library to obtain a second comparison result and feeds the result back to the gate 11.
Step S230, the gate 11 performs a switch operation according to the comparison result and the second comparison result;
in step S230, when the face feature value matches the face information in the second sub-face library, the second comparison result fed back to the gate by the second gate 12 includes hit information, and the gate receives the hit information, the gate barrier is opened, and the gate stops the comparison process between the face feature value and the sub-face library, thereby reducing unnecessary calculation processes and avoiding waste of calculation resources.
According to the processing forbidding method for the face recognition , the gate 11 and the second gate 12 simultaneously compare the face characteristic value with the sub-face library and the second sub-face library, in the aspect of , the two gates perform parallel calculation, so that the calculation efficiency is improved, the utilization rate of a gate CPU is also improved, in addition, in the aspect of , the sending of the face characteristic value and the feedback of a comparison result are local communication, the communication efficiency is high, and therefore the efficiency of a gate forbidding face recognition process is improved.
In various embodiments, fig. 3 is a schematic view 1 of an application scenario of 0 face recognition banning processing method according to another embodiment of the present invention, as shown in fig. 3, a 2 nd gate 11 and a third gate 31 both belong to a 3 rd gate group, and each gate in a 4 th gate group stores a 5 th sub-face library, a 6 th gate 11 acquires an 8 th load state of a 7 th gate 11 and a third load state of the third gate 31, selects a 9 th processing gate according to a preset load balancing algorithm, and is responsible for face comparison of the 0 th gate group, after detecting a face and calculating a face feature value, the 1 st gate 11 sends the face feature value to a 2 nd processing gate and a second gate 12 for comparison, the 3 rd processing gate compares the face feature value with the 4 th sub-face library to obtain a 5 th comparison result, and sends the 6 th comparison result to the 7 th gate 11, and similarly, the second gate 12 compares the face feature value with the second sub-face library to obtain a second comparison result, and the second comparison result compares the face feature value with the 4 th sub-face library to obtain a 5 th comparison result, and sends the 6 th comparison result to the gate 11, and the second gate 11, and the gate 11 sends the second gate 11 a random polling method to determine that the face comparison result and the gate group can be performed by a random polling method, and the gate group can only send the gate group with the gate group, and the gate group can perform a random comparison result in a random comparison efficiency in a random comparison method, and the random comparison result, and the random comparison method can be performed in a random comparison method, and the gate group with the gate group, and the gate group can be more than the gate group, and the gate group can be performed the gate group, and the gate group can.
In various embodiments, fig. 4 is a schematic diagram of an application scenario of a 0-type face recognition 7-banned processing method according to another embodiment of the present invention, as shown in fig. 4, a second gate 12 and a fourth gate 41 both belong to a second gate group, and each gate in the second gate group stores a second sub-face library, as shown in fig. 4, the first gate 11 acquires a second load state of the second gate 12 and a fourth load state of the fourth gate 41, selects a second processing gate according to a preset load balancing algorithm, and is responsible for face comparison of the second gate group, after detecting a face and calculating a face feature value, the gate 11 sends the face feature value to the gate 11 of the 3 and the second processing gate, and compares the face feature value with the face library of the 5 to obtain a comparison result of the gate 6, and feeds back the comparison result of the gate 8 to the gate 11 of the 9, as well as the second processing gate compares the feature value with the second sub-library to obtain a second comparison result of the gate 7, and the gate 6, and the gate 7, and the gate 6, as well as a gate 11, and a gate 7, and a gate 11, and a gate 7, and a gate 6, and a gate 7, which are all of a gate 7, and a gate 7, all of a gate 7, and a gate comparison method, all of a gate 7, and a gate 7, all of a gate 7, and a gate comparison method, all of a gate, and a gate, all of a gate method, all of a gate, and a gate group, and a gate method, and a gate 7, all of a gate method, and a gate method, all of a gate method, and a gate method, all of a gate method, and a gate method, all of a gate method, and a gate method, all of a gate.
In embodiments, fig. 6 is a schematic view of an application scenario of face recognition processing prohibiting methods in another embodiments of the present invention, fig. 7 is a flowchart of face recognition processing prohibiting methods in another embodiments of the present invention, and as shown in fig. 7, face recognition processing prohibiting methods are provided, which are applied to the gate in fig. 6 as an example to explain, and include the following steps:
step S710, the control end 61 receives the face feature value sent by the gate 11;
step S720, the control terminal 61 sends the face feature value to the second gate 12, wherein the gate stores the sub-face library of the face library, the gate compares the face feature value with the sub-face library to obtain the comparison result, the second gate 12 stores the second sub-face library of the face library, the second gate 12 compares the face feature value with the second sub-face library to obtain the second comparison result, and the second gate 12 sends the second comparison result to the gate 11.
In the face recognition forbidden processing method, in the aspect of , the two gates perform parallel calculation to improve the calculation efficiency and also improve the utilization rate of the gates CPU, the control end 61 is responsible for forwarding the face feature value to reduce the processing load of the forwarding process of the gate 11, when the number of gates needing to be forwarded is large, the processing efficiency can be further improved by , in addition, in the aspect of , the sending of the face feature value and the feedback of the comparison result are local communication, the communication efficiency is high, and therefore the efficiency of the face recognition process of forbidden gates is improved.
In embodiments, fig. 8 is a schematic diagram illustrating an application scenario of a face recognition disable processing method according to 0 in another embodiments of the present invention, as shown in fig. 8, gate 11 and third gate 31 both belong to a gate group, each gate in an gate group stores a sub-face library, 5 gate 11 detects a face and calculates a face feature value and then sends the face feature value to control terminal 61, control terminal 61 obtains load state of gate 11 and third load state of third gate 31, gate is processed according to a preset load balancing algorithm , control terminal 61 sends the face feature value to processing gate, gate is responsible for face comparison of by the processing gate, gate is selected by the 361 processing gate to compare the face feature value with 362 sub-face library to obtain a 363 comparison result, gate is selected by the processing gate, gate is responsible for face comparison of according to a second gate 72, gate 72 processing gate 72, gate is selected by a processing gate, gate 72 processing gate 72, gate 72 is selected by a switch for a weighted comparison, and a switch for selecting a , and a switch for selecting a random load comparison, and a , wherein the gate 72, a gate 72 is selected by a switch for a random comparison method for selecting a plurality of a gate 72, a gate 72 is selected gate 72, a gate 72.
In several embodiments, fig. 9 is a schematic diagram of an application scenario of a 0-type face recognition 6-banned processing method according to another embodiment of the present invention, as shown in fig. 9, a second gate 12 and a fourth gate 41 both belong to a second gate group, and each gate in the second gate group stores a second sub-face library, as shown in fig. 9, a face feature value is sent to a control terminal 61 after the face is detected and a face feature value is calculated, the control terminal 61 obtains a second load state of the second gate 12 and a fourth load state of the fourth gate 41, a second processing gate is selected according to a preset load balancing algorithm, the second processing gate is responsible for a face comparison of the second gate group, the control terminal 61 sends the face feature value to a 2-th processing gate and a second processing gate for face comparison, the 3-th processing gate compares the face feature value with a 4-sub-face library to obtain a 5-comparison result, the 7-comparison result is fed back to the gate 11 of the 8, as the same model, the second processing gate compares the face feature value with the second sub-gate library 4-face library to obtain a 5-comparison result, and a comparison result, the comparison result obtained by the second processing gate 7, the second processing gate 11, the second processing gate 7, the second processing gate, the gate group, the second processing gate group, the gate receives the comparison result, the comparison gate 7-gate, the comparison result obtained by the comparison gate, the comparison result obtained by the second processing gate 7 comparison gate group, the gate, the comparison gate 7, the comparison gate, the comparison result obtained by the gate 7, the comparison method, the gate, the comparison method, the gate group, the comparison method, the gate group.
In embodiments, fig. 11 is a schematic structural diagram of a gate inhibition machine for kinds of face recognition in embodiments of the present invention, as shown in fig. 11, the gate inhibition machine includes:
a capturing module 111, configured to obtain a face feature value,
a sending module 112, configured to send the face feature value to the second gate 12, where the th gate 11 stores a th sub-face library of the face library, and the second gate 12 stores a second sub-face library of the face library;
a comparison module 113, configured to compare the face feature value with an th sub-face library to obtain a th comparison result, and a collection module 114, configured to receive a second comparison result of the second gate 12, where the second comparison result is a result obtained by the second gate 12 comparing the face feature value with the second sub-face library;
the gate interceptor 115 performs a switching operation according to the th comparison result and the second comparison result.
In embodiments, fig. 12 is a schematic structural diagram of another embodiments of face recognition gate-inhibition machines according to the present invention, as shown in fig. 12, the gate further includes:
the load module 121 is configured to obtain a load state of the gate 11 and a third load state of the third gate 31, and determine the processing gate according to a load balancing algorithm, where the gate 11 and the third gate 31 both belong to a gate group, and the third gate 31 stores an sub-face library;
the gathering module 114 is further configured to obtain th comparison results, wherein the th processing gate compares the face feature values with the th sub-face library to obtain th comparison results.
In embodiments, fig. 13 is a schematic structural diagram of a gate inhibition machine for face recognition in embodiments according to the present invention, as shown in fig. 13, the gate inhibition machine further includes:
the second load module 131 is configured to obtain a second load state of the second gate 12 and a fourth load state of the fourth gate 41, and determine a second processing gate according to a load balancing algorithm, where the second gate 12 and the fourth gate 41 both belong to a second gate group, the fourth gate 41 stores a second sub-face library, and a second comparison result is a result obtained by comparing the face feature value with the second sub-face library by the second processing gate.
Wherein the capture module 111 is generally located within a camera portion of the gate, the transmit module 112, the compare module 113, the gather module 114, the th load module 121, and the second load module 131 are generally located within a processor of the gate, and the gate discourager 115 is generally a rotatable gate or gate of the gate.
The gate detects a human face, calculates a human face characteristic value and sends the human face characteristic value to the second gate, the gate compares the human face characteristic value with an th sub-face library stored in the gate, the second gate compares the human face characteristic value with a second sub-face library stored in the second gate and feeds back a comparison result to the gate, the two gates perform parallel calculation, the calculation efficiency is improved, the utilization rate of a gate CPU is also improved, meanwhile, the sending of the human face characteristic value and the feedback of the result are both local communication, the communication efficiency is high, and therefore the efficiency of a gate-forbidden human face recognition process is improved.
In embodiments, fig. 14 is a schematic structural diagram of a control-forbidden terminal for face recognition in embodiments according to the present invention, as shown in fig. 14, the control terminal includes:
the forwarding module 141 is configured to receive the face feature value sent by the th gate 11, and send the face feature value to the second gate 12, where the th gate 11 stores a th sub-face library of the face library, the th gate 11 performs face comparison on the face feature value and a th sub-face library to obtain a th comparison result, the second gate 12 stores a second sub-face library of the face library, the second gate 12 performs face comparison on the face feature value and the second sub-face library to obtain a second comparison result, and the second gate 12 sends the second comparison result to the th gate 11.
Among embodiments, fig. 15 is a schematic structural diagram of a control-forbidden terminal for face recognition in another embodiments according to the present invention, as shown in fig. 15, the control terminal further includes:
the selection module 151 is configured to obtain a th load state of the gate 11 and a third load state of the third gate 31, and determine the th processing gate according to a load balancing algorithm, where the th gate 11 and the third gate 31 both belong to a th gate group, and the third gate 31 stores an th sub-face library;
the forwarding module 141 is further configured to send the face feature value to the th processing gate, where the th processing gate performs face comparison on the face feature value and the th sub-face library to obtain a th comparison result.
Among embodiments, fig. 16 is a schematic structural diagram of a second control-forbidden terminal for face recognition in embodiments according to the present invention, and as shown in fig. 16, the control terminal further includes:
the second selecting module 161 is configured to obtain a second load state of the second gate 12 and a fourth load state of the fourth gate 41, and determine a second processing gate according to a load balancing algorithm, where the second gate 12 and the fourth gate 41 both belong to a second gate group, and the fourth gate 41 stores a second child face library;
the forwarding module 141 is further configured to send the face feature value to a second processing gate, where the second processing gate 12 compares the face feature value with a second sub-face library to obtain a second comparison result.
The control end receives the face characteristic value detected and calculated by the gate 11 and sends the face characteristic value to the second gate 12 for face comparison, the sending of the face characteristic value and the feedback of the comparison result are local communication in aspect, the communication efficiency is high, in addition, aspect, the two gates perform parallel calculation, the calculation efficiency is improved, the utilization rate of the gate CPU is also improved, the control end is responsible for forwarding the face characteristic value, the processing load of the forwarding process of the gate 11 is reduced, when the number of gates needing forwarding is large, the processing efficiency can be improved by steps, and therefore the efficiency of the gate forbidden face recognition process is improved.
The specific limitations of the gate and the control end can be referred to the limitations of the face recognition forbidden processing method, which are not described herein again.
In specific embodiments, fig. 17 is a schematic view of an application scenario of the human face recognition prohibited processing method in another embodiments of the present invention, as shown in fig. 17, when the total number of the human face libraries is 60 thousands, and the size of the human face library that can be stored by a single machine is 15 thousands, the gates in each group can be divided into 4 groups, each gate in each group can store 15 thousands of human faces, and the human faces stored by each gate in the group are the same.
Fig. 18 is a flowchart of face recognition forbidden processing methods in another embodiments according to the present invention, and as shown in fig. 18, the load balancing face comparison process in this embodiment is as follows:
the gate calculates face characteristic values after the gate 11 captures faces, the face characteristic values are sent to the control end 61, the control end 61 randomly selects devices according to loads of the current devices in the gate groups to compare the characteristic values and feeds the characteristic values back to the gate , the gate 11 judges switch processing according to comparison results, after gates feed back hit results, the gate is opened , and if the feedback results are not hit, the gate is not opened .
In embodiments, computer devices are provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
the gate acquires a face feature value, the face feature value is sent to the gate , the gate stores a sub-face library of the face library, the gate stores a second sub-face library of the face library, the gate compares the face feature value with the sub-face library to obtain a comparison result, the gate receives a second comparison result of the gate, wherein the second comparison result is a result obtained by the gate comparing the face feature value with the second sub-face library, and the gate performs switching according to the comparison result and the second comparison result.
In of these embodiments, the processor when executing the computer program further performs the steps of:
the gate obtains a 0 th load state of a gate and a third load state of a third gate, a gate 1 processing is determined according to a load balancing algorithm, wherein the gate and the third gate both belong to a gate , the third gate stores a th sub-face library, and the gate obtains a th comparison result, wherein the gate compares the face feature value with the th sub-face library to obtain a th comparison result.
In of these embodiments, the processor when executing the computer program further performs the steps of:
, the gate acquires a second load state of the second gate and a fourth load state of the fourth gate, and determines a second processing gate according to a load balancing algorithm, wherein the second gate and the fourth gate both belong to a second gate set, the fourth gate stores a second sub-face library, and the second comparison result is a result obtained by the second processing gate comparing the face feature value with the second sub-face library.
In of these embodiments, the processor when executing the computer program further performs the steps of:
the control end receives the face characteristic value sent by the th gate, the control end sends the face characteristic value to the second gate, the th gate stores the th sub-face library of the face library, and the second gate stores the second sub-face library of the face library.
In of these embodiments, the processor when executing the computer program further performs the steps of:
the control end obtains a load state of an gate and a third load state of a third gate, a processing gate is determined according to a load balancing algorithm, wherein the gate and the third gate both belong to a gate set, the third gate stores a sub-face library, and the control end sends a face feature value to an processing gate.
In of these embodiments, the processor when executing the computer program further performs the steps of:
the control end obtains a second load state of a second gate and a fourth load state of a fourth gate, a second processing gate is determined according to a load balancing algorithm, wherein the second gate and the fourth gate both belong to a second gate set, the fourth gate stores a second sub-face library, and the control end sends the face characteristic value to the second processing gate.
The computer equipment detects a human face through the th gate and calculates a human face characteristic value, the human face characteristic value is sent to the second gate, the th gate compares the human face characteristic value with the th sub-human face library stored in the th gate, the second gate compares the human face characteristic value with the second sub-human face library stored in the second gate and feeds back a comparison result to the th gate, the two gates calculate in parallel, the calculation efficiency is improved, the utilization rate of a gate CPU is also improved, meanwhile, the sending of the human face characteristic value and the feedback of the result are local communication, the communication efficiency is high, and therefore the efficiency of a gate-forbidden human face recognition process is improved.
readable storage medium having stored thereon an executable program which when executed by a processor implements the face recognition disable processing method described above.
The readable storage medium detects a human face through the th gate, calculates a human face characteristic value, sends the human face characteristic value to the second gate, compares the human face characteristic value with the th sub-face library stored in the th gate at the th gate, compares the human face characteristic value with the second sub-face library stored in the second gate and feeds back the comparison result to the th gate, the two gates calculate in parallel, the calculation efficiency is improved, the utilization rate of a gate CPU is also improved, meanwhile, the sending of the human face characteristic value and the feedback of the result are local communication, the communication efficiency is high, and therefore the efficiency of the gate-forbidden face recognition process is improved.
It will be understood by those of ordinary skill in the art that all or a portion of the processes of the methods of the embodiments described above may be implemented by a computer program that may be stored in a non-volatile computer-readable storage medium, which when executed, may include the processes of the embodiments of the methods described above, wherein any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (16)
1, A face recognition processing prohibition method, which comprises the following steps:
the gate obtains a face feature value, and sends the face feature value to a second gate, wherein the gate stores a sub-face library of a face library, and the second gate stores a second sub-face library of the face library;
the gate compares the face feature value with the sub-face library to obtain a comparison result, and the gate receives a second comparison result of the second gate, wherein the second comparison result is obtained by the second gate comparing the face feature value with the second sub-face library;
the gate is switched according to the comparison result and the second comparison result.
2. The method of claim 1, wherein the gate comparing the face feature value to the sub-face library to obtain a comparison result comprises:
the gate acquires a th load state of the gate and a third load state of a third gate, and determines a th processing gate according to a load balancing algorithm, wherein the th gate and the third gate both belong to a th gate group, and the third gate stores the th sub-face library;
the gate obtains th comparison result, wherein the th processing gate compares the face feature value with the th sub-face library to obtain th comparison result.
3. The method according to claim 1 or 2, wherein the second comparison result is a result of the second gate comparing the face feature value with the second sub-face library, and comprises:
the gate acquires a second load state of the second gate and a fourth load state of a fourth gate, and determines a second processing gate according to a load balancing algorithm, wherein the second gate and the fourth gate both belong to a second gate group, the fourth gate stores the second child face library,
and the second comparison result is obtained by comparing the face characteristic value with the second sub-face library by the second processing gate.
4, A face recognition processing prohibition method, which comprises the following steps:
the control end receives a face feature value sent by an th gate, the control end sends the face feature value to a second gate, the th gate stores a th sub-face library of a face library, the th gate performs face comparison on the face feature value and the th sub-face library to obtain a th comparison result, the second gate stores a second sub-face library of the face library, the second gate performs face comparison on the face feature value and the second sub-face library to obtain a second comparison result, and the second gate sends the second comparison result to the th gate.
5. The method of claim 4, wherein the gate performing the face comparison of the face feature value with the sub-face library to obtain a comparison result comprises:
the control end obtains a load state of the gate and a third load state of a third gate, and determines a processing gate according to a load balancing algorithm, wherein the gate and the third gate both belong to a gate group, the third gate stores the sub-face library, the control end sends the face feature value to the processing gate, and the processing gate performs face comparison on the face feature value and the sub-face library to obtain an comparison result.
6. The method of any of claim 4 or 5, wherein the second gate comparing the face feature values with the second sub-face library to obtain a second comparison result comprises:
the control end obtains a second load state of the second gate and a fourth load state of a fourth gate, and determines a second processing gate according to a load balancing algorithm, wherein the second gate and the fourth gate both belong to a second gate set, the fourth gate stores the second sub-face library, the control end sends the face feature value to the second processing gate, and the second processing gate compares the face feature value with the second sub-face library to obtain a second comparison result.
7, A gate machine, characterized in that, the gate machine includes:
a capturing module for obtaining the face characteristic value,
the receiving and sending module is used for sending the face characteristic value to a second gate, wherein the th gate stores a th sub-face library of a face library, and the second gate stores a second sub-face library of the face library;
a comparison module for comparing the face feature value with the th sub-face library to obtain th comparison result,
a collecting module, configured to receive, by the th gate, a second comparison result of the second gate, where the second comparison result is a result obtained by the second gate comparing the face feature value with the second sub-face library;
and the gate interceptor performs switching according to the th comparison result and the second comparison result.
8. The gate of claim 7, further comprising:
an load module, configured to obtain a load state of the gate and a third load state of a third gate, and determine a processing gate according to a load balancing algorithm, where the gate and the third gate both belong to a gate group, and the third gate stores the sub-face library;
the gathering module is further configured to obtain th comparison results, wherein the th processing gate compares the face feature values with the th sub-face library to obtain th comparison results.
9. The gate of any of claim 7 or claim 8, further comprising:
and the second load module is used for acquiring a second load state of a second gate and a fourth load state of a fourth gate, and determining a second processing gate according to a load balancing algorithm, wherein the second gate and the fourth gate both belong to a second gate set, the fourth gate stores the second sub-face library, and the second comparison result is obtained by comparing the face feature value with the second sub-face library by the second processing gate.
10, control terminal, characterized in that, the control terminal includes:
a forwarding module, configured to receive a face feature value sent by an th gate, and send the face feature value to a second gate, where the th gate stores a th sub-face library of a face library, the th gate performs face comparison on the face feature value and the th sub-face library to obtain a th comparison result, the second gate stores a second sub-face library of the face library, the second gate performs face comparison on the face feature value and the second sub-face library to obtain a second comparison result, and the second gate sends the second comparison result to the th gate.
11. The control terminal of claim 10, further comprising:
an selecting module, configured to obtain a th load state of the gate and a third load state of a third gate, and determine a th processing gate according to a load balancing algorithm, where the th gate and the third gate both belong to a th gate group, and the third gate stores the th sub-face library;
the forwarding module is further configured to send the face feature value to the th processing gate, where the th processing gate performs face comparison on the face feature value and the th sub-face library to obtain a th comparison result.
12. The control terminal according to any of of claims 10 or 11, further comprising:
the second selection module is used for acquiring a second load state of the second gate and a fourth load state of a fourth gate, and determining a second processing gate according to a load balancing algorithm, wherein the second gate and the fourth gate both belong to a second gate set, and the fourth gate stores the second sub-face library;
the forwarding module is further configured to send the face feature value to the second processing gate, where the second processing gate compares the face feature value with the second sub-face library to obtain a second comparison result.
13, A face recognition disable system, the system includes gate and second gate:
the gate obtains a face feature value, and sends the face feature value to a second gate, wherein the gate stores a sub-face library of a face library, and the second gate stores a second sub-face library of the face library;
the gate compares the face feature value with the sub-face library to obtain a comparison result, the second gate compares the face feature value with the second sub-face library to obtain a second comparison result, and the second gate sends the second comparison result to the gate;
the gate is switched according to the comparison result and the second comparison result.
14, A face recognition forbidden system, which is characterized in that the system comprises a control terminal, a gate, a second gate:
, acquiring a face characteristic value and sending the face characteristic value to a control end;
the control end receives the face characteristic value sent by the th gate, the control end sends the face characteristic value to a second gate, the th gate stores a th sub-face library of a face library, and the second gate stores a second sub-face library of the face library;
the gate compares the face feature value with the sub-face library to obtain a comparison result, the second gate compares the face feature value with the second sub-face library to obtain a second comparison result, the second gate sends the second comparison result to the gate, and the gate performs switching according to the comparison result and the second comparison result.
15, computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method of any of claims 1-6 when executing the computer program.
16, computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of any of claims 1 to 6, wherein is defined.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910910550.XA CN110738770A (en) | 2019-09-25 | 2019-09-25 | Face recognition forbidden processing method, gate, control end and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910910550.XA CN110738770A (en) | 2019-09-25 | 2019-09-25 | Face recognition forbidden processing method, gate, control end and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110738770A true CN110738770A (en) | 2020-01-31 |
Family
ID=69269522
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910910550.XA Pending CN110738770A (en) | 2019-09-25 | 2019-09-25 | Face recognition forbidden processing method, gate, control end and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110738770A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110991431A (en) * | 2020-03-03 | 2020-04-10 | 支付宝(杭州)信息技术有限公司 | Face recognition method, device, equipment and storage medium |
| CN111489093A (en) * | 2020-04-15 | 2020-08-04 | 山东琅玡科技有限公司 | Novel labor management system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007172177A (en) * | 2005-12-20 | 2007-07-05 | Matsushita Electric Works Ltd | Gate management system |
| CN103679907A (en) * | 2014-01-08 | 2014-03-26 | 佛山科学技术学院 | Biological information identification method and biological information identification system applied to building talk-back |
| CN105046303A (en) * | 2015-08-03 | 2015-11-11 | 深圳市科锐奇科技有限公司 | Distributed data interaction based biological identification method and system |
| CN106710049A (en) * | 2017-01-06 | 2017-05-24 | 微美光速资本投资管理（北京）有限公司 | Holographic door control system and control method |
| CN108062499A (en) * | 2016-11-08 | 2018-05-22 | 中国移动通信有限公司研究院 | A kind of method of recognition of face, system and platform |
| CN109300204A (en) * | 2018-08-20 | 2019-02-01 | 新疆民航凯亚信息网络有限责任公司 | Self-service double-gate clearance safety check apparatus and its clearance safety inspection method |
| CN109756497A (en) * | 2018-12-29 | 2019-05-14 | 北京旷视科技有限公司 | Method and system, forwarding server and the readable storage medium storing program for executing of message are called in transmission |
| CN110111458A (en) * | 2019-04-16 | 2019-08-09 | 深圳市多度科技有限公司 | Realize method, system and door access machine that face is deployed to ensure effective monitoring and control of illegal activities |
-
2019
- 2019-09-25 CN CN201910910550.XA patent/CN110738770A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007172177A (en) * | 2005-12-20 | 2007-07-05 | Matsushita Electric Works Ltd | Gate management system |
| CN103679907A (en) * | 2014-01-08 | 2014-03-26 | 佛山科学技术学院 | Biological information identification method and biological information identification system applied to building talk-back |
| CN105046303A (en) * | 2015-08-03 | 2015-11-11 | 深圳市科锐奇科技有限公司 | Distributed data interaction based biological identification method and system |
| CN108062499A (en) * | 2016-11-08 | 2018-05-22 | 中国移动通信有限公司研究院 | A kind of method of recognition of face, system and platform |
| CN106710049A (en) * | 2017-01-06 | 2017-05-24 | 微美光速资本投资管理（北京）有限公司 | Holographic door control system and control method |
| CN109300204A (en) * | 2018-08-20 | 2019-02-01 | 新疆民航凯亚信息网络有限责任公司 | Self-service double-gate clearance safety check apparatus and its clearance safety inspection method |
| CN109756497A (en) * | 2018-12-29 | 2019-05-14 | 北京旷视科技有限公司 | Method and system, forwarding server and the readable storage medium storing program for executing of message are called in transmission |
| CN110111458A (en) * | 2019-04-16 | 2019-08-09 | 深圳市多度科技有限公司 | Realize method, system and door access machine that face is deployed to ensure effective monitoring and control of illegal activities |
Non-Patent Citations (1)
| Title |
|---|
| 陆赛杰: "《电子世界》", 23 December 2018 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110991431A (en) * | 2020-03-03 | 2020-04-10 | 支付宝(杭州)信息技术有限公司 | Face recognition method, device, equipment and storage medium |
| CN111489093A (en) * | 2020-04-15 | 2020-08-04 | 山东琅玡科技有限公司 | Novel labor management system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110096996B (en) | Biological information identification method, device, terminal, system and storage medium | |
| TWI756687B (en) | Coding model training method and device for preventing privacy data leakage | |
| CN108985162B (en) | Target real-time tracking method and device, computer equipment and storage medium | |
| CN107333292B (en) | A kind of SIM card state judging method and device | |
| CN109889538B (en) | User abnormal behavior detection method and system | |
| CN107623754B (en) | WiFi acquisition system and method based on authenticity MAC identification | |
| CN110738770A (en) | Face recognition forbidden processing method, gate, control end and system | |
| CN108574668B (en) | DDoS attack flow peak value prediction method based on machine learning | |
| CN109195011B (en) | Video processing method, device, equipment and storage medium | |
| CN105654095B (en) | feature selection method and device | |
| CN111241938A (en) | Face recognition method and device based on image verification and computer equipment | |
| CN108471413B (en) | Edge network security admittance defense system and method thereof | |
| CN113691783A (en) | Converter station video monitoring method, system, device and computer equipment | |
| CN106604253B (en) | Wireless networking method and system | |
| CN113934545A (en) | Video data scheduling method, system, electronic equipment and readable medium | |
| CN108647106A (en) | Using abnormality eliminating method, storage medium and computer equipment | |
| US11570187B1 (en) | Detection of cyberattacks and operational issues of internet of things devices | |
| JP7434690B2 (en) | Method, apparatus, system, device and computer program for detecting user data of user equipment UE | |
| CN106886746A (en) | A kind of recognition methods and back-end server | |
| CN113691483B (en) | Method, device and equipment for detecting abnormal user equipment and storage medium | |
| CN116385857B (en) | Calculation power distribution method based on AI intelligent scheduling | |
| CN115578765A (en) | Target identification method, device, system and computer readable storage medium | |
| CN113628248B (en) | Pedestrian residence time length determining method and device and computer readable storage medium | |
| KR101587845B1 (en) | Method for detecting distributed denial of services attack apparatus thereto | |
| CN101599969B (en) | Method for extracting worm features and device thereof |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200131 |