CN111223305B - Method and device for detecting illegal card involvement - Google Patents

Method and device for detecting illegal card involvement Download PDF

Info

Publication number
CN111223305B
CN111223305B CN201911353595.8A CN201911353595A CN111223305B CN 111223305 B CN111223305 B CN 111223305B CN 201911353595 A CN201911353595 A CN 201911353595A CN 111223305 B CN111223305 B CN 111223305B
Authority
CN
China
Prior art keywords
license plate
passing
detected
vehicle
probability
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201911353595.8A
Other languages
Chinese (zh)
Other versions
CN111223305A (en
Inventor
展昭臣
周坤
余思
王工艺
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chengdu Huawei Technology Co Ltd
Original Assignee
Huawei Technologies Co Ltd
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 Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CN201911353595.8A priority Critical patent/CN111223305B/en
Publication of CN111223305A publication Critical patent/CN111223305A/en
Application granted granted Critical
Publication of CN111223305B publication Critical patent/CN111223305B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
    • G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C5/00—Registering or indicating the working of vehicles
    • G07C5/08—Registering or indicating performance data other than driving, working, idle, or waiting time, with or without registering driving, working, idle or waiting time
    • G07C5/0808—Diagnosing performance data

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The embodiment of the application provides a method and a device for detecting a card-related violation, wherein when detecting whether a certain to-be-detected license plate has the card-related violation, the confidence coefficient of each card gate in M card gates through which the to-be-detected license plate passes is obtained, and the confidence coefficient of the license plate in each vehicle-passing record in the vehicle-passing records of the to-be-detected license plate passing through the M card gates is obtained; and determining a detection result of the license plate to be detected according to the confidence coefficient of each gate and the confidence coefficient of the license plate in each vehicle passing record, so that the lower probability of the license plate to be detected having a card-related violation due to wrong gate position labeling can be avoided by calculating the gate confidence coefficient of each gate, and the lower probability of the license plate to be detected having a card-related violation due to wrong license plate recognition can be avoided by calculating the confidence coefficient of the license plate in each vehicle passing record, thereby effectively improving the accuracy of the probability of the license plate to be detected having a card-related violation.

Description

Method and device for detecting illegal card-related law
Technical Field
The application relates to the technical field of terminals, in particular to a method and a device for detecting illegal card-related law.
Background
The vehicle number plate is the 'ID card' of the vehicle and is an important information for distinguishing other motor vehicles. Along with the restrictive growth of high insurance quantity of motor vehicles, especially the stricter index control and forbidden measures for passenger cars, the illegal behaviors of counterfeiting, changing and using motor vehicle license plates and the like are increasingly prominent. Common card-related violations include: blocking, fouling, fake cards, and the like. The fake plate refers to the number plate, the model and the color of a real plate vehicle counterfeited and illegally copied by lawless persons, so that the surface of the vehicle which is smuggled, assembled, scrapped and stolen is covered with a legal coat, and the fake plate vehicle is called a clone vehicle, which is called a fake plate for short.
In the prior art, when illegal card-related detection is performed, illegal card-related detection is performed on a license plate by a method based on space-time contradiction. When the method based on the space-time contradiction is used for detection, the illegal detection result of the card-related law may be wrong due to wrong license plate identification or wrong bayonet coordinate calibration. Therefore, the accuracy of the detection result is lower by adopting the existing illegal card-related detection method.
Disclosure of Invention
The embodiment of the application provides a method and a device for detecting illegal card-related law, and the accuracy of a detection result is improved when illegal card-related law detection is carried out.
In a first aspect, an embodiment of the present application provides a method for detecting a violation of a card, where the method for detecting a violation of a card may include:
acquiring the confidence coefficient of each checkpoint in M checkpoints through which the license plate to be detected passes, and the confidence coefficient of the license plate in each passing record in the passing records of the license plate to be detected passing through the M checkpoints; wherein M is an integer greater than or equal to 2.
And determining the detection result of the license plate to be detected according to the checkpoint confidence coefficient of each checkpoint and the confidence coefficient of the license plate in each vehicle passing record.
It can be seen that, different from the prior art, the method for detecting the illegal involvement of the license plates provided in the embodiment of the application detects whether the illegal involvement of a certain license plate to be detected exists according to the confidence level of each card slot and the confidence level of the license plate in each vehicle passing record, so that the lower probability of the illegal involvement of the license plate to be detected caused by the wrong labeling of the card slot position can be avoided by calculating the confidence level of each card slot, the lower probability of the illegal involvement of the license plate to be detected caused by the wrong recognition of the license plate can be avoided by calculating the confidence level of the license plate in each vehicle passing record, and the accuracy of the probability of the illegal involvement of the license plate to be detected is effectively improved.
In a possible implementation manner, determining a detection result of a license plate to be detected according to a checkpoint confidence of each checkpoint and a confidence of the license plate in each vehicle passing record may include:
determining the number of vehicle passing records meeting preset conditions in the vehicle passing records with time-space contradictions according to the checkpoint confidence coefficient of each checkpoint and the confidence coefficient of the license plate in each vehicle passing record; determining the probability of the license plate to be detected that the license plate is illegal by means of the license plate related law according to the number of the vehicle passing records with the time-space contradiction and the number of the vehicle passing records meeting the preset conditions; the preset condition is used for indicating that the reason of the space-time contradiction, namely the position marking error of the non-bayonet is caused and the recognition of the non-license plate is wrong; the vehicle passing records with the time-space contradiction are determined in the vehicle passing records for determining the M checkpoints according to the checkpoint marks in each vehicle passing record, the time for passing the checkpoints and the position information of the checkpoints corresponding to the checkpoint marks.
Therefore, the probability of the license plate violation of the license plate to be detected can be determined according to the number of the vehicle-passing records with the space-time contradiction, the non-bayonet position marking errors caused by the space-time contradiction and the number of the vehicle-passing records with the non-license plate recognition errors, so that the space-time contradiction caused by the license plate to be detected due to the bayonet position marking errors and the license plate recognition errors can be eliminated, and the accuracy of the probability of the license plate violation of the license plate to be detected is effectively improved.
In a possible implementation manner, obtaining a bayonet confidence of each bayonet from among M bayonets through which a license plate to be detected passes may include:
obtaining the vehicle passing record of a license plate to be detected passing through a first gate and the vehicle passing record of the license plate to be detected passing through a second gate in the time neighborhood of the license plate to be detected passing through the first gate in the M gates; the first bayonet is any one of the M bayonets, and the second bayonet is different from the first bayonet; the bayonet confidence coefficient of the first bayonet is determined according to the vehicle passing record of the first bayonet and the vehicle passing record of the second bayonet, so that the bayonet confidence coefficient of the first bayonet can be calculated through the vehicle passing record of the license plate to be detected passing the second bayonet in the time neighbor of the license plate to be detected passing the first bayonet, and the accuracy of the bayonet confidence coefficient of the first bayonet is improved.
In one possible implementation manner, determining a gate confidence of the first gate according to the vehicle passing record of the first gate and the vehicle passing record of the second gate may include:
and determining the number of the vehicle passing records with the time-space contradiction in the vehicle passing records passing through the first bayonet according to the vehicle passing records passing through the first bayonet and the vehicle passing records passing through the second bayonet.
And determining the probability of the time-space contradiction of the vehicle passing records at the first gate according to the number of the vehicle passing records passing through the first gate and the number of the vehicle passing records with the time-space contradiction.
And determining the checkpoint confidence of the first checkpoint according to the probability of the time-space contradiction of the vehicle-passing records at the first checkpoint and the number of the vehicle-passing records with the time-space contradiction.
Therefore, the confidence coefficient of each bayonet can be understood as the accuracy of the coordinate marked by each bayonet by calculating the confidence coefficient of each bayonet, so that whether a certain license plate to be detected is illegal by card interference or not is detected subsequently, the confidence coefficient of each bayonet through which the license plate to be detected passes is fully considered, the low accuracy of the detection result caused by the wrong calibration of the coordinates of the bayonet in the vehicle passing record can be avoided, and the accuracy of the detection result is effectively improved.
In a possible implementation manner, in obtaining the vehicle passing records that the license plate to be detected passes through the M checkpoints, the confidence of the license plate in each vehicle passing record may include:
and determining the probability that the license plate to be detected accords with the motor vehicle license plate rule and the probability that the license plate to be detected has registration information.
And determining a first probability that the license plate to be detected in each vehicle passing record is an even current license plate error and a second probability that the license plate to be detected in each vehicle passing record is a frequent current license plate error according to the vehicle passing records of the M checkpoints.
And determining the confidence coefficient of the license plate in each passing record according to the probability that the license plate to be detected accords with the motor vehicle license plate coding rule, the probability with the registration information, the first probability and the second probability.
It can be seen that, in the vehicle passing records that the license plates to be detected pass through the M checkpoints, the confidence coefficient of the license plate in each vehicle passing record is calculated, and the confidence coefficient of the license plate in each vehicle passing record can be understood as the accuracy of license plate identification in the vehicle passing record, so that whether a certain license plate to be detected is subjected to license plate interference violation is detected subsequently, the confidence coefficient of the license plate in each vehicle passing record is fully considered, and thus the condition that the accuracy of the detection result is low due to the fact that the license plate in the vehicle passing record is identified incorrectly can be avoided, and the accuracy of the detection result is effectively improved.
In a possible implementation manner, determining the confidence level of the license plate in each vehicle passing record according to the probability that the license plate to be detected meets the motor vehicle license plate coding rule, the probability that the license plate has the registration information, the first probability, and the second probability may include:
and calculating a first product of a first weight corresponding to the even-present license plate error and a first probability, a second product of a second weight corresponding to the frequent license plate error and a second probability, and a third product of the probability according with the motor vehicle license plate coding rule and the probability with the registration information.
Calculating a fourth product of the sum of the first product and the second product and the third product; and determining the confidence coefficient of the license plate in each vehicle passing record according to the fourth product.
It can be seen that when the confidence level of the license plate in each passing record is calculated, the confidence level of the license plate in the passing record based on the semantic meaning of the license plate (determined by the probability that the to-be-detected license plate meets the motor vehicle license plate coding rule and the probability with the registration information) and the confidence level of the license plate based on the semantic meaning of the track (determined by the first probability that the to-be-detected license plate in each passing record is an occasional license plate error and the second probability that the to-be-detected license plate in each passing record is a frequent license plate error) need to be determined respectively, and the confidence level of the license plate in each passing record is determined by combining the confidence level of the license plate based on the semantic meaning of the license plate and the confidence level of the license plate based on the semantic meaning of the track, so that the accuracy of the confidence level of the license plate in the obtained passing record can be improved.
In a possible implementation manner, determining, according to the vehicle-passing records of the M checkpoints, a first probability that the license plate to be detected in each vehicle-passing record is an even license plate error may include:
obtaining a first vehicle passing record of a license plate to be detected passing through a third gate and a second vehicle passing record of the license plate to be detected passing through a fourth gate in the space-time neighbor of the license plate to be detected passing through the third gate in the M gates; determining a first probability according to the license plate identification in the first vehicle passing record and the license plate identification in the second vehicle passing record; the third bayonet is any one of the M bayonets, and the fourth bayonet is different from the third bayonet, so that the first probability that the license plate to be detected in the vehicle passing record is an accidental license plate error can be obtained, and a basis is provided for subsequently calculating the confidence coefficient of the license plate in the vehicle passing record.
In a possible implementation manner, determining the first probability according to the license plate identifier in the first vehicle passing record and the license plate identifier in the second vehicle passing record may include:
if the license plate identifier in the first passing record is the license plate to be detected, the license plate identifier in the second passing record is different from the license plate to be detected and is a similar license plate of the license plate to be detected, determining that the first probability is greater than 0; and if the license plate identifier in the first vehicle passing record is the license plate to be detected, and the license plate identifier in the second vehicle passing record is the same as the license plate to be detected, determining that the first probability is 0.
In a possible implementation manner, the passing record includes a gate identifier and time of passing through the gate, and the second probability that the license plate to be detected in each passing record is a frequent license plate error is determined according to the passing records of M gates, which may include:
according to the bayonet identification and the time of passing through the bayonet of each vehicle passing record in the M bayonets and the position information of the bayonet corresponding to the bayonet identification, clustering the vehicle passing records of the M bayonets; and determining a second probability according to the number of the cluster sets obtained by the clustering processing, so that the second probability that the license plate to be detected in the vehicle passing record is a frequent vehicle license plate error can be obtained, and a basis is provided for subsequently calculating the confidence coefficient of the license plate in the vehicle passing record.
In a possible implementation, according to M bayonets, each bayonet identification of passing the car record and the time of passing through the bayonet to and the positional information of the bayonet corresponding to the bayonet identification, the car record of passing of M bayonets is clustered, and may include:
sequencing the vehicle passing records of the M checkpoints according to the time sequence, and taking the first vehicle passing record as a first element in a first cluster set; calculating N first moving speeds of the license plate to be detected according to the time of passing through the bayonet in the ith vehicle passing record, the position information of the bayonet corresponding to the bayonet identification in the ith vehicle passing record, the time of passing through the bayonet in the last vehicle passing record in each of the N existing cluster sets and the position information of the bayonet corresponding to the bayonet identification in the last vehicle passing record; the number of the cluster sets of all the vehicle-passing records before the ith vehicle-passing record is N, and i is greater than or equal to 2; comparing the N first moving speeds with a preset threshold value respectively, and if a second moving speed in the N first moving speeds is smaller than or equal to the preset threshold value, taking the ith vehicle passing record as the last element in a clustering set corresponding to the second moving speed; if the N first moving speeds are all larger than the preset threshold value, the ith vehicle passing record is used as a first element in a new cluster set, so that the number of the cluster sets can be obtained through clustering processing, a second probability that the license plate to be detected in the vehicle passing record is a frequent license plate error can be determined through the number of the cluster sets, and a basis is provided for subsequently calculating the confidence coefficient of the license plate in the vehicle passing record.
In a second aspect, an embodiment of the present application further provides a device for detecting a violation of a card, where the device for detecting a violation of a card may include:
the processing unit is used for acquiring the confidence coefficient of each gate in M gates through which the license plate to be detected passes, and the confidence coefficient of the license plate in each passing record in the passing records of the license plate to be detected passing through the M gates; wherein M is an integer greater than or equal to 2.
And the determining unit is used for determining the detection result of the license plate to be detected according to the confidence coefficient of each gate and the confidence coefficient of the license plate in each vehicle passing record.
In a possible implementation manner, the determining unit is specifically configured to determine, according to a checkpoint confidence of each checkpoint and a confidence of a license plate in each vehicle passing record, the number of vehicle passing records meeting a preset condition in the vehicle passing records with a time-space contradiction; determining the probability of the license plate to be detected that the license plate is illegal by means of the license plate related law according to the number of the vehicle passing records with the time-space contradiction and the number of the vehicle passing records meeting the preset conditions; the preset condition is used for indicating that the reason of the space-time contradiction, namely the position marking error of the non-bayonet is caused and the recognition of the non-license plate is wrong; the vehicle passing records with the time-space contradiction are determined in the vehicle passing records for determining the M checkpoints according to the checkpoint marks in each vehicle passing record, the time for passing the checkpoints and the position information of the checkpoints corresponding to the checkpoint marks.
In a possible implementation manner, the processing unit is specifically configured to obtain, from the M gates, a vehicle passing record that a license plate to be detected passes through the first gate, and a vehicle passing record that the license plate to be detected passes through the second gate within a time neighborhood of the license plate to be detected passing through the first gate; determining the bayonet confidence of the first bayonet according to the vehicle passing record passing through the first bayonet and the vehicle passing record passing through the second bayonet; wherein, first bayonet socket is any one bayonet socket in M bayonet sockets, and the second bayonet socket is different with first bayonet socket.
In a possible implementation manner, the processing unit is specifically configured to determine, from the vehicle-passing records passing through the first gate and the vehicle-passing records passing through the second gate, the number of vehicle-passing records in which the time-space conflict exists in the vehicle-passing records passing through the first gate; determining the probability of the time-space contradiction of the vehicle passing records at the first gate according to the number of the vehicle passing records passing through the first gate and the number of the vehicle passing records with the time-space contradiction; and determining the checkpoint confidence of the first checkpoint according to the probability of the time-space contradiction of the vehicle-passing records at the first checkpoint and the number of the vehicle-passing records with the time-space contradiction.
In a possible implementation manner, the processing unit is specifically configured to determine a probability that the license plate to be detected meets the license plate rule of the motor vehicle and a probability that the license plate has registration information; determining a first probability that the license plate to be detected in each vehicle passing record is an even current license plate error and a second probability that the license plate to be detected in each vehicle passing record is a frequent current license plate error according to the vehicle passing records of the M checkpoints; and determining the confidence coefficient of the license plate in each passing record according to the probability that the license plate to be detected accords with the motor vehicle license plate coding rule, the probability with the registration information, the first probability and the second probability.
In a possible implementation manner, the processing unit is specifically configured to calculate a first product of a first weight corresponding to the even license plate error and a first probability, a second product of a second weight corresponding to the frequent license plate error and a second probability, and a third product of a probability according with the motor vehicle license plate coding rule and a probability with the registration information; and calculating a fourth product of the sum of the first product and the second product and the third product; and determining the confidence of the license plate in each vehicle passing record according to the fourth product.
In a possible implementation manner, the processing unit is specifically configured to obtain a first vehicle passing record that a license plate to be detected passes through a third gate and a second vehicle passing record that the license plate to be detected passes through a fourth gate in a space-time neighborhood passing through the third gate from among the M gates; determining a first probability according to the license plate identification in the first vehicle passing record and the license plate identification in the second vehicle passing record; the third bayonet is any one of the M bayonets, and the fourth bayonet is different from the third bayonet.
In a possible implementation manner, the processing unit is specifically configured to determine that the first probability is greater than 0 if the license plate identifier in the first passing record is a license plate to be detected, and the license plate identifier in the second passing record is different from the license plate to be detected and is a similar license plate of the license plate to be detected; and if the license plate identifier in the first vehicle passing record is the license plate to be detected, and the license plate identifier in the second vehicle passing record is the same as the license plate to be detected, determining that the first probability is 0.
In a possible implementation manner, the vehicle passing records include a bayonet identification and time of passing through the bayonet, and the processing unit is specifically configured to perform clustering processing on the vehicle passing records of the M bayonets according to the bayonet identification and the time of passing through the bayonet of each vehicle passing record in the M bayonets and position information of the bayonet corresponding to the bayonet identification; and determining a second probability according to the number of the cluster sets obtained by the clustering.
In a possible implementation manner, the processing unit is specifically configured to sort the vehicle passing records of the M checkpoints according to a time sequence, and use a first vehicle passing record as a first element in the first cluster set; calculating N first moving speeds of the license plate to be detected according to the time of passing through the bayonet in the ith vehicle passing record, the position information of the bayonet corresponding to the bayonet identification in the ith vehicle passing record, the time of passing through the bayonet in the last vehicle passing record in each of the N existing cluster sets and the position information of the bayonet corresponding to the bayonet identification in the last vehicle passing record; comparing the N first moving speeds with a preset threshold value respectively, and if a second moving speed in the N first moving speeds is smaller than or equal to the preset threshold value, taking the ith vehicle passing record as the last element in a clustering set corresponding to the second moving speed; if the N first moving speeds are all larger than a preset threshold value, taking the ith vehicle passing record as a first element in a new cluster set; and N is the number of the cluster sets in which all the vehicle passing records before the ith vehicle passing record are located, and i is greater than or equal to 2.
In a third aspect, the present application further provides an apparatus for detecting illegal card involvement, which may include at least one processor and at least one memory, wherein,
the memory is to store program instructions;
the processor is configured to execute the program instructions in the memory, so as to enable the device for detecting a illegal card-related law to implement the method for detecting a illegal card-related law in any one of the possible implementation manners of the first aspect.
In a fourth aspect, an embodiment of the present application further provides a chip, where a computer program is stored on the chip, and when the computer program is executed by a processor, the method for detecting a illegal card involvement recited in any one of the foregoing possible implementation manners of the first aspect is executed.
In a fifth aspect, an embodiment of the present application further provides a computer storage medium, which includes instructions that, when executed by one or more processors, cause a device for detecting a violation of a card-related law to perform the method for detecting a violation of a card-related law described in any one of the foregoing possible implementations of the first aspect.
The embodiment of the application provides a method and a device for detecting card-related violation, wherein when detecting whether a certain to-be-detected license plate has card-related violation, the confidence of each card slot in M card slots through which the to-be-detected license plate passes is obtained, and the confidence of the license plate in each vehicle-passing record in the vehicle-passing records of the to-be-detected license plate passing through the M card slots is obtained; and determining the detection result of the license plate to be detected according to the checkpoint confidence coefficient of each checkpoint and the confidence coefficient of the license plate in each vehicle passing record. It can be seen that, different from the prior art, the method for detecting the illegal involvement of the license plates provided in the embodiment of the application detects whether the illegal involvement of a certain license plate to be detected exists according to the confidence level of each card slot and the confidence level of the license plate in each vehicle passing record, so that the lower probability of the illegal involvement of the license plate to be detected caused by the wrong labeling of the card slot position can be avoided by calculating the confidence level of each card slot, the lower probability of the illegal involvement of the license plate to be detected caused by the wrong recognition of the license plate can be avoided by calculating the confidence level of the license plate in each vehicle passing record, and the accuracy of the probability of the illegal involvement of the license plate to be detected is effectively improved.
Drawings
Fig. 1 is a schematic diagram of a possible application scenario provided in an embodiment of the present application;
fig. 2 is a schematic flow chart of a method for detecting a card-related violation according to an embodiment of the present application;
FIG. 3 is a diagram of a system architecture according to an embodiment of the present application;
fig. 4 is a schematic flowchart of a method for detecting a card-related violation according to an embodiment of the present application;
FIG. 5 is a schematic diagram illustrating the existence of spatiotemporal contradiction according to the embodiment of the present application;
FIG. 6 is a schematic diagram of an occasional license plate error provided in accordance with an embodiment of the present application;
FIG. 7 is a schematic diagram of a frequent license plate error provided by an embodiment of the present application;
fig. 8 is a schematic structural diagram of a device for detecting a card-related violation according to an embodiment of the present application;
fig. 9 is a schematic structural diagram of another device for detecting a card-related violation according to an embodiment of the present application.
Detailed Description
The embodiment of the application can be applied to the field of illegal card-related detection, and can also be applied to the field of portrait detection in an expanded manner, and when the embodiment is applied to the field of portrait detection, people (similar to the license plate in the application) and collection equipment (similar to the card slot in the application) need to be distinguished, or other similar fields need to be distinguished, and the embodiment of the application is not limited herein.
Fig. 1 is a schematic view of a possible application scenario provided in an embodiment of the present application, for example, please refer to fig. 1, when a vehicle passes through a gate (i.e., a monitoring system of a road traffic security gate) arranged on a road, the gate sends a vehicle passing record recorded by the gate to an electronic device, where the vehicle passing record includes a license plate identifier, a gate identifier, and time of passing through the gate, so that the electronic device performs illegal card passing detection based on the vehicle passing record. For example, the electronic device may be a server or a terminal. Under the normal condition, the electronic equipment adopts a space-time contradiction method to carry out illegal card-related detection based on the acquired vehicle-passing record, so as to obtain illegal card-related detection results. However, when the time-space contradiction method is adopted for detection, the card identification in the vehicle-passing record is wrong, and/or the coordinate (i.e., longitude and latitude) calibration of the card port in the vehicle-passing record is wrong, so that the illegal card-related detection result is wrong, and the accuracy of the detection result is low. The time-space contradiction means that the same moving object moves from one place to another at a speed exceeding the common knowledge, and if a certain license plate moves from one place to another at a speed exceeding the common knowledge, the license plate has the time-space contradiction.
It can be seen that, when the spatio-temporal contradiction method is adopted for detection, the accuracy of the detection result is lower just because the license plate in the vehicle passing record is identified incorrectly and/or the bayonet coordinate in the vehicle passing record is calibrated incorrectly. In order to improve the accuracy of the detection result, when the illegal card-related detection is carried out, two aspects of consideration can be tried, on the first aspect, the accuracy of the identification of the license plate in the vehicle-passing record can be improved, so that the low accuracy of the detection result caused by the error identification of the license plate in the vehicle-passing record can be avoided; in the second aspect, when the coordinates of the bayonet are marked, the marked coordinates are accurate, so that the problem that the accuracy of a detection result is low due to calibration errors of the coordinates of the bayonet in the vehicle passing record is avoided. However, for the first aspect, when the license plate recognition technology is used to recognize the license plate in the vehicle passing record, the license plate may be affected by factors such as illumination and weather, and it is not ensured that the license plate in the vehicle passing record can be accurately recognized; for the second aspect, when coordinates of the bayonets are marked, because the coordinates are usually marked manually, it cannot be ensured that the coordinates marked by each bayonet are accurate. Therefore, in order to improve the accuracy of the detection result, it is obviously not feasible to improve the accuracy of the identification of the license plate in the vehicle passing record and the accuracy of the coordinates marked on each mount. Based on this, although the accuracy of the license plate recognition in the vehicle passing record and the accuracy of the coordinate marked by each gate cannot be improved, the vehicle passing record with higher accuracy of the license plate recognition can be determined in a plurality of vehicle passing records of gates through which vehicles pass based on the existing license plate recognition technology and gate marking technology, the gates with higher accuracy of some coordinates can be determined in the plurality of gates through which vehicles pass, and the illegal card passing detection is performed based on the process record with higher accuracy of the license plate recognition and the gates with higher accuracy of the coordinates, so that the accuracy of the detection result can be effectively improved. The gate is a short-term monitoring system of a road traffic public security gate, and refers to a road traffic field monitoring system for shooting, recording and processing all motor vehicles passing through the gate point by means of the gate point on a specific place on a road, such as a toll station, a traffic or public security inspection station and the like.
It can be understood that, if a process record with higher accuracy of license plate recognition is determined from a plurality of vehicle-passing records, the accuracy of license plate recognition in each vehicle-passing record needs to be calculated, for example, in this embodiment, it may be considered that the confidence of the license plate in each vehicle-passing record is used to represent the accuracy of license plate recognition in the vehicle-passing record; similarly, if a bayonet with higher accuracy of coordinates is determined from a plurality of bayonets passed by the vehicle, it is generally required to calculate the accuracy of the coordinate label of each bayonet from the plurality of bayonets passed by the vehicle, and for example, it may also be considered to use the bayonet confidence of each bayonet to represent the accuracy of the coordinate label of the bayonet. The confidence coefficient is used for representing the credibility of a certain object, and generally takes a decimal number with a value range of 0.0 to 1.0, and generally, the larger the value of the confidence coefficient is, the higher the credibility is represented; conversely, a smaller confidence value indicates a lower confidence.
In summary, for effectively improving the accuracy of the detection result, for example, please refer to fig. 2, where fig. 2 is a schematic flow chart of a method for detecting a card-related violation provided in the embodiment of the present application, so that when detecting whether a certain to-be-detected license plate has a card-related violation, S201 is first executed to obtain a checkpoint confidence of each checkpoint in M checkpoints through which the to-be-detected license plate passes, and a confidence of the license plate in each vehicle-passing record in vehicle-passing records in which the to-be-detected license plate passes through the M checkpoints; the gate confidence of each gate can be understood as the accuracy of the coordinates marked by each gate, the confidence of the license plate in each passing record can be understood as the accuracy of the license plate identification in the passing record, and then S202 is executed to determine the detection result of the license plate to be detected according to the gate confidence of each gate and the confidence of the license plate in each passing record. It can be seen that, different from the prior art, the method for detecting the illegal involvement of the license plates provided in the embodiment of the application detects whether the illegal involvement of a certain license plate to be detected exists according to the confidence level of the card slot of each card slot and the confidence level of the license plate in each vehicle passing record, and avoids that the probability of the illegal involvement of the license plate to be detected is low due to wrong labeling of the card slot position by calculating the confidence level of the card slot of each card slot, and avoids that the probability of the illegal involvement of the license plate to be detected is low due to wrong identification of the license plate by calculating the confidence level of the license plate in each vehicle passing record, so that the accuracy of the probability of the illegal involvement of the license plate to be detected is effectively improved. Wherein M is an integer greater than or equal to 2.
Wherein, a bayonet socket can correspond to there being the car record, also can correspond to there being many to pass the car record. Wherein, to each record of passing the car, can include license plate sign, bayonet socket sign and the time of passing the bayonet socket usually, this time of passing the bayonet socket is the license plate that this license plate sign instructed passes the time of the bayonet socket that this bayonet socket sign instructed.
It should be noted that, referring to fig. 3, a system architecture diagram corresponding to the method for detecting a card-related violation provided by the embodiment of the present application is shown, where fig. 3 is a system architecture diagram provided by the embodiment of the present application, when detecting whether a certain card to be detected has a card-related violation, two inputs, namely, a pass record of a gate and a gate information table (used for obtaining coordinates of the gate), may be obtained first, and a confidence level of the gate and a confidence level of a license plate in each pass record are calculated according to the pass record of the gate and the gate information table, and then a probability that the license plate is a card-related violation is determined according to the gate confidence level of each gate and the confidence level of the license plate in each pass record, so as to obtain a detection result. It can be understood that by adopting the method for detecting the illegal card-related law provided by the embodiment of the application, the probability that the illegal card-related law exists in each license plate can be detected. After the probability that each license plate has a license plate related violation is determined, the first K license plates with higher probability can be pushed to the staff, so that the staff can make a corresponding decision according to the detection result; certainly, the calculated probability of the license plate illegal by referring to the license plate, the confidence coefficient of the license plate and the pass confidence coefficient of the license plate can be stored in a detail database, so that a reference basis is provided for the subsequent calculation of the probability of the license plate illegal by referring to the license plate.
It is to be understood that, in the embodiments of the present application, "at least one" means one or more, "a plurality" means two or more. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists singly, A and B exist simultaneously, and B exists singly, wherein A and B can be singular or plural. In the description of the text of the present application, the character "/" generally indicates that the former and latter associated objects are in an "or" relationship.
Hereinafter, the method for detecting a card-related violation provided by the present application will be described in detail by way of a detailed example. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.
Fig. 4 is a schematic flowchart of a method for detecting a violation of a card in an embodiment of the present application, where the method for detecting a violation of a card may be executed by software and/or a hardware device, for example, the hardware device may be a device for detecting a violation of a card, and the device for detecting a violation of a card may be disposed in an electronic device. For example, referring to fig. 4, the method for detecting a illegal card-related law may include:
s401, obtaining the confidence coefficient of each gate in M gates through which the license plate to be detected passes, and the confidence coefficient of the license plate in each passing record in the passing records of the license plate to be detected passing through the M gates.
For example, in obtaining M gates through which a license plate to be detected passes, a gate confidence of each gate is taken as an example of any one of the M gates, for example, a first gate, when obtaining the gate confidence of the first gate, a vehicle passing record that the license plate to be detected passes through the first gate and a vehicle passing record that the license plate to be detected passes through a second gate in a time neighborhood of the time when the license plate to be detected passes through the first gate may be obtained first, and the second gate is different from the first gate; determining the number of the vehicle passing records with time-space contradiction in the vehicle passing records passing through the first bayonet according to the vehicle passing records passing through the first bayonet and the vehicle passing records passing through the second bayonet; determining the probability of the time-space contradiction of the vehicle passing records at the first gate according to the number of the vehicle passing records passing through the first gate and the number of the vehicle passing records with the time-space contradiction; and then determining the checkpoint confidence of the first checkpoint according to the probability of the time-space contradiction of the vehicle-passing records at the first checkpoint and the number of the vehicle-passing records with the time-space contradiction.
Wherein, the time neighbors of the license plate to be detected passing through the first gate can be understood as: in the time dimension, a preset time period before the time when the license plate to be detected passes through the first gate is within, and/or a preset time period after the time when the license plate to be detected passes through the first gate is within. In general, the preset time period may be any value within 10 minutes to 30 minutes, and of course, may also be any value within 31 minutes to 40 minutes, and may be specifically set according to actual needs. For example, when the time for the license plate to be detected to pass through the first gate is 10:10, and the preset time period is 20 minutes, the time for the license plate to be detected to pass through the first gate may be within a time period of 9:50-10:10, and/or within a time period of 10:10-10: 30. It can be understood that when the time period when the license plate to be detected passes through the first gate is within a time period of 9:50-10:30, the time period is within a time period of 9:50-10:10 compared with the time period when the license plate to be detected passes through the first gate, or the time period when the license plate to be detected passes through the first gate is within a time period of 10:10-10:30 compared with the time period when the license plate to be detected passes through the second gate, in general, in the time period when the license plate to be detected passes through the first gate is within a time period of 9:50-10:10, the collected passing records of the license plate to be detected through the second gate are more, and therefore, the accuracy of the gate confidence coefficient of the first gate calculated based on the collected passing records of the license plate to be detected passing through the first gate and the passing records of the license plate to be detected through the second gate is higher.
It should be noted that, in this embodiment of the application, when calculating the gate confidence of the first gate, it is necessary to first obtain the vehicle passing record that the license plate to be detected passes through the first gate, and the vehicle passing record that the license plate to be detected passes through the second gate in the time neighborhood of passing through the first gate, and the purpose of the record is as follows: in the vehicle passing records that the license plate to be detected passes through the first bayonet, each vehicle passing record comprises the bayonet identification of the first bayonet and the time of the license plate to pass through the first bayonet, and the license plate to be detected passes through the second bayonet in time neighborhood, each vehicle passing record comprises the bayonet identification of the second bayonet and the time of the license plate to pass through the second bayonet, so that the distance between the first bayonet and the second bayonet can be determined according to the position of the first bayonet and the position of the second bayonet, the interval time period is determined according to the time of the license plate to pass through the first bayonet and the time of the second bayonet, then the distance between the first bayonet and the second bayonet and the interval time period are calculated to calculate the moving speed of the license plate to be detected, and the number of the vehicle passing records with time-space contradiction in the vehicle passing records of the first bayonet is determined according to the moving speed of the license plate to be detected. For example, please refer to fig. 5, fig. 5 is a schematic diagram of the present application, which is provided in an embodiment of the present application and has a time-space conflict, in fig. 5, because the coordinates of the bayonet E are deviated, (i.e., the coordinates are calibrated incorrectly, and the correct position should be at the bayonet F) and pass through the bayonet E, most of the time passes through the bayonet a, the bayonet B, the bayonet C, and the bayonet D, and is unreachable in speed, so the time-space conflict exists in the process record of the bayonet E; most of the passing vehicles through the bayonet F pass through the bayonet A, the bayonet B, the bayonet C and the bayonet D within reasonable time, and the speed can be achieved, so that the time-space contradiction exists in the process record of the bayonet F. It can be understood that, when the distance between the first bayonet and the second bayonet is determined according to the position of the first bayonet and the position of the second bayonet, in order to accurately determine the distance between the first bayonet and the second bayonet, the position of the first bayonet and the position of the second bayonet may be obtained by means of a road network to determine the distance between the first bayonet and the second bayonet.
Therefore, after the number of the passing records with the time-space contradiction is determined in the passing records passing through the first gate, the probability of the time-space contradiction occurring in the first gate of the passing records can be determined according to the number of the passing records passing through the first gate and the number of the passing records with the time-space contradiction, and then the gate confidence coefficient of the first gate can be calculated according to the probability of the time-space contradiction occurring in the first gate of the passing records and the number of the passing records with the time-space contradiction. For example, in the embodiment of the present application, the method can be according to CFirst, theAnd f (y/x, y), calculating the bayonet confidence of the first bayonet. Wherein, CFirst, theOne bayonet represents the bayonet confidence coefficient of the first bayonet, x represents the number of vehicle passing records of which the license plate to be detected passes through the first bayonet, and y represents the number of vehicle passing records of which the existence of the space-time contradiction is determined in the vehicle passing records of the first bayonet.
It should be noted that, in the embodiment of the present application, when performing the illegal fording card detection, it is considered that in the prior art, the accuracy of the detection result is low due to the calibration error of the bayonet coordinates in the vehicle-passing record, and therefore, by calculating the confidence coefficient of each bayonet, the bayonet confidence coefficient of each bayonet can be understood as the accuracy of the coordinate labeled by each bayonet, so that whether a certain vehicle card to be detected has illegal fording card or not is subsequently detected, and the bayonet confidence coefficient of each bayonet through which the vehicle card to be detected passes is fully considered, so that the low accuracy of the detection result due to the calibration error of the bayonet coordinates in the vehicle-passing record can be avoided, and the accuracy of the detection result is effectively improved.
After describing in detail how to obtain the bayonet confidence of each of the M bayonets through which the license plate to be detected passes, as described above, since the accuracy of the detection result is low due to the wrong identification of the license plate in the vehicle passing record in the prior art, therefore, when the illegal detection of the related license plate is carried out, the license plate can be detected by calculating the vehicle passing records of the license plate to be detected passing through M checkpoints, the confidence of the license plate in each vehicle passing record can be understood as the accuracy of the license plate identification in the vehicle passing record, so that whether a certain license plate to be detected has a license plate-related violation or not is detected subsequently, the confidence coefficient of the license plate in each vehicle-passing record is fully considered, therefore, the lower accuracy of the detection result caused by the wrong identification of the license plate in the vehicle passing record can be avoided, and the accuracy of the detection result is effectively improved.
For example, when the confidence level of the license plate in each passing record is obtained in the passing records of the license plate to be detected passing through M checkpoints, the confidence level of the license plate in each passing record may include confidence levels in two aspects, one confidence level is the confidence level of the license plate based on the semantics of the license plate, and the other confidence level is the confidence level of the license plate based on the semantics of the track. In other words, when the confidence level of the license plate in each passing record is calculated, the confidence level of the license plate in each passing record based on the license plate semantics and the confidence level of the license plate based on the trajectory semantics need to be determined respectively, and the confidence level of the license plate in each passing record is determined by combining the confidence level of the license plate based on the license plate semantics and the confidence level of the license plate based on the trajectory semantics. It should be noted that, when determining the license plate confidence based on the license plate semantics and the license plate confidence based on the trajectory semantics in the vehicle passing record, the license plate confidence based on the license plate semantics in the vehicle passing record may be determined first, and then the license plate confidence based on the trajectory semantics in the vehicle passing record may be determined, or the license plate confidence based on the trajectory semantics in the vehicle passing record may be determined first, and then the license plate confidence based on the license plate semantics in the vehicle passing record may be determined.
When the confidence coefficient of the license plate based on the semantic meaning of the license plate is calculated, the confidence coefficient can be considered from two aspects, firstly, whether the license plate to be detected accords with the motor vehicle license plate rule or not is judged, and if the license plate to be detected accords with the motor vehicle license plate rule, the probability that the license plate to be detected accords with the motor vehicle license plate rule is determined to be 1; if the license plate does not accord with the motor vehicle license plate rule, the probability that the license plate to be detected accords with the motor vehicle license plate rule is determined to be 0, and the license plate to be detected is definitely an abnormal license plate. It can be seen that the probability value of the license plate to be detected according with the motor vehicle license plate rule is 1 or 0. Secondly, judging whether the license plate to be detected has registration information, and if so, determining that the probability that the license plate to be detected has the registration information is 1; and if the license plate to be detected does not have the registration information, determining that the probability that the license plate to be detected has the registration information is 0. Similarly, the probability value of the registration information of the license plate to be detected is 1 or 0. If partial registration information (for example, of a certain province) is stored locally at present, whether a license plate marked as the province is in a library or not can be determined through the registration information in the province, and whether the license plate marked as the province is in the library or not can not be determined through the license plate marked as the extraprovince. After the probability that the license plate to be detected accords with the motor vehicle license plate rule and the probability with the registration information are respectively determined, a third product of the probability that the license plate accords with the motor vehicle license plate coding rule and the probability with the registration information can be calculated, and the third product is determined as the license plate confidence coefficient based on the license plate semantics.
The license plate confidence based on the license plate semantics can be calculated through the description, and then the license plate confidence based on the track semantics can be calculated. When calculating the confidence coefficient of the license plate with the track semantics, the method can also be considered from two aspects, namely firstly, judging whether the license plate to be detected in each vehicle passing record is an even license plate error or not, and obtaining a first probability that the license plate to be detected in each vehicle passing record is the even license plate error; secondly, judging whether the license plate to be detected in each process record is a frequent license plate error or not, and obtaining a second probability that the license plate to be detected in each vehicle passing record is a frequent license plate error; and then determining the confidence level of the license plate based on the track semantics according to the first probability and the second probability. It should be noted that, when calculating the confidence level of the license plate based on the track semantics according to the first probability that the license plate to be detected in each passing record is an even current license plate error and the second probability that the license plate to be detected in each passing record is a frequent license plate error, the first probability that the license plate to be detected in each passing record is an even current license plate error may be determined, the second probability that the license plate to be detected in each passing record is a frequent license plate error may be determined, the first probability that the license plate to be detected in each passing record is an even current license plate error may be determined, and the second probability that the license plate to be detected in each passing record is a frequent license plate error may be determined as an example, but do not represent a limitation of the embodiments of the present application.
For example, when the confidence level of the license plate based on the track semantics is calculated according to the first probability that the license plate to be detected in each vehicle passing record is an even current license plate error and the second probability that the license plate to be detected in each vehicle passing record is a frequent current license plate error, the first weight corresponding to the even current license plate error and the second weight corresponding to the current license plate error can be respectively determined; calculating the product of the first weight and the first probability, and recording the product as a first product; calculating the product of the second weight and the second probability, and recording the product as a second product; and after the first product and the second product are obtained through calculation respectively, calculating the sum of the first product and the second product, wherein the sum is the confidence coefficient of the license plate based on the track semantics.
After the license plate confidence based on the license plate semantics and the license plate confidence based on the trajectory semantics are respectively obtained through calculation, a fourth product of the license plate confidence based on the license plate semantics (a third product of the probability according with the motor vehicle license plate coding rule and the probability with the registration information) and the license plate confidence based on the trajectory semantics (a sum of the first product and the second product) can be calculated, and the fourth product is the confidence of the license plate in each passing record. For example, in the embodiment of the present application, the formula can be used
Figure BDA0002335310720000101
And calculating the confidence coefficient of the license plate in each vehicle passing record. Wherein, CConfidence of license plate in vehicle passing recordThe confidence coefficient of the license plate in each passing record is represented, n represents the number of accumulation factors, m represents the number of accumulation factors, and for example, in the embodiment of the application, n is 2, and m is 2, two accumulation factors are respectively whether the license plate error is an even license plate error and whether the license plate error is a frequent license plate error, two accumulation factors are respectively whether the license plate error accords with the motor vehicle number plate coding rule and whether the license plate error has registration information, w isiWeight representing the ith accumulation factor, CiRepresenting the probability of the i-th accumulation factor, CjRepresenting the probability of the jth multiplicative factor.
It should be noted that, in the embodiment of the present application, when performing the illegal detection of the license plate involvement, it is considered that in the prior art, the accuracy of the detection result is low due to the fact that the license plate identification error in the vehicle passing record is incorrect, therefore, by calculating the confidence coefficient of the license plate in each vehicle passing record in the vehicle passing records in which the license plate to be detected passes through the M checkpoints, the confidence coefficient of the license plate in each vehicle passing record can be understood as the accuracy of the license plate identification in the vehicle passing record, so that whether the license plate involvement violation exists in the subsequent detection of a certain license plate to be detected or not is fully considered, the confidence coefficient of the license plate in each vehicle passing record is fully considered, and thus, the accuracy of the detection result is low due to the license plate identification error in the vehicle passing record can be avoided, and the accuracy of the detection result is effectively improved.
It can be understood that in the embodiment of the application, the confidence of each gate can be calculated first, and then the confidence of the license plate in each vehicle passing record can be calculated; or the confidence coefficient of the license plate in each passing record can be calculated firstly, and then the checkpoint confidence coefficient of each checkpoint is calculated; here, the embodiment of the present application is only described as an example in which the confidence level of each gate is calculated first, and then the confidence level of the license plate in each passing record is calculated, but the embodiment of the present application is not limited thereto.
After the confidence coefficient of each gate and the confidence coefficient of the license plate in each vehicle passing record are respectively calculated, the number of vehicle passing records meeting preset conditions in the vehicle passing records with the time-space contradiction can be determined according to the confidence coefficient of each gate and the confidence coefficient of the license plate in each vehicle passing record, namely the following S402 is executed:
s402, determining the number of the vehicle passing records meeting the preset conditions in the vehicle passing records with the time-space contradiction according to the entrance confidence coefficient of each entrance and the confidence coefficient of the license plate in each vehicle passing record.
The preset condition is used for indicating that the reason of the space-time contradiction is not a bayonet position marking error and is not a license plate recognition error. In other words, the determined reason that the space-time contradiction exists in the passing record meeting the preset condition is not the space-time contradiction existing due to the fact that the license plate identification error in the passing record exists and the bayonet coordinate calibration error in the passing record exists, so that the space-time contradiction existing in the license plate to be detected due to the bayonet position marking error and the license plate identification error can be eliminated, and the accuracy of the probability that the license plate to be detected has the license plate violation is effectively improved.
It can be understood that, before determining the number of the passing records meeting the preset conditions in the passing records with the time-space contradiction according to the confidence coefficient of each gate and the confidence coefficient of the license plate in each passing record, the number of the passing records meeting the preset conditions in the passing records with the time-space contradiction needs to be determined in the passing records with the time-space contradiction, and then the number of the passing records meeting the preset conditions in the passing records with the time-space contradiction can be determined according to the confidence coefficient of each gate and the confidence coefficient of the license plate in each passing record. For example, when the vehicle passing records with the time-space contradiction are determined to exist in the vehicle passing records of the M checkpoints, the vehicle passing records with the time-space contradiction can be determined to exist in the vehicle passing records of the M checkpoints according to the checkpoint identifier and the time of passing the checkpoint of each vehicle passing record and the position information of the checkpoint corresponding to the checkpoint identifier. In general, when determining that there is a vehicle-passing record with a time-space conflict, it is necessary to determine that there is a time-space conflict through two checkpoints and two vehicle-passing records, and the determination can be made according to the following steps:
f(Cbayonet 1，CBayonet 2，CVehicle passing record 1，CVehicle passing record 2) And determining the passing record with the time-space contradiction. Wherein, CBayonet 1And CBayonet 2Showing two bayonets, CVehicle passing record 1And CVehicle passing record 2Two vehicle passing records are shown.
Thus, after the number of the vehicle-passing records meeting the preset condition is determined in the vehicle-passing records with the time-space contradiction according to the confidence level of each gate and the confidence level of the license plate in each vehicle-passing record, the probability of the license plate to be detected that the license plate has the card-related violation can be determined according to the number of the vehicle-passing records with the time-space contradiction and the number of the vehicle-passing records meeting the preset condition, namely, the following S403 is executed:
s403, determining the probability of the license plate to be detected that the license plate is illegal by means of involving the license plate according to the number of the vehicle passing records with the space-time contradiction and the number of the vehicle passing records meeting the preset conditions.
Exemplary, may be according to CLaw of relation to cardsF (p, q) determining the probability of the license plate to be detected that the license plate is illegal; wherein, CLaw of relation to cardsF (p, q) represents the probability of detecting the existence of the license plate illegal by referring to the license plate, p represents the number of vehicle passing records with space-time contradiction, and q represents the number of vehicle passing records meeting the preset condition.
Therefore, in the embodiment of the application, when the probability of the existence of the card-related violation of the license plate to be detected is calculated, the confidence coefficient of each card port in the M card ports through which the license plate to be detected passes and the confidence coefficient of the license plate in each vehicle-passing record in the vehicle-passing records of the M card ports through which the license plate to be detected passes can be firstly obtained, the number of the vehicle-passing records with the time-space conflict and the number of the vehicle-passing records with the non-card port position marking error and the non-license plate recognition error in the vehicle-passing records are determined according to the confidence coefficient of each card port and the confidence coefficient of the license plate in each vehicle-passing record, and the number of the vehicle-passing records with the time-space conflict and the number of the vehicle-passing records with the non-license plate recognition error in the vehicle-passing records with the time-space conflict, the probability of the existence of the card-related violation of the license plate to be detected due to the card port position marking error and the license plate recognition error can be eliminated The accuracy of the probability of the illegal license plate involvement to be detected is effectively improved.
Based on the embodiment shown in fig. 4, in S401, before calculating the confidence level of the license plate based on the trajectory semantics according to the first probability that the license plate to be detected in each passing record is the even-current license plate error and the second probability that the license plate to be detected in each passing record is the frequent-current license plate error, it is necessary to determine the first probability that the license plate to be detected in each passing record is the even-current license plate error and the second probability that the license plate to be detected in each passing record is the frequent-current license plate error, so that the confidence level of the license plate based on the trajectory semantics can be calculated according to the first probability that the license plate to be detected in each passing record is the even-current license plate error and the second probability that the license plate to be detected in each passing record is the frequent-current license plate error. In the following, how to determine the first probability that the license plate to be detected in each vehicle-passing record is an even-current license plate error and the second probability that the license plate to be detected in each vehicle-passing record is a frequent license plate error in the embodiment of the present application will be described in detail. It should be noted that, in the embodiment of the present application, a first probability that the license plate to be detected in each vehicle passing record is an even license plate error may be determined, and then a second probability that the license plate to be detected in each vehicle passing record is a frequent license plate error may be determined; the second probability that the license plate to be detected in each passing record is the frequent license plate error may be determined first, and then the first probability that the license plate to be detected in each passing record is the even license plate error is determined, but the embodiment of the present application is only described by taking as an example the first probability that the license plate to be detected in each passing record is the even license plate error and then the second probability that the license plate to be detected in each passing record is the frequent license plate error, and the embodiment of the present application is not limited thereto.
For example, when a first probability that a license plate to be detected in each vehicle passing record is an accidental vehicle plate error is determined, a first vehicle passing record that the license plate to be detected passes through a third gate and a second vehicle passing record that the license plate to be detected passes through a fourth gate in space-time neighbors of the third gate in the M gates can be obtained; the third bayonet is any one of the M bayonets, and the fourth bayonet is different from the third bayonet; if the license plate identifier in the first passing record is the license plate to be detected, the license plate identifier in the second passing record is different from the license plate to be detected and is a similar license plate of the license plate to be detected, determining that the first probability is greater than 0; and if the license plate identifier in the first vehicle passing record is the license plate to be detected, and the license plate identifier in the second vehicle passing record is the same as the license plate to be detected, determining that the first probability is 0.
The space-time neighbors of the license plate to be detected passing through the third checkpoint comprise neighbors in two dimensions of time dimension and space dimension, and the time neighbors of the license plate to be detected passing through the third checkpoint in the time dimension can be understood as follows: and/or within a preset time period before the license plate to be detected passes through the third gate and/or within a preset time period after the license plate to be detected passes through the third gate. In general, the preset time period may be any value within 10 minutes to 30 minutes, and of course, may also be any value within 31 minutes to 40 minutes, and may be specifically set according to actual needs. For example, when the time that the license plate to be detected passes through the third checkpoint is 10:10, and the preset time period is 20 minutes, the time that the license plate to be detected passes through the third checkpoint is within a time period of 9:50-10:10, and/or within a time period of 10:10-10: 30. It can be understood that when the time when the license plate to be detected passes through the third gate is within the time period of 9:50-10:30, the time when the license plate to be detected passes through the third gate is within the time period of 9:50-10:10, or the time when the license plate to be detected passes through the third gate is within the time period of 10:10-10:30, under the normal condition, in the time period of 9:50-10:10, the collected passing records of the license plate to be detected passing through the fourth gate are more, and therefore, the accuracy of the first probability that the license plate to be detected is an even-present license plate error, which is calculated based on the collected passing records of the license plate to be detected passing through the third gate and the passing records of the license plate to be detected passing through the fourth gate, is higher. The spatial neighbors of the license plate to be detected on the time dimension through the third checkpoint can be understood as follows: the space adjacent region of the license plate to be detected passing through the third bayonet can be regarded as a circular region with the position of the third bayonet as the center and the preset space section as the radius. For example, the preset space segment may be any value within 1 km to 5 km, and of course, may also be any value within 5 km to 6 km, and may be specifically set according to actual needs. It can be understood that the larger the value of the preset space section is, the lower the calculation efficiency is due to the fact that the collected vehicle passing records of the license plate to be detected passing through the fourth bayonet are more, but the accuracy of the first probability that the license plate to be detected is an even vehicle license plate error, which is obtained through calculation, is higher based on the collected vehicle passing records of the license plate to be detected passing through the third bayonet and the vehicle passing records of the license plate to be detected passing through the fourth bayonet.
When the first probability that the license plate to be detected in each vehicle passing record is an accidental license plate error is determined, if the license plate to be detected passes through the third gate, but the license plate to be detected does not pass through the fourth card slot in the space-time neighbor passing through the third card slot (which can be determined according to the first vehicle passing record that the license plate to be detected passes through the third card slot and the second vehicle passing record that the license plate to be detected passes through the fourth card slot in the space-time neighbor passing through the third card slot), but the license plate similar to the license plate to be detected passes through the fourth card slot, that is, the license plate identification recorded by the second passing of the fourth checkpoint is different from the license plate to be detected and is a similar license plate of the license plate to be detected, the license plate to be detected is identified to be wrong to a great extent, which may be an occasional license plate error, under the condition, the first probability that the license plate to be detected is an occasional license plate error is more than 0; if the license plate to be detected passes through the third bayonet and the license plate to be detected passes through the fourth bayonet in the space-time neighbor passing through the third bayonet (the license plate to be detected can be determined according to the first vehicle passing record of the license plate to be detected passing through the third bayonet and the second vehicle passing record of the license plate to be detected passing through the fourth bayonet in the space-time neighbor passing through the third bayonet), the fact that the license plate to be detected has no recognition error is proved, the license plate to be detected cannot be an even current license plate error, and under the condition, the first probability that the license plate to be detected is the even current license plate error can be determined to be 0. For example, please refer to fig. 6, where fig. 6 is a schematic diagram of an even-appearing license plate error provided in the embodiment of the present application, in general, the number of tracks of the even-appearing license plate error is 2, and the setting may be specifically performed according to actual needs, and here, the embodiment of the present application only takes the number of tracks of the even-appearing license plate error as an example, but the embodiment of the present application is not limited thereto. For example, when determining whether two license plates are similar license plates, the error-prone character pair list can be searched and determined, the error-prone character pair list can be obtained through a license plate error correction model, and the error-prone character pair list can be obtained by counting the occurrence frequency of error characters in the error license plate.
Through the description, a first probability that the license plate to be detected in each vehicle passing record is an occasional license plate error can be calculated, and then a second probability that the license plate to be detected in each vehicle passing record is a frequent license plate error can be calculated. For example, when determining that the license plate to be detected in each vehicle passing record is the second probability of a frequent vehicle license plate error, clustering the vehicle passing records of the M gates according to the gate identifier and the time of passing through the gate of each vehicle passing record and the position information of the gate corresponding to the gate identifier in the M gates; and determining a second probability according to the number of the cluster sets obtained by clustering, so as to obtain a second probability that the license plate to be detected in each vehicle passing record is a frequent license plate error. The number of cluster sets obtained according to the clustering processing can be used for indicating the number of tracks of the license plate.
It can be understood that, when determining the second probability according to the number of cluster sets obtained by the clustering process, since frequent license plate errors generally occur in many places, the number of tracks is much greater than 2, and therefore, if the number of cluster sets is much greater than 2, it is indicated that frequent license plate errors occur to a large extent. For example, please refer to fig. 7, fig. 7 is a schematic diagram of a frequent license plate error provided in the embodiment of the present application, and it can be seen that the number of frequent license plate errors is greater than 2.
For example, when the vehicle passing records of the M slots are clustered according to the slot identifier and the time of passing through the slot of each vehicle passing record and the position information of the slot corresponding to the slot identifier, the vehicle passing records of the M slots may be sorted according to the time sequence, and the first vehicle passing record is used as the first element in the first clustering set; calculating N first moving speeds of the license plate to be detected according to the time of passing through the bayonet in the ith vehicle passing record, the position information of the bayonet corresponding to the bayonet identification in the ith vehicle passing record, the time of passing through the bayonet in the last vehicle passing record in each of the N existing cluster sets and the position information of the bayonet corresponding to the bayonet identification in the last vehicle passing record; wherein N is the number of the cluster sets in which all the vehicle passing records before the ith vehicle passing record are located, and i is greater than or equal to 2; comparing the N first moving speeds with a preset threshold value respectively, and if a second moving speed in the N first moving speeds is smaller than or equal to the preset threshold value, taking the ith vehicle passing record as the last element in a clustering set corresponding to the second moving speed; and if the N first moving speeds are all larger than a preset threshold value, taking the ith vehicle passing record as a first element in the new cluster set.
Taking the example that the vehicle passing records of the M gates include 5 vehicle passing records, each vehicle passing record includes the time when the license plate to be detected passes through the gate, so that the 5 vehicle passing records can be sorted according to the sequence of the time in the 5 vehicle passing records, and the first vehicle passing record is taken as the first element of the first clustering set; for a second vehicle passing record, determining the distance between two bayonets according to the position information of the bayonets corresponding to the bayonet identifiers in the second vehicle passing record and the position information of the bayonets corresponding to the bayonet identifiers in the element first vehicle passing record in the first cluster set, determining a time interval according to the time of passing through the bayonets in the second vehicle passing record and the time of passing through the bayonets in the element first vehicle passing record in the first cluster set, calculating a first moving speed of the license plate according to the distance between the two bayonets and the time interval, and dividing the second vehicle passing record into the first cluster set as the last element of the first cluster set if the first moving speed is smaller than or equal to a preset threshold value; for a third vehicle passing record, determining the distance between two bayonets according to the position information of the bayonets corresponding to the bayonet identifications in the third vehicle passing record and the position information of the bayonets corresponding to the bayonet identifications in the second vehicle passing record of the last element in the first cluster set, determining a time interval according to the time of passing through the bayonets in the third vehicle passing record and the time of passing through the bayonets in the second vehicle passing record of the last element in the first cluster set, calculating a first moving speed of the license plate according to the distance and the time interval between the two bayonets, and increasing a new second cluster set and taking the third vehicle passing record as the first element of the second cluster set if the first moving speed is greater than a preset threshold value; for the fourth vehicle passing record, because two cluster sets exist, the distance between the two bayonets needs to be determined according to the position information of the bayonets corresponding to the bayonet identifications in the fourth vehicle passing record and the position information of the bayonets corresponding to the bayonet identifications in the last element second vehicle passing record in the first cluster set, a time interval is determined according to the time of passing through the bayonets in the fourth vehicle passing record and the time of passing through the bayonets in the last element second vehicle passing record in the first cluster set, and then the first moving speed of the license plate is calculated according to the distance between the two bayonets and the time interval; determining the distance between the two bayonets according to the position information of the bayonets corresponding to the bayonet identifications in the fourth passing record and the position information of the bayonets corresponding to the bayonet identifications in the last element third passing record in the second clustering set, determining a time interval according to the time of passing through the bayonets in the fourth passing record and the time of passing through the bayonets in the last element third passing record in the second clustering set, and calculating the first moving speed of the license plate according to the distance between the two bayonets and the time interval; assuming that one of the two calculated first moving speeds is smaller than a preset threshold value, and the first moving speed is calculated by the distance and time interval between the bayonet in the fourth passing record and the bayonet in the second passing record, dividing the fourth passing record into a first clustering set as the last element in the first clustering set; for the fifth vehicle passing record, because there are already two cluster sets, it is necessary to determine the distance between the two bayonets according to the position information of the bayonet corresponding to the bayonet identifier in the fifth vehicle passing record and the position information of the bayonet corresponding to the bayonet identifier in the last element fourth vehicle passing record in the first cluster set, determine the time interval according to the time of passing through the bayonet in the fifth vehicle passing record and the time of passing through the bayonet in the last element fourth vehicle passing record in the first cluster set, and calculate the first moving speed of the license plate according to the distance between the two bayonets and the time interval; determining the distance between the two bayonets according to the position information of the bayonets corresponding to the bayonet identifications in the fifth passing record and the position information of the bayonets corresponding to the bayonet identifications in the last element third passing record in the second clustering set, determining a time interval according to the time of passing through the bayonets in the fifth passing record and the time of passing through the bayonets in the last element third passing record in the second clustering set, and calculating the first moving speed of the license plate according to the distance between the two bayonets and the time interval; and if the two calculated first moving speeds are both greater than a preset threshold value and the first moving speed is calculated by the distance and time interval between the bayonet in the fifth passing record and the bayonet in the fourth passing record, adding a new third clustering set and taking the fifth passing record as a first element of the third clustering set. Of course, in the practical application process, the number of the vehicle-passing records is much larger than 5, and the embodiment of the present application only takes the clustering process on the 5 vehicle-passing records as an example, but does not represent that the embodiment of the present application is limited thereto.
After the first probability that the license plate to be detected in each vehicle passing record is an even current license plate error and the second probability that the license plate to be detected in each vehicle passing record is a frequent current license plate error are respectively calculated, the confidence coefficient of the license plate based on the track semantics can be calculated according to the first probability that the license plate to be detected in each vehicle passing record is an even current license plate error and the second probability that the license plate to be detected in each vehicle passing record is a frequent current license plate error, and the confidence coefficient of the license plate in each vehicle passing record can be further obtained; and then determining the number of passing records with non-bayonet position labeling errors and non-license plate recognition errors caused by the time-space contradictions in the passing records with the time-space contradictions according to the confidence coefficient of each bayonet and the confidence coefficient of the license plate in each passing record, and then determining the probability of existence of the license plate violation of the license plate to be detected according to the number of the passing records with the time-space contradictions and the number of the passing records with the non-license plate recognition errors, so that the probability of existence of the time-space contradictions caused by the license plate violation of the license plate to be detected due to the bayonet position labeling errors and the license plate recognition errors can be eliminated, and the accuracy of the probability of existence of the license plate violation of the license plate to be detected can be effectively improved.
In order to facilitate understanding of the method for detecting a card-related violation provided in the embodiment of the present application, the method for detecting a card-related violation provided in the embodiment of the present application will be described below by way of example. Taking license plates to be detected as a Shaan AXXXXX and a Shaan CYYYYY as examples, wherein 2 bayonets through which the license plate AXXXXXXX to be detected passes are respectively a first bayonet and a second bayonet, the Shaan CYYYY passes through a third bayonet, the bayonet identification of the first bayonet is 32013414, the bayonet identification of the second bayonet is 32013415 and the bayonet identification of the third bayonet is 32013416, the time that the license plate AXXXXXXXXXXX to be detected passes through the first bayonet is 20190912041342, the time that the license plate AXXXXXXX to be detected passes through the second bayonet is 20190912044212, and the time that the license plate CYYYYYY to be detected passes through the third bayonet is 20190912041342, then the license plate AXXXXXXXXX to be detected passes through the 2 bayonets and correspondingly corresponds to 2 vehicle passing records, the first bayonet record comprises the license plate identification AXXXXXXX to be detected, the first bayonet identification and the time that the second bayonet passes through the second bayonet; the third passing record of the license plate passing through the third gate of the to-be-detected license plate label shan CYYYYYY comprises the license plate detecting identifier shan CYYYYYY, the third gate identifier and the time of passing through the third gate, and can be shown in the following table 1. In addition, the position of the gate may also be obtained, and the position may be represented by longitude and latitude, where the longitude of the first gate is 120.1513, the latitude is 32.12345, the longitude of the second gate is 120.1613, the latitude is 32.11345, the longitude of the third gate is 120.1713, and the latitude is 32.14345, as shown in table 2 below.
TABLE 1
License plate to be detected Bayonet sign Time of passing through bayonet
Shan AXXXXXXX 32013414 20190912041342
Shan AXXXXXXX 32013415 20190912044212
Shaan CYYYYY 32013416 20190912041342
TABLE 2
Bayonet sign Longitude (G) Latitude
32013414 120.1513 32.12345
32013415 120.1613 32.11345
32013416 120.1713 32.14345
By combining the table 1 and the expression 2, after the passing records of the passes of the to-be-detected license plates AXXXXX and the pass of the shan cyyyyyy are obtained, the pass confidence of each of the first pass, the second pass and the third pass can be respectively calculated according to the passing records of the passes of the to-be-detected license plates AXXXXX and the pass of the shan CYYYYY, and the confidence of the license plate in each of the pass records can be shown in the following table 3 and table 4, wherein the table 3 is the pass confidence of each of the first pass, the second pass and the third pass, and fig. 4 is the confidence of the license plate in the first pass record, the confidence of the license plate in the second pass record and the confidence of the third pass record.
TABLE 3
Bayonet sign Bayonet confidence
32013414 100％
32013415 77％
32013416 13％
TABLE 4
Vehicle passing record Confidence of license plate
First record of vehicle passing 90％
Second record of passing 75％
Third record of passing 95％
In obtaining the pass records of each of the first, second and third checkpoints and the pass records of the checkpoints through which the license plates to be detected pass through, the number of pass records meeting the preset conditions in the pass records with the time-space contradiction can be determined according to the checkpoint confidence of each checkpoint and the confidence of the license plate in each pass record after obtaining the confidence of the license plate in each pass record, as shown in the following table 5:
TABLE 5
Figure BDA0002335310720000161
As shown in table 5, comparing the data in the second column and the data in the third column in table 5 shows that, for the license plate to be detected, AXXXXX, 79 pass records with spatio-temporal contradictions due to data quality (for example, position calibration errors of a checkpoint and/or recognition errors of a license plate) can be filtered out by calculating the confidence of the checkpoint and the confidence of the license plate in the pass record. Similarly, for the license plate CYYYYY to be detected, 23 vehicle passing records with time-space contradiction caused by data quality (such as position calibration error of a checkpoint and/or vehicle plate recognition error) can be filtered out by calculating the confidence coefficient of the checkpoint and the confidence coefficient of the license plate in the vehicle passing record.
After the number of the vehicle passing records meeting the preset condition is determined, the probability that the license plate AXXXXX and the license plate CYYYYY of the license plate to be detected have card-related violation is determined according to the number of the vehicle passing records meeting the preset condition, which can be shown in the following table 6:
TABLE 6
Figure BDA0002335310720000162
When the probability of the license plate violation of the card involvement is calculated, the confidence coefficient of each card entrance in M card entrances where the license plate to be detected passes through and the confidence coefficient of the license plate in each vehicle passing record in the vehicle passing records where the license plate to be detected passes through the M card entrances can be firstly obtained, the number of the vehicle passing records with the wrong recognition of the license plate due to the space-time contradiction and the number of the vehicle passing records with the wrong recognition of the license plate are determined according to the confidence coefficient of each card entrance in each vehicle passing record and the confidence coefficient of the license plate in each vehicle passing record in the vehicle passing records with the space-time contradiction, the probability of the license plate violation of the card involvement in the license plate to be detected is determined according to the number of the vehicle passing records with the space-time contradiction and the number of the vehicle passing records with the wrong recognition of the space-time contradiction, so that the license plate to be detected has the space-time contradiction due to the position marking error of the license plate involvement in the card entrances and the number of the license plate violation of the license plate can be eliminated, the accuracy of the probability of the illegal license plate involvement to be detected is effectively improved.
Fig. 8 is a schematic structural diagram of a device 80 for detecting illegal card-related law provided in an embodiment of the present application, for example, please refer to fig. 8, where the device 80 for detecting illegal card-related law may include:
the processing unit 801 is configured to acquire a gate confidence of each gate in M gates through which a license plate to be detected passes, and a license plate confidence of each pass record in the pass records of the license plate to be detected passing through the M gates; wherein M is an integer greater than or equal to 2.
The determining unit 802 is configured to determine a detection result of the license plate to be detected according to the checkpoint confidence of each checkpoint and the confidence of the license plate in each vehicle passing record.
Optionally, the determining unit 802 is specifically configured to determine, according to the confidence level of each gate and the confidence level of the license plate in each vehicle passing record, the number of vehicle passing records meeting the preset condition in the vehicle passing records with the time-space contradiction; determining the probability of the license plate to be detected that the license plate is illegal by means of the license plate related law according to the number of the vehicle passing records with the time-space contradiction and the number of the vehicle passing records meeting the preset conditions; the preset condition is used for indicating that the reason of the space-time contradiction, namely the position marking error of the non-bayonet is caused and the recognition of the non-license plate is wrong; the vehicle passing records with the time-space contradiction are determined in the vehicle passing records for determining the M checkpoints according to the checkpoint marks in each vehicle passing record, the time for passing the checkpoints and the position information of the checkpoints corresponding to the checkpoint marks.
Optionally, the processing unit 801 is specifically configured to obtain, from among the M gates, a vehicle passing record that a license plate to be detected passes through the first gate, and a vehicle passing record that the license plate to be detected passes through the second gate within a time neighborhood of the license plate to be detected passing through the first gate; determining the bayonet confidence of the first bayonet according to the vehicle passing record passing through the first bayonet and the vehicle passing record passing through the second bayonet; wherein, first bayonet socket is any one bayonet socket in M bayonet sockets, and the second bayonet socket is different with first bayonet socket.
Optionally, the processing unit 801 is specifically configured to determine, according to the vehicle passing record passing through the first gate and the vehicle passing record passing through the second gate, the number of vehicle passing records in which the time-space conflict exists in the vehicle passing records passing through the first gate; determining the probability of the time-space contradiction of the vehicle passing records at the first gate according to the number of the vehicle passing records passing through the first gate and the number of the vehicle passing records with the time-space contradiction; and determining the checkpoint confidence of the first checkpoint according to the probability of the time-space contradiction of the vehicle-passing records at the first checkpoint and the number of the vehicle-passing records with the time-space contradiction.
Optionally, the processing unit 801 is specifically configured to determine the probability that the license plate to be detected meets the license plate rule of the motor vehicle and the probability that the license plate has registration information; determining a first probability that the license plate to be detected in each vehicle passing record is an even current license plate error and a second probability that the license plate to be detected in each vehicle passing record is a frequent current license plate error according to the vehicle passing records of the M checkpoints; and determining the confidence coefficient of the license plate in each passing record according to the probability that the license plate to be detected accords with the motor vehicle license plate coding rule, the probability with the registration information, the first probability and the second probability.
Optionally, the processing unit 801 is specifically configured to calculate a first product of a first weight corresponding to the even-present license plate error and a first probability, a second product of a second weight corresponding to the frequent license plate error and a second probability, and a third product of a probability according to the motor vehicle license plate encoding rule and a probability with registration information; and calculating a fourth product of the sum of the first product and the second product and the third product; and determining the confidence of the license plate in each vehicle passing record according to the fourth product.
Optionally, the processing unit 801 is specifically configured to obtain, from among the M gates, a first vehicle passing record that a license plate to be detected passes through a third gate, and a second vehicle passing record that the license plate to be detected passes through a fourth gate in space-time vicinity passing through the third gate; determining a first probability according to the license plate identification in the first vehicle passing record and the license plate identification in the second vehicle passing record; the third bayonet is any one of the M bayonets, and the fourth bayonet is different from the third bayonet.
Optionally, the processing unit 801 is specifically configured to determine that the first probability is greater than 0 if the license plate identifier in the first vehicle passing record is a license plate to be detected, and the license plate identifier in the second vehicle passing record is different from the license plate to be detected and is a similar license plate of the license plate to be detected; and if the license plate identifier in the first vehicle passing record is the license plate to be detected, and the license plate identifier in the second vehicle passing record is the same as the license plate to be detected, determining that the first probability is 0.
Optionally, the vehicle passing records include a bayonet identifier and time of passing through the bayonet, and the processing unit 801 is specifically configured to perform clustering processing on the vehicle passing records of the M bayonets according to the bayonet identifier and time of passing through the bayonet of each vehicle passing record in the M bayonets and position information of the bayonet corresponding to the bayonet identifier; and determining a second probability according to the number of the cluster sets obtained by the clustering processing.
Optionally, the processing unit 801 is specifically configured to sort the vehicle passing records of the M checkpoints according to a time sequence, and use the first vehicle passing record as a first element in the first cluster set; calculating N first moving speeds of the license plate to be detected according to the time of passing through the bayonet in the ith vehicle passing record, the position information of the bayonet corresponding to the bayonet identification in the ith vehicle passing record, the time of passing through the bayonet in the last vehicle passing record in each of the N existing cluster sets and the position information of the bayonet corresponding to the bayonet identification in the last vehicle passing record; comparing the N first moving speeds with a preset threshold value respectively, and if a second moving speed in the N first moving speeds is smaller than or equal to the preset threshold value, taking the ith vehicle passing record as the last element in a clustering set corresponding to the second moving speed; if the N first moving speeds are all larger than a preset threshold value, taking the ith vehicle passing record as a first element in a new cluster set; and N is the number of the cluster sets in which all the vehicle passing records before the ith vehicle passing record are located, and i is greater than or equal to 2.
The device 80 for detecting illegal involvement with cards shown in the embodiment of the present application can execute the method for detecting illegal involvement with cards in the embodiment shown in any of the above figures, and the implementation principle and the beneficial effects thereof are similar to those of the method for detecting illegal involvement with cards, and are not described herein again.
Fig. 9 is a schematic structural diagram of another device 90 for detecting illegal card involvement provided in the embodiment of the present application, for example, referring to fig. 9, the device 90 for detecting illegal card involvement may include at least one processor 901 and at least one memory 902, wherein,
the memory 902 is used for storing program instructions;
the processor 901 is configured to execute the program instructions in the memory 902, so that the device 90 for detecting illegal card-related law executes the method for detecting illegal card-related law in any one of the embodiments shown in the above drawings, and the implementation principle and the beneficial effect thereof are similar to those of the method for detecting illegal card-related law, and are not described herein again.
The embodiment of the present application further provides a chip, where a computer program is stored on the chip, and when the computer program is executed by a processor, the method for detecting a illegal card-related law in the embodiments shown in any of the above drawings is executed, and the implementation principle and the beneficial effect of the method are similar to those of the method for detecting a illegal card-related law, and are not described herein again.
The embodiment of the present application further provides a computer storage medium, which includes instructions, and when the instructions are executed by one or more processors, the apparatus for detecting a illegal card-related law executes the method for detecting a illegal card-related law in any of the embodiments shown in the above drawings, and the implementation principle and the beneficial effects thereof are similar to those of the method for detecting a illegal card-related law, and are not described herein again.
The processor in the above embodiments may be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, or discrete hardware components. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules may be located in a Random Access Memory (RAM), a flash memory, a read-only memory (ROM), a programmable ROM, an electrically erasable programmable memory, a register, or other storage media that are well known in the art. The storage medium is located in a memory, and a processor reads instructions in the memory and combines hardware thereof to complete the steps of the method.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.

Claims (22)

1. A method for detecting illegal card-related law is characterized by comprising the following steps:
acquiring the confidence coefficient of each checkpoint in M checkpoints through which a license plate to be detected passes, and the confidence coefficient of the license plate in each passing record in the passing records of the license plate to be detected passing through the M checkpoints; wherein M is an integer greater than or equal to 2; the confidence coefficient of the license plate in each vehicle passing record is used for representing the recognition accuracy of the license plate in the vehicle passing record; the confidence coefficient of the license plate in each vehicle passing record is determined based on the license plate semantic confidence coefficient and the track semantic confidence coefficient of the license plate; the bayonet confidence coefficient of each bayonet is used for representing the accuracy of the coordinates of the bayonet; the semantic confidence of the license plate is determined by the probability that the license plate meets the motor vehicle license plate number coding rule and the probability that the license plate has registration information; the semantic confidence of the track of the license plate is determined by a first probability that the license plate is an even license plate error and a second probability that the license plate is a frequent license plate error;
determining the detection result of the license plate to be detected according to the checkpoint confidence coefficient of each checkpoint and the confidence coefficient of the license plate in each vehicle passing record; the detection result of the license plate of the vehicle to be detected comprises the following steps: and the probability of illegal license plate involvement of the license plate to be detected is determined.
2. The method according to claim 1, wherein the determining the detection result of the license plate to be detected according to the confidence level of each gate and the confidence level of the license plate in each passing record comprises:
determining the number of vehicle passing records meeting preset conditions in the vehicle passing records with time-space contradictions according to the checkpoint confidence coefficient of each checkpoint and the confidence coefficient of the license plate in each vehicle passing record; the preset condition is used for indicating that the reason of the space-time contradiction, namely the position marking error of the non-bayonet is caused and the recognition of the non-license plate is wrong; the vehicle passing records with the time-space contradiction are determined in the vehicle passing records for determining the M checkpoints according to the checkpoint identifications and the checkpoint passing time in each vehicle passing record and the position information of the checkpoints corresponding to the checkpoint identifications;
and determining the probability of the license plate to be detected that the license plate is involved with the license plate violation according to the number of the vehicle passing records with the space-time contradiction and the number of the vehicle passing records meeting the preset condition.
3. The method according to claim 1 or 2, wherein the obtaining of the bayonet confidence of each of the M bayonets through which the license plate to be detected passes comprises:
obtaining the vehicle passing record of the license plate to be detected passing through a first gate and the vehicle passing record of the license plate to be detected passing through a second gate in the time neighborhood of the license plate to be detected passing through the first gate in the M gates; the first bayonet is any one of the M bayonets, and the second bayonet is different from the first bayonet;
and determining the confidence coefficient of the first gate according to the vehicle passing record of the first gate and the vehicle passing record of the second gate.
4. The method of claim 3, wherein determining a checkpoint confidence for the first checkpoint from the vehicle passing record for the first checkpoint and the vehicle passing record for the second checkpoint comprises:
determining the number of the vehicle passing records with time-space contradiction in the vehicle passing records passing through the first gate according to the vehicle passing records passing through the first gate and the vehicle passing records passing through the second gate;
determining the probability of the time-space contradiction of the vehicle passing records at the first gate according to the number of the vehicle passing records passing through the first gate and the number of the vehicle passing records with the time-space contradiction;
and determining the checkpoint confidence of the first checkpoint according to the probability of the time-space contradiction of the vehicle-passing records at the first checkpoint and the number of the vehicle-passing records with the time-space contradiction.
5. The method according to claim 1, wherein obtaining the confidence level of the license plate in each passing record in the passing records of the license plate to be detected passing through the M gates comprises:
determining the probability that the license plate to be detected accords with the motor vehicle license plate rule and the probability that the license plate to be detected has registration information;
determining a first probability that the license plate to be detected in each vehicle passing record is an even current license plate error and a second probability that the license plate to be detected in each vehicle passing record is a frequent current license plate error according to the vehicle passing records of the M checkpoints;
and determining the confidence coefficient of the license plate in each passing record according to the probability that the license plate to be detected accords with the motor vehicle license plate coding rule, the probability with the registration information, the first probability and the second probability.
6. The method of claim 5, wherein the determining the confidence level of the license plate in each passing record according to the probability that the license plate to be detected meets the motor vehicle license plate coding rule, the probability of having the registration information, the first probability, and the second probability comprises:
calculating a first product of a first weight corresponding to the even-present license plate error and the first probability, a second product of a second weight corresponding to the frequent license plate error and the second probability, and a third product of the probability according with the motor vehicle license plate coding rule and the probability with the registration information;
calculating a fourth product of the sum of the first product and the second product, and the third product;
and determining the confidence of the license plate in each vehicle passing record according to the fourth product.
7. The method according to claim 5 or 6, wherein the determining a first probability that the license plate to be detected in each pass record is an accidental license plate error according to the pass records of the M checkpoints comprises:
obtaining a first vehicle passing record of the license plate to be detected passing through a third checkpoint and a second vehicle passing record of the license plate to be detected passing through a fourth checkpoint in the space-time neighbors of the license plate to be detected passing through the third checkpoint in the M checkpoints; the third bayonet is any one of the M bayonets, and the fourth bayonet is different from the third bayonet;
and determining the first probability according to the license plate identification in the first vehicle passing record and the license plate identification in the second vehicle passing record.
8. The method of claim 7, wherein determining the first probability based on the license plate identifier in the first pass record and the license plate identifier in the second pass record comprises:
if the license plate identifier in the first passing record is the license plate to be detected, and the license plate identifier in the second passing record is different from the license plate to be detected and is a similar license plate of the license plate to be detected, determining that the first probability is greater than 0;
and if the license plate identifier in the first vehicle passing record is the license plate to be detected, and the license plate identifier in the second vehicle passing record is the same as the license plate to be detected, determining that the first probability is 0.
9. The method according to any one of claims 5, 6 and 8, wherein the passing records comprise a gate mark and time for passing a gate, and determining a second probability that the license plate to be detected in each passing record is a frequent license plate error according to the passing records of the M gates comprises:
according to the bayonet identification and the time of passing through the bayonet of each vehicle passing record in the M bayonets and the position information of the bayonet corresponding to the bayonet identification, clustering the vehicle passing records of the M bayonets;
and determining the second probability according to the number of the cluster sets obtained by clustering.
10. The method according to claim 9, wherein the clustering the vehicle passing records of the M checkpoints according to the checkpoint identifier and the checkpoint passing time of each vehicle passing record of the M checkpoints and the position information of the checkpoint corresponding to the checkpoint identifier comprises:
sequencing the vehicle passing records of the M checkpoints according to the time sequence, and taking the first vehicle passing record as a first element in a first cluster set;
calculating N first moving speeds of the license plate to be detected according to the time of passing through the bayonet in the ith vehicle passing record, the position information of the bayonet corresponding to the bayonet identification in the ith vehicle passing record, the time of passing through the bayonet in the last vehicle passing record in each of the N existing cluster sets and the position information of the bayonet corresponding to the bayonet identification in the last vehicle passing record; wherein N is the number of the cluster sets in which all the vehicle passing records before the ith vehicle passing record are located, and i is greater than or equal to 2;
comparing the N first moving speeds with a preset threshold respectively, and if a second moving speed in the N first moving speeds is smaller than or equal to the preset threshold, taking the ith vehicle passing record as the last element in a cluster set corresponding to the second moving speed; and if the N first moving speeds are all larger than a preset threshold value, taking the ith vehicle passing record as a first element in a new cluster set.
11. A detection device for illegal card-related law, comprising:
the processing unit is used for acquiring the confidence coefficient of each gate in M gates through which the license plate to be detected passes, and the confidence coefficient of the license plate in each passing record in the passing records of the license plate to be detected passing through the M gates; wherein M is an integer greater than or equal to 2; the confidence coefficient of the license plate in each vehicle passing record is used for representing the recognition accuracy of the license plate in the vehicle passing record; the confidence coefficient of the license plate in each vehicle passing record is determined based on the license plate semantic confidence coefficient and the track semantic confidence coefficient of the license plate; the bayonet confidence coefficient of each bayonet is used for representing the accuracy of the coordinates of the bayonet; the semantic confidence of the license plate is determined by the probability that the license plate meets the motor vehicle license plate number coding rule and the probability that the license plate has registration information; the semantic confidence of the track of the license plate is determined by a first probability that the license plate is an even license plate error and a second probability that the license plate is a frequent license plate error;
the determining unit is used for determining the detection result of the license plate to be detected according to the checkpoint confidence coefficient of each checkpoint and the confidence coefficient of the license plate in each passing record; the detection result of the license plate of the vehicle to be detected comprises the following steps: and the probability of illegal license plate involvement of the license plate to be detected is determined.
12. The apparatus of claim 11,
the determining unit is specifically configured to determine, according to the confidence level of each gate and the confidence level of the license plate in each passing record, the number of passing records meeting preset conditions in the passing records with time-space contradictions; determining the probability of the license plate to be detected of having the illegal card-related law according to the number of the vehicle-passing records with the time-space contradiction and the number of the vehicle-passing records meeting the preset condition; the preset condition is used for indicating that the reason of the space-time contradiction, namely the position marking error of the non-bayonet is caused and the recognition of the non-license plate is wrong; the vehicle passing records with the space-time contradiction are determined in the vehicle passing records for determining the M checkpoints according to the checkpoint identifications in each vehicle passing record, the time for passing the checkpoints and the position information of the checkpoints corresponding to the checkpoint identifications.
13. The apparatus of claim 11 or 12,
the processing unit is specifically configured to obtain, from the M gates, a vehicle passing record that the license plate to be detected passes through a first gate, and a vehicle passing record that the license plate to be detected passes through a second gate within a time neighborhood of the license plate to be detected passing through the first gate; determining the checkpoint confidence of the first checkpoint according to the vehicle passing record of the first checkpoint and the vehicle passing record of the second checkpoint; wherein, first bayonet socket is in any one bayonet socket in M bayonet sockets, the second bayonet socket with first bayonet socket is different.
14. The apparatus of claim 13,
the processing unit is specifically configured to determine, according to the vehicle passing record passing through the first gate and the vehicle passing record passing through the second gate, the number of vehicle passing records in which a temporal-spatial contradiction exists in the vehicle passing records passing through the first gate; determining the probability of the time-space contradiction of the vehicle passing records at the first gate according to the number of the vehicle passing records passing through the first gate and the number of the vehicle passing records with the time-space contradiction; and determining the checkpoint confidence of the first checkpoint according to the probability of the time-space contradiction of the vehicle-passing records at the first checkpoint and the number of the vehicle-passing records with the time-space contradiction.
15. The apparatus of claim 11,
the processing unit is specifically used for determining the probability that the license plate to be detected meets the motor vehicle license plate rule and the probability that the license plate to be detected has registration information; determining a first probability that the license plate to be detected in each vehicle passing record is an even current license plate error and a second probability that the license plate to be detected in each vehicle passing record is a frequent current license plate error according to the vehicle passing records of the M checkpoints; and determining the confidence of the license plate in each passing record according to the probability that the license plate to be detected accords with the motor vehicle license plate coding rule, the probability with the registration information, the first probability and the second probability.
16. The apparatus of claim 15,
the processing unit is specifically configured to calculate a first product of a first weight corresponding to the even-present license plate error and the first probability, a second product of a second weight corresponding to the frequent license plate error and the second probability, and a third product of the probability according with the motor vehicle license plate coding rule and the probability with the registration information; and calculating a fourth product of the sum of the first product and the second product, and the third product; and determining the confidence of the license plate in each vehicle passing record according to the fourth product.
17. The apparatus of claim 15 or 16,
the processing unit is specifically configured to obtain a first vehicle passing record that the license plate to be detected passes through a third gate and a second vehicle passing record that the license plate to be detected passes through a fourth gate in the space-time neighborhood of the license plate to be detected passing through the third gate from among the M gates; determining the first probability according to the license plate identification in the first vehicle passing record and the license plate identification in the second vehicle passing record; wherein, the third bayonet socket is in any one bayonet socket in M bayonet sockets, the fourth bayonet socket with the third bayonet socket is different.
18. The apparatus of claim 17,
the processing unit is specifically configured to determine that the first probability is greater than 0 if the license plate identifier in the first vehicle passing record is the license plate to be detected, and the license plate identifier in the second vehicle passing record is different from the license plate to be detected and is a similar license plate of the license plate to be detected; and if the license plate identification in the first passing record is the license plate to be detected and the license plate identification in the second passing record is the same as the license plate to be detected, determining that the first probability is 0.
19. The apparatus according to any one of claims 15, 16, and 18, wherein the vehicle passing record includes a gate identifier and a time of passing through a gate, and the processing unit is specifically configured to perform clustering processing on the vehicle passing records of the M gates according to the gate identifier and the time of passing through a gate of each of the M gates, and position information of the gate corresponding to the gate identifier; and determining the second probability according to the number of cluster sets obtained by clustering.
20. The apparatus of claim 19,
the processing unit is specifically configured to sort the vehicle passing records of the M checkpoints according to a time sequence, and use a first vehicle passing record as a first element in a first cluster set; calculating N first moving speeds of the license plate to be detected according to the time of passing through the bayonet in the ith vehicle passing record, the position information of the bayonet corresponding to the bayonet identification in the ith vehicle passing record, the time of passing through the bayonet in the last vehicle passing record in each of the N existing cluster sets and the position information of the bayonet corresponding to the bayonet identification in the last vehicle passing record; comparing the N first moving speeds with a preset threshold respectively, and if a second moving speed in the N first moving speeds is smaller than or equal to the preset threshold, taking the ith vehicle passing record as the last element in a cluster set corresponding to the second moving speed; if the N first moving speeds are all larger than a preset threshold value, taking the ith vehicle passing record as a first element in a new cluster set; and N is the number of the cluster sets in which all the vehicle passing records before the ith vehicle passing record are located, and i is greater than or equal to 2.
21. A device for detecting a violation of a card, comprising at least one processor and at least one memory, wherein,
the memory is to store program instructions;
the processor is configured to execute the program instructions in the memory to enable the device for detecting the illegal card-related law to implement the method for detecting the illegal card-related law according to any one of the above claims 1 to 10.
22. A computer storage medium comprising instructions which, when executed by one or more processors, cause a device for detecting a violation of a card of claim 1-10 to perform the method of detecting a violation of a card of claim.
CN201911353595.8A 2019-12-25 2019-12-25 Method and device for detecting illegal card involvement Active CN111223305B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911353595.8A CN111223305B (en) 2019-12-25 2019-12-25 Method and device for detecting illegal card involvement

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911353595.8A CN111223305B (en) 2019-12-25 2019-12-25 Method and device for detecting illegal card involvement

Publications (2)

Publication Number Publication Date
CN111223305A CN111223305A (en) 2020-06-02
CN111223305B true CN111223305B (en) 2022-05-17

Family

ID=70832109

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911353595.8A Active CN111223305B (en) 2019-12-25 2019-12-25 Method and device for detecting illegal card involvement

Country Status (1)

Country Link
CN (1) CN111223305B (en)

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102789690A (en) * 2012-07-17 2012-11-21 公安部道路交通安全研究中心 Illegal vehicle identifying method and system
CN103246876A (en) * 2013-05-10 2013-08-14 苏州祥益网络科技有限公司 Image feature comparison based counterfeit vehicle registration plate identification method
CN105336164A (en) * 2015-10-27 2016-02-17 杭州电子科技大学 Error checkpoint positional information automatic identification method based on big data analysis
CN105513368A (en) * 2015-11-26 2016-04-20 银江股份有限公司 Uncertain information-based method for screening vehicles with false license plates
CN105702047A (en) * 2016-03-04 2016-06-22 浙江宇视科技有限公司 License plate identification error filtering method and apparatus in fake-license plate analysis
CN105719489A (en) * 2016-03-24 2016-06-29 银江股份有限公司 Fake-licensed vehicle detection method based on bayonet vehicle flow direction probability
CN105976610A (en) * 2016-03-21 2016-09-28 江苏智通交通科技有限公司 Bayonet passing vehicle detection reliability control method and system
CN106373406A (en) * 2016-12-02 2017-02-01 南京航空航天大学 Inspecting system against fake-licensed vehicles
CN107305736A (en) * 2016-04-22 2017-10-31 贵州众成信息产业有限公司 A kind of intelligent error-checking method of Car license recognition based on car plate rule and space-time accessibility and system
CN108242153A (en) * 2018-03-12 2018-07-03 小草数语（北京）科技有限公司 Abnormal bayonet recognition methods and device
CN108681717A (en) * 2018-05-18 2018-10-19 贵州云腾志远科技发展有限公司 City-level traffic video detection equipment quality detection method

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103914986B (en) * 2014-03-14 2016-07-13 浙江宇视科技有限公司 The method and apparatus that a kind of deck is analyzed
CN105654735A (en) * 2016-03-24 2016-06-08 安徽四创电子股份有限公司 Rapid fake-licensed car recognizing method based on outlier analysis algorithm
CN108073923B (en) * 2016-11-09 2021-09-17 杭州海康威视数字技术股份有限公司 License plate correction method and device

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102789690A (en) * 2012-07-17 2012-11-21 公安部道路交通安全研究中心 Illegal vehicle identifying method and system
CN103246876A (en) * 2013-05-10 2013-08-14 苏州祥益网络科技有限公司 Image feature comparison based counterfeit vehicle registration plate identification method
CN105336164A (en) * 2015-10-27 2016-02-17 杭州电子科技大学 Error checkpoint positional information automatic identification method based on big data analysis
CN105513368A (en) * 2015-11-26 2016-04-20 银江股份有限公司 Uncertain information-based method for screening vehicles with false license plates
CN105702047A (en) * 2016-03-04 2016-06-22 浙江宇视科技有限公司 License plate identification error filtering method and apparatus in fake-license plate analysis
CN105976610A (en) * 2016-03-21 2016-09-28 江苏智通交通科技有限公司 Bayonet passing vehicle detection reliability control method and system
CN105719489A (en) * 2016-03-24 2016-06-29 银江股份有限公司 Fake-licensed vehicle detection method based on bayonet vehicle flow direction probability
CN107305736A (en) * 2016-04-22 2017-10-31 贵州众成信息产业有限公司 A kind of intelligent error-checking method of Car license recognition based on car plate rule and space-time accessibility and system
CN106373406A (en) * 2016-12-02 2017-02-01 南京航空航天大学 Inspecting system against fake-licensed vehicles
CN108242153A (en) * 2018-03-12 2018-07-03 小草数语（北京）科技有限公司 Abnormal bayonet recognition methods and device
CN108681717A (en) * 2018-05-18 2018-10-19 贵州云腾志远科技发展有限公司 City-level traffic video detection equipment quality detection method

Also Published As

Publication number Publication date
CN111223305A (en) 2020-06-02

Similar Documents

Publication Publication Date Title
CN106991820B (en) Illegal vehicle processing method and device
CN107305736B (en) License plate recognition intelligent error checking method and system based on license plate rule and space-time reachability
CN112309126B (en) License plate detection method and device, electronic equipment and computer readable storage medium
CN111369801B (en) Vehicle identification method, device, equipment and storage medium
CN107329977B (en) A kind of false-trademark vehicle postsearch screening method based on probability distribution
CN112183367B (en) Vehicle data error detection method, device, server and storage medium
CN111369805B (en) Fake plate detection method and device, electronic equipment and computer readable storage medium
CN109344886B (en) Occlusion number plate distinguishing method based on convolutional neural network
CN101587643A (en) Identification method of fake-licensed cars
CN108364474B (en) Data fusion processing method and system based on image and RFID information
Beibut et al. Effective algorithms and methods for automatic number plate recognition
CN110164137B (en) Method, system and medium for identifying fake-licensed vehicle based on driving time of bayonet pair
CN112115946B (en) License plate fake-license plate identification method and device, storage medium and electronic equipment
CN111078973A (en) Fake-licensed vehicle identification method and equipment based on big data and storage medium
CN111369790B (en) Vehicle passing record correction method, device, equipment and storage medium
CN111223305B (en) Method and device for detecting illegal card involvement
CN108205670B (en) License plate recognition method and device
CN110674749B (en) Fake-licensed vehicle identification method and device and storage medium
CN114201530B (en) Early screening and preventive supervision method for suspected abnormal operation passenger car
CN110718070A (en) Accompanying vehicle identification method, device, equipment and storage medium
CN114023085B (en) Intersection signal timing parameter inference method based on bayonet detection data
CN111708769B (en) Method, system, computer storage medium and terminal for realizing information processing
CN110728276A (en) License plate recognition method and device
CN110880244A (en) Method and system for distinguishing whether taxi is cloned taxi
CN117349688B (en) Track clustering method, device, equipment and medium based on peak track

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20220908

Address after: No. 1899 Xiyuan Avenue, high tech Zone (West District), Chengdu, Sichuan 610041

Patentee after: Chengdu Huawei Technologies Co.,Ltd.

Address before: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen

Patentee before: HUAWEI TECHNOLOGIES Co.,Ltd.