Traffic violation detection method and system based on mimicry architecture

Abstract

The present invention provides a traffic violation detection method and system based on a mimicry architecture, the method comprising the following steps: step 1, obtaining target image information and identification information thereof; step 2, performing preliminary violation detection, and if a violation is detected, transmitting a violation identification code and the like to an input agent; if no violation is detected, transmitting a non-violation identification code to the input agent; step 3, the input agent determining whether the parsed identification code belongs to a violation detection result of the same group of target image information; step 4, judging the extracted identification code by an input arbiter, and generating a violation mark if the judgment is passed; the input agent encapsulating the violation mark and the like into a violation data packet I, and then copying and distributing the data to M business processing backends; step 5, the business processing backend encapsulating the vehicle owner information and the violation data packet I into a violation data packet II, and then transmitting the data packet II to an output arbiter; thereby performing mimicry processing on the traffic violation detection process.

Description

Technical Field

[0001] The present invention relates to the field of mimicry defense technology, and in particular to a traffic violation detection method and system based on a mimicry architecture. Background Art

[0002] "Electronic eyes," also known as "electronic police," are the common name for an intelligent traffic violation monitoring and management system. Combining vehicle detection, photoelectric imaging, automatic control, network communications, and computer technologies, these systems provide 24 / 7 surveillance of motor vehicle violations such as running red lights, driving against traffic, speeding, crossing lanes, and illegal parking, capturing images and text of these violations. Once the captured images are downloaded and transmitted to the command center, they are registered, numbered, and publicly announced before being transferred to the central computer database for access by various agencies.

[0003] Road violation detection technology is more difficult than other detection technologies because the objects of road violation identification are significantly more complex: 1) The background is more complex, and vehicles on the road move at high speeds; 2) Vehicle trajectories may change constantly, even appearing irregularly, which may lead to misjudgments due to different vehicle trajectories; 3) Vehicle shadows are difficult to identify on rainy and foggy days, and the human eye can easily misjudge them in such situations; 4) Insufficient light at night can also affect the identification results, causing misjudgments.

[0004] At present, electronic eyes generally use one-way equipment to take pictures. Although the probability of misjudgment of traffic violations is low, once the data processing results are misjudged, it will cause unnecessary trouble to the driver.

[0005] In order to solve the above problems, people have been seeking an ideal technical solution. Summary of the Invention

[0006] The purpose of the present invention is to address the deficiencies of the existing technology and thus provide a traffic violation detection method and system based on a mimicry architecture.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] A first aspect of the present invention provides a method for detecting traffic violations based on a mimic architecture, the method comprising the following steps:

[0009] Step 1: Each violation detection module obtains target image information and identification information of the target image information respectively;

[0010] The identification information includes license plate information, capture time information and electronic eye terminal information;

[0011] Step 2: Each violation detection module performs preliminary violation detection based on the corresponding target image information.

[0012] If a traffic violation is detected, the corresponding target image information is used as a traffic violation image, and the traffic violation image, the identification information of the traffic violation image and the pre-agreed traffic violation identification code are encapsulated into a preliminary detection result, which is then transmitted to the input agent;

[0013] If no violation is detected, the pre-agreed non-violation identification code and the identification information of the target image information are packaged into a preliminary detection result and transmitted to the input agent;

[0014] Step 3: The input agent parses the identification code and identification information from the received preliminary detection results, and determines whether the parsed identification code belongs to the violation detection results of the same set of target image information based on the parsed identification information; wherein the identification code includes a violation identification code and a non-violation identification code, and the same set of target image information refers to images collected by the same mimetic electronic eye within the same time period, and each mimetic electronic eye includes N information collection terminals, and each information collection terminal corresponds to one violation detection module;

[0015] Step 4, when it is determined that the parsed identification code belongs to the same set of violation detection results of the target image information, the extracted identification code is adjudicated by calling the input adjudicator, and if the adjudication is passed, a violation mark is generated; wherein the identification code includes a violation identification code and a non-violation identification code;

[0016] The input agent encapsulates the violation mark, violation picture and identification information of the violation picture into a violation data packet I, and copies and distributes it to M business processing backends;

[0017] Step 5: After receiving the traffic violation data packet I from the input agent, the business processing backend obtains the owner information of the violating vehicle based on the identification information of the violation image in the traffic violation data packet I, encapsulates the owner information and the traffic violation data packet I into a traffic violation data packet II, and transmits the data packet II to the output arbitrator of the output agent;

[0018] Step 6: the output arbitrator makes an adjudication on the received violation data packet II. If the adjudication is successful, the violation data packet II that complies with the adjudication strategy is transmitted to the violation database for storage.

[0019] A second aspect of the present invention provides a traffic violation detection system based on a mimetic architecture, the system comprising a mimetic electronic eye, a violation detector, an input agent, an output agent, a heterogeneous execution platform, and a violation database, the mimetic electronic eye comprising N information collection terminals, the violation detector comprising N violation detection modules, the heterogeneous execution platform comprising M business processing backends, the input agent being configured with an input arbitrator, and the output agent being configured with an output arbitrator;

[0020] N information collection terminals are respectively connected to the N violation detection modules in communication, the N violation detection modules are respectively connected to the input agent in communication, the input agent is respectively connected to the M business processing backends in communication, the M business processing backends are respectively connected to the output agent in communication, and the output agent is connected to the violation database in communication; wherein,

[0021] The mimic electronic eye is used to collect target image information in real time, generate identification information of the target image information, and transmit the target image information and the identification information of the target image information to the corresponding violation detection module; wherein the identification information includes license plate information, capture time information and electronic eye terminal information;

[0022] The violation detection module is configured to obtain corresponding target image information and identification information of the target image information, perform preliminary violation detection based on the corresponding target image information, and if a violation is detected, use the corresponding target image information as a violation image, encapsulate the violation image, the identification information of the violation image, and a pre-agreed violation identification code into a preliminary detection result, which is then transmitted to an input agent; if no violation is detected, encapsulate a pre-agreed non-violation identification code and the identification information of the target image information into a preliminary detection result, which is then transmitted to an input agent;

[0023] The input agent is configured to parse the identification code and identification information from the received preliminary detection result, and determine whether the parsed identification code belongs to the same set of violation detection results of the target image information based on the parsed identification information; further configured to, when it is determined that the parsed identification code belongs to the same set of violation detection results of the target image information, call the input adjudicator to adjudicate the extracted identification code, and generate a violation mark if the adjudication is passed; further configured to encapsulate the violation mark, the violation image, and the identification information of the violation image into a violation data packet I, and then copy and distribute the data packet to M business processing backends;

[0024] The identification code includes a violation identification code and a non-violation identification code. The same set of target image information refers to images collected by the same mimetic electronic eye in the same time period. Each mimetic electronic eye includes N information collection terminals, and each information collection terminal corresponds to one violation detection module.

[0025] The business processing backend is configured to, upon receiving a violation data packet I from an input agent, obtain the owner information of the violating vehicle based on the identification information of the violation image in the violation data packet I, encapsulate the owner information and the violation data packet I into a violation data packet II, and transmit the data packet II to an output arbitrator of the output agent;

[0026] The output arbiter is used to make an adjudication on the received traffic violation data packet II. If the adjudication is successful, the traffic violation data packet II that complies with the adjudication strategy is transmitted to the traffic violation database for storage.

[0027] The present invention has outstanding substantive features and significant progress compared to the prior art. Specifically:

[0028] 1) This invention proposes a traffic violation detection method and system based on a mimetic architecture. This system incorporates mimetic concepts to mimeticize the traffic violation detection process. Even if a violation detection result from one of the information collection terminals is misjudged, the invention can ensure the accuracy of the final detection result by inputting it into the arbiter for decision processing.

[0029] At the same time, the M business processing backends on the heterogeneous executive platform are used to perform mimicry processing on the process of calling and querying the owner information, and the output arbitrator is used to adjudicate the included violation data packets, thereby preventing the violation data packet II stored in the violation database from being maliciously tampered with;

[0030] 2) The present invention provides a violation identification code and a non-violation identification code to ensure data transmission efficiency between the violation detection module and the input agent; and the determination object of the input arbitrator in step 4 is the identification code, which can effectively improve the determination efficiency;

[0031] 3) The input arbiter of the present invention also marks and records abnormalities for electronic eye terminals that fail the identification code arbitration, and sends maintenance information to the maintenance management terminal when the number of abnormalities exceeds a threshold, reminding maintenance personnel to carry out maintenance in a timely manner;

[0032] 4) The business processing backend of the present invention can communicate and interconnect with the vehicle management database, so that after receiving the violation data packet I from the input agent, the business processing backend can query and obtain the owner information of the violating vehicle based on the identification information extracted from the violation image;

[0033] 5) The present invention can reduce the misjudgment rate of traffic violations, thereby improving government service efficiency and saving manpower costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of the traffic violation detection method based on mimicry architecture of the present invention;

[0035] Figure 2 is a flow chart of a method for identifying preliminary detection results of the same set of target image information of the present invention;

[0036] Figure 3 This is a schematic diagram of the structure of the traffic violation detection system based on the mimic architecture of the present invention. Figure 1 ;

[0037] Figure 4 This is a schematic diagram of the structure of the traffic violation detection system based on the mimic architecture of the present invention. Figure 2 . DETAILED DESCRIPTION

[0038] The technical solution of the present invention is further described in detail below through specific implementation methods.

[0039] Example 1

[0040] As attached Figure 1 and 3 As shown, a traffic violation detection method based on a mimic architecture includes the following steps:

[0041] Step 1: Each violation detection module obtains target image information and identification information of the target image information respectively;

[0042] The identification information includes license plate information, capture time information, and electronic eye terminal information, and the target image information is the image information captured by the information collection terminal;

[0043] Step 2: Each violation detection module performs preliminary violation detection based on the corresponding target image information.

[0044] If a traffic violation is detected, the corresponding target image information is used as a traffic violation image, and the traffic violation image, the identification information of the traffic violation image and the pre-agreed traffic violation identification code are encapsulated into a preliminary detection result, which is then transmitted to the input agent;

[0045] If no violation is detected, the pre-agreed non-violation identification code and the identification information of the target image information are packaged into a preliminary detection result and transmitted to the input agent;

[0046] Step 3: The input agent parses the identification code and identification information from the received preliminary detection result, and determines whether the parsed identification code belongs to the same set of violation detection results of the target image information based on the parsed identification information; wherein the identification code includes a violation identification code and a non-violation identification code;

[0047] Step 4: When it is determined that the parsed identification code belongs to the same set of violation detection results of the target image information, the extracted identification code is adjudicated by calling the input adjudicator of the input agent, and a violation mark is generated if the adjudication is passed; wherein the identification code includes a violation identification code and a non-violation identification code;

[0048] The input agent encapsulates the violation mark, violation picture and identification information of the violation picture into a violation data packet I, and copies and distributes it to M business processing backends;

[0049] Step 5: After receiving the traffic violation data packet I from the input agent, the business processing backend obtains the owner information of the violating vehicle based on the identification information of the violation image in the traffic violation data packet I, encapsulates the owner information and the traffic violation data packet I into a traffic violation data packet II, and transmits the data packet II to the output arbitrator of the output agent;

[0050] The business processing backend is connected to the vehicle management office database, and the owner information includes license plate number, owner's name, mobile phone number, ID number, home address, etc.

[0051] Step 6: the output arbitrator makes an adjudication on the received violation data packet II. If the adjudication is successful, the violation data packet II that complies with the adjudication strategy is transmitted to the violation database for storage.

[0052] It can be understood that in the process of traffic violation detection, there are two decisions, which adopt the majority decision strategy or the consistent decision strategy. The majority decision strategy or the consistent decision strategy will not be repeated here; the first time is that the input decision device makes a decision on the extracted identification code, and the decision object is mainly the identification code, which includes the violation identification code and the non-violation identification code; the second time is that the output decision device makes a decision on the received violation data packet II, and the decision object is mainly the vehicle owner information.

[0053] It should be noted that the input arbiter receives multiple preliminary detection results from the violation detection module within the same time period, and when the input arbiter makes an adjudication on the extracted identification code, it is based on the preliminary detection results of the same group of target image information; therefore, it is necessary to filter out preliminary detection results belonging to the same group of target image information from these preliminary detection results. The same group of target image information refers to images collected by the same mimetic electronic eye within the same time period. Each mimetic electronic eye includes N information collection terminals, and each information collection terminal corresponds to one violation detection module.

[0054] Attachment Figure 2A flow chart of a method for identifying preliminary detection results of the same set of target image information is shown. In step 3, when the input agent determines whether the parsed identification code belongs to the violation detection result of the same set of target image information based on the parsed identification information, the following is executed:

[0055] Extracting the license plate information from the identification information, determining whether the extracted license plate information is the same license plate information, and if so, extracting the electronic eye terminal information from the identification information; wherein the electronic eye terminal information includes terminal location information and terminal number;

[0056] Determine whether the extracted terminal numbers are continuous. If so, determine whether the terminal location information is in the same area. If so, determine whether the parsed identification codes belong to the same group of violation detection results of target image information.

[0057] It can be understood that when the equipment is installed and configured, the terminal numbers of the N information collection terminals in the same mimetic electronic eye are configured with the first few digits being the same and the last few digits being consecutive, so as to facilitate the grouping of preliminary detection results when making violation decisions.

[0058] In addition, the terminal location information of N information collection terminals in the same mimetic electronic eye is in the same area, and the terminal location information can be in the following format, such as the intersection of xx Road and xx Road, or xx meters east of xx Road.

[0059] It should be noted that in different application scenarios, the N information collection terminals (electronic eyes) of the same mimetic electronic eye are placed in different ways, that is, arranged in different ways; for example, application scenarios such as capturing red light running and seat belt fastening: these can be completed in one location by simply arranging N electronic eyes horizontally, and appropriately increasing the distance between each other to form a suitable angle to capture the lane; if it is an application scenario such as interval speed measurement, these N electronic eyes can be placed longitudinally along the road, and the distance between each other can be appropriately increased.

[0060] Furthermore, in step 2, when each violation detection module performs preliminary violation detection based on the corresponding target image information, it executes:

[0061] Extracting electronic eye terminal information from the identification information, wherein the electronic eye terminal information includes a terminal arrangement mode;

[0062] If the terminal arrangement in the identification information is horizontal arrangement, performing a first type of violation detection on the target image information;

[0063] If the terminal arrangement in the identification information is vertical arrangement, the second type of violation behavior detection is performed on the target image information.

[0064] Furthermore, the first type of violation detection is running a red light or a yellow light (if it is considered a violation or requires a warning to the driver), and the second type of violation detection is instantaneous speed measurement, interval speed measurement, solid line lane change, etc. The violation detection algorithm used is the existing technology known to those skilled in the art and will not be repeated here;

[0065] It should be noted that when N violation detection modules are detecting the same type of violation, they can use homogeneous algorithm modules (same algorithm) or heterogeneous algorithm modules (different algorithms). In order to further reduce the misjudgment rate, N violation detection modules use heterogeneous algorithm modules when detecting the same type of violation.

[0066] Furthermore, the input arbiter and the violation detection module pre-agreed that the violation identification code is the violation type identifier + 1, and the non-violation identification code is the violation type identifier + 0. The violation type identifier is used to distinguish specific violation behaviors, and 0 and 1 are used to make signal transmission between the input arbiter and the violation detection module more covert and more efficient.

[0067] Specifically, the violation type identifier may be composed of pre-agreed letters, symbols or data.

[0068] Specifically, the information collection terminal is an image collector, and the license plate information is extracted from the target image information by the image collector using existing image recognition technology;

[0069] N is a natural number greater than or equal to 3, such as an integer multiple of 3, 5, 7, etc., and the specific number can be adaptively modified according to cost and accuracy; M is a natural number greater than or equal to 3, such as 3, 5, 6, etc., and the specific number can be adaptively modified according to cost and accuracy.

[0070] Example 2

[0071] It should be noted that, in addition to storing the violation data packet II in the violation database for query, this embodiment also provides another specific implementation of the traffic violation behavior detection method based on the mimic architecture;

[0072] Based on Example 1, step 6 in the traffic violation detection method based on mimic architecture further includes the following steps:

[0073] When the traffic violation database receives a new traffic violation data packet II, it sends a traffic violation prompt message to the client according to the vehicle owner information in the corresponding traffic violation data packet II.

[0074] Specifically, the client can be a mobile phone or PC of the car owner, and the violation reminder information can be sent in the form of text messages or emails.

[0075] It is understandable that the violation database is configured with triggers when it is designed. When new violation data (violation data package II) is inserted, the trigger is triggered and a violation information is sent to the car owner's mobile phone number in violation data package II.

[0076] Example 3

[0077] It should be noted that after receiving the violation reminder information, the car owner can choose to log in to the client (such as 12123 app) to process the violation information; therefore, based on the above embodiment, this embodiment provides another specific implementation method of the traffic violation behavior detection method based on the mimetic architecture;

[0078] As attached Figure 4 As shown, the traffic violation detection method based on the mimic architecture further includes step 7, which includes:

[0079] Step 701: The client generates a traffic violation query request and sends the request to the input agent, which then copies and distributes the request to M business processing backends. The request includes vehicle owner information or license plate information.

[0080] Step 702: The business processing backend parses the vehicle owner information or license plate information in the violation query request, generates a database query statement, and sends the database query statement to the output arbiter of the output agent;

[0081] Step 703: the output arbitrator adjudicates the received database query statement. If the adjudication is successful, the database query statement that complies with the adjudication strategy is transmitted to the violation database.

[0082] Step 704: the violation database generates a violation query result based on the received database query statement, and copies and distributes the violation query result to M business processing backends through the output agent;

[0083] Step 705: After receiving the violation query result, the business processing backend generates a response result corresponding to the violation query request and transmits it to the input arbitrator of the input agent;

[0084] In step 706, the input arbitrator adjudicates the received response result, generates a normalized response result after passing the adjudication, and returns it to the client.

[0085] In a specific embodiment, if a car owner wants to view historical violation records, he can click on the history record to send a violation query request including a request to view the history record to the input agent;

[0086] The input agent distributes the received violation query requests to different business processing backends. Each business processing backend combines the query statements into database queries based on the specified conditions and sends them to the output agent. After the output arbiter of the output agent makes the decision and normalizes the query statement, only one query statement is sent to the violation database.

[0087] The violation database DB distributes the violation query results (queried data) to different business processing backends through the output module. After processing by the business processing backend (for example, the data format returned by the DB is different from the data format returned to the input module, the data returned by the DB needs to be reorganized according to the response format, and then the response result is returned), it is normalized through the input arbiter of the input module and the final response result is returned to the client.

[0088] It should be noted that by performing mimicry processing on database query statements and violation query results and controlling access to the violation database, the security and reliability of the violation database DB can be improved, and the violation database DB can be prevented from being illegally invaded or tampered with.

[0089] Example 4

[0090] Based on the above embodiment, this embodiment provides another specific implementation of a traffic violation detection method based on a mimic architecture;

[0091] In step 4, when the input arbitrator makes an adjudication on the extracted identification code, if the adjudication fails, the electronic eye terminal information in the identification information is extracted, and the corresponding electronic eye terminal and the violation detection module are marked as abnormal;

[0092] The input arbiter also records the electronic eye terminal information and the number of abnormalities of the violation detection module, generates maintenance information when the number of abnormalities exceeds a threshold, and sends it to the maintenance management terminal; wherein the maintenance information includes the electronic eye terminal information.

[0093] It should be noted that, through the above improvements, this embodiment can monitor in real time whether the electronic eye terminal and the corresponding violation detection module are abnormal, and automatically generate maintenance information in a timely manner to remind maintenance personnel to perform maintenance in a timely manner.

[0094] Example 5

[0095] Based on the above embodiment, this embodiment provides a specific implementation of a traffic violation detection system based on a mimic architecture, as shown in the attached Figure 3 and 4 As shown;

[0096] The traffic violation detection system based on the mimetic architecture includes a mimetic electronic eye, a violation detector, an input agent, an output agent, a heterogeneous execution platform, and a violation database. The mimetic electronic eye includes N information collection terminals, the violation detector includes N violation detection modules, the heterogeneous execution platform includes M business processing backends, the input agent is configured with an input arbitrator, and the output agent is configured with an output arbitrator.

[0097] N information collection terminals are respectively connected to the N violation detection modules in communication, the N violation detection modules are respectively connected to the input agent in communication, the input agent is respectively connected to the M business processing backends in communication, the M business processing backends are respectively connected to the output agent in communication, and the output agent is connected to the violation database in communication; wherein,

[0098] The mimic electronic eye is used to collect (capture) target image information in real time, generate identification information of the target image information, and transmit the target image information and the identification information of the target image information to the corresponding violation detection module; wherein the identification information includes license plate information, capture time information and electronic eye terminal information;

[0099] The violation detection module is configured to obtain corresponding target image information and identification information of the target image information, perform preliminary violation detection based on the corresponding target image information, and if a violation is detected, use the corresponding target image information as a violation image, encapsulate the violation image, the identification information of the violation image, and a pre-agreed violation identification code into a preliminary detection result, which is then transmitted to an input agent; if no violation is detected, encapsulate a pre-agreed non-violation identification code and the identification information of the target image information into a preliminary detection result, which is then transmitted to an input agent;

[0100] The input agent is configured to parse the identification code and identification information from the received preliminary detection result, and determine whether the parsed identification code belongs to the same set of violation detection results of the target image information based on the parsed identification information; further configured to, when it is determined that the parsed identification code belongs to the same set of violation detection results of the target image information, call the input adjudicator to adjudicate the extracted identification code, and generate a violation mark if the adjudication is passed; further configured to encapsulate the violation mark, the violation image, and the identification information of the violation image into a violation data packet I, and then copy and distribute the data packet to M business processing backends;

[0101] Wherein, the identification code includes a violation identification code and a non-violation identification code;

[0102] The business processing backend is configured to, upon receiving a violation data packet I from an input agent, obtain the owner information of the violating vehicle based on the identification information of the violation image in the violation data packet I, encapsulate the owner information and the violation data packet I into a violation data packet II, and transmit the data packet II to an output arbitrator of the output agent;

[0103] The output arbiter is used to make an adjudication on the received traffic violation data packet II. If the adjudication is successful, the traffic violation data packet II that complies with the adjudication strategy is transmitted to the traffic violation database for storage.

[0104] Furthermore, the traffic violation detection system based on the mimic architecture further includes a client terminal connected to the violation database for communication, which is used to:

[0105] Receive traffic violation prompt information from the traffic violation database, wherein the traffic violation prompt information is information generated by the traffic violation database when receiving a new traffic violation data packet II.

[0106] Furthermore, the client is also connected to the input agent for performing a violation query. When performing the violation query:

[0107] The client is further configured to generate a violation query request and send the violation query request to the input agent, which is then copied and distributed to M business processing backends via the input agent;

[0108] The business processing backend is further configured to parse the vehicle owner information or license plate information in the violation query request, generate a database query statement, and send the database query statement to the output arbiter of the output agent; and is further configured to generate a response result corresponding to the violation query request after receiving the violation query result, and transmit the response result to the input arbiter of the input agent;

[0109] The output arbiter is further configured to adjudicate the received database query statement, and if the adjudication is successful, transmit the database query statement that complies with the adjudication strategy to the violation database;

[0110] The input arbiter is further configured to adjudicate the received response result, generate a normalized response result after the adjudication is passed, and return the result to the client;

[0111] The violation database is further configured to generate violation query results based on received database query statements, and copy and distribute the violation query results to M business processing backends via the output agent.

[0112] Furthermore, when the input agent determines whether the parsed identification code belongs to the same group of violation detection results of target image information based on the parsed identification information, it is used to:

[0113] Extracting the license plate information from the identification information, determining whether the extracted license plate information is the same license plate information, and if so, extracting the electronic eye terminal information from the identification information; wherein the electronic eye terminal information includes terminal location information and terminal number;

[0114] Determine whether the extracted terminal numbers are continuous. If so, determine whether the terminal location information is in the same area. If so, determine whether the parsed identification codes belong to the same group of violation detection results of target image information.

[0115] Furthermore, the violation detection module, when performing preliminary violation detection based on the corresponding target image information, is specifically used to:

[0116] Extracting electronic eye terminal information from the identification information, wherein the electronic eye terminal information includes a terminal arrangement mode;

[0117] If the terminal arrangement in the identification information is horizontal arrangement, performing a first type of violation detection on the target image information;

[0118] If the terminal arrangement in the identification information is vertical arrangement, the second type of violation behavior detection is performed on the target image information.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or some technical features can be replaced by equivalents without departing from the spirit of the technical solutions of the present invention. They should all be included in the scope of the technical solutions claimed for protection by the present invention.

Claims

1. A traffic violation detection method based on mimicry architecture, characterized in that: The following steps are involved: Step 1: Each violation detection module obtains target image information and identification information of the target image information respectively; The identification information includes license plate information, capture time information and electronic eye terminal information; Step 2: Each violation detection module performs preliminary violation detection based on the corresponding target image information. If a traffic violation is detected, the corresponding target image information is used as a traffic violation image, and the traffic violation image, the identification information of the traffic violation image and the pre-agreed traffic violation identification code are encapsulated into a preliminary detection result, which is then transmitted to the input agent; If no violation is detected, the pre-agreed non-violation identification code and the identification information of the target image information are packaged into a preliminary detection result and transmitted to the input agent; Step 3: The input agent parses the identification code and identification information from the received preliminary detection results, and determines whether the parsed identification code belongs to the violation detection results of the same set of target image information based on the parsed identification information; wherein the identification code includes a violation identification code and a non-violation identification code, and the same set of target image information refers to images collected by the same mimetic electronic eye within the same time period, and each mimetic electronic eye includes N information collection terminals, and each information collection terminal corresponds to one violation detection module; Step 4, when it is determined that the parsed identification code belongs to the same set of violation detection results of the target image information, the extracted identification code is adjudicated by calling the input adjudicator, and if the adjudication is passed, a violation mark is generated; wherein the identification code includes a violation identification code and a non-violation identification code; The input agent encapsulates the violation mark, violation picture and identification information of the violation picture into a violation data packet I, and copies and distributes it to M business processing backends; Step 5: After receiving the traffic violation data packet I from the input agent, the business processing backend obtains the owner information of the violating vehicle based on the identification information of the violation image in the traffic violation data packet I, encapsulates the owner information and the traffic violation data packet I into a traffic violation data packet II, and transmits the data packet II to the output arbitrator of the output agent; Step 6: the output arbitrator makes an adjudication on the received violation data packet II. If the adjudication is successful, the violation data packet II that complies with the adjudication strategy is transmitted to the violation database for storage.

2. The traffic violation detection method based on mimic architecture according to claim 1 is characterized in that: The step 6 further comprises the following steps: When the traffic violation database receives a new traffic violation data packet II, it sends a traffic violation prompt message to the client according to the vehicle owner information in the corresponding traffic violation data packet II.

3. The traffic violation detection method based on mimic architecture according to claim 2 is characterized in that: The method further includes step 7, wherein the step 7 includes: Step 701: The client generates a traffic violation query request and sends the request to the input agent, which then copies and distributes the request to M business processing backends. The request includes vehicle owner information or license plate information. Step 702: The business processing backend parses the vehicle owner information or license plate information in the violation query request, generates a database query statement, and sends the database query statement to the output arbiter of the output agent; Step 703: the output arbitrator adjudicates the received database query statement. If the adjudication is successful, the database query statement that complies with the adjudication strategy is transmitted to the violation database. Step 704: the violation database generates a violation query result based on the received database query statement, and copies and distributes the violation query result to M business processing backends through the output agent; Step 705: After receiving the violation query result, the business processing backend generates a response result corresponding to the violation query request and transmits it to the input arbitrator of the input agent; In step 706, the input arbitrator adjudicates the received response result, generates a normalized response result after passing the adjudication, and returns it to the client.

4. The traffic violation detection method based on mimic architecture according to claim 1 is characterized in that: In step 4, when the input arbitrator makes an adjudication on the extracted identification code, if the adjudication fails, the electronic eye terminal information in the identification information is extracted, and the corresponding electronic eye terminal and the violation detection module are marked as abnormal; The input arbitrator also records the number of abnormalities of the electronic eye terminal and the violation detection module, generates maintenance information when the number of abnormalities exceeds a threshold, and sends it to the maintenance management terminal.

5. The traffic violation detection method based on mimicry architecture according to claim 1 is characterized in that: In step 3, when the input agent determines whether the parsed identification code belongs to the same group of violation detection results of target image information based on the parsed identification information, the following steps are executed: Extracting the license plate information from the identification information, determining whether the extracted license plate information is the same license plate information, and if so, extracting the electronic eye terminal information from the identification information; wherein the electronic eye terminal information includes terminal location information and terminal number; Determine whether the extracted terminal numbers are continuous. If so, determine whether the terminal location information is in the same area. If so, determine whether the parsed identification codes belong to the same group of violation detection results of target image information.

6. The traffic violation detection method based on mimic architecture according to claim 1 is characterized in that: In step 2, when each violation detection module performs preliminary violation detection based on the corresponding target image information, the following steps are executed: Extracting electronic eye terminal information from the identification information, wherein the electronic eye terminal information includes a terminal arrangement mode; If the terminal arrangement in the identification information is horizontal arrangement, performing a first type of violation detection on the target image information; If the terminal arrangement in the identification information is vertical arrangement, the second type of violation behavior detection is performed on the target image information.

7. A traffic violation detection system based on a mimic architecture, characterized by: It includes a mimetic electronic eye, a violation detector, an input agent, an output agent, a heterogeneous execution platform and a violation database. The mimetic electronic eye includes N information collection terminals, the violation detector includes N violation detection modules, the heterogeneous execution platform includes M business processing backends, the input agent is configured with an input arbitrator, and the output agent is configured with an output arbitrator. N information collection terminals are respectively connected to the N violation detection modules in communication, the N violation detection modules are respectively connected to the input agent in communication, the input agent is respectively connected to the M business processing backends in communication, the M business processing backends are respectively connected to the output agent in communication, and the output agent is connected to the violation database in communication; wherein, The mimic electronic eye is used to collect target image information in real time, generate identification information of the target image information, and transmit the target image information and the identification information of the target image information to the corresponding violation detection module; wherein the identification information includes license plate information, capture time information and electronic eye terminal information; The violation detection module is configured to obtain corresponding target image information and identification information of the target image information, perform preliminary violation detection based on the corresponding target image information, and if a violation is detected, use the corresponding target image information as a violation image, encapsulate the violation image, the identification information of the violation image, and a pre-agreed violation identification code into a preliminary detection result, which is then transmitted to an input agent; if no violation is detected, encapsulate a pre-agreed non-violation identification code and the identification information of the target image information into a preliminary detection result, which is then transmitted to an input agent; The input agent is configured to parse the identification code and identification information from the received preliminary detection result, and determine whether the parsed identification code belongs to the same set of violation detection results of the target image information based on the parsed identification information; further configured to, when it is determined that the parsed identification code belongs to the same set of violation detection results of the target image information, call the input adjudicator to adjudicate the extracted identification code, and generate a violation mark if the adjudication is passed; further configured to encapsulate the violation mark, the violation image, and the identification information of the violation image into a violation data packet I, and then copy and distribute the data packet to M business processing backends; The identification code includes a violation identification code and a non-violation identification code. The same set of target image information refers to images collected by the same mimetic electronic eye in the same time period. Each mimetic electronic eye includes N information collection terminals, and each information collection terminal corresponds to one violation detection module. The business processing backend is configured to, upon receiving a violation data packet I from an input agent, obtain the owner information of the violating vehicle based on the identification information of the violation image in the violation data packet I, encapsulate the owner information and the violation data packet I into a violation data packet II, and transmit the data packet II to an output arbitrator of the output agent; The output arbiter is used to make an adjudication on the received traffic violation data packet II. If the adjudication is successful, the traffic violation data packet II that complies with the adjudication strategy is transmitted to the traffic violation database for storage.

8. The traffic violation detection system based on mimicry architecture according to claim 7 is characterized in that: It also includes a client terminal connected to the violation database, which is used to: Receive traffic violation prompt information from the traffic violation database, wherein the traffic violation prompt information is information generated by the traffic violation database when receiving a new traffic violation data packet II.

9. The traffic violation detection system based on mimicry architecture according to claim 8 is characterized in that: The client is also connected to the input agent for performing violation query. When performing violation query: The client is further configured to generate a traffic violation query request and send the traffic violation query request to the input agent, which is then copied and distributed to M business processing backends via the input agent; wherein the traffic violation query request includes vehicle owner information or license plate information; The business processing backend is further configured to parse the vehicle owner information or license plate information in the violation query request, generate a database query statement, and send the database query statement to the output arbiter of the output agent; and is further configured to generate a response result corresponding to the violation query request after receiving the violation query result, and transmit the response result to the input arbiter of the input agent; The output arbiter is further configured to adjudicate the received database query statement, and if the adjudication is successful, transmit the database query statement that complies with the adjudication strategy to the violation database; The input arbiter is further configured to adjudicate the received response result, generate a normalized response result after the adjudication is passed, and return the result to the client; The violation database is further configured to generate violation query results based on received database query statements, and copy and distribute the violation query results to M business processing backends via the output agent.

10. The traffic violation detection system based on mimicry architecture according to claim 7, characterized in that: The input agent, when determining whether the parsed identification code belongs to the same group of violation detection results of target image information based on the parsed identification information, is used to: Extracting the license plate information from the identification information, determining whether the extracted license plate information is the same license plate information, and if so, extracting the electronic eye terminal information from the identification information; wherein the electronic eye terminal information includes terminal location information and terminal number; Determine whether the extracted terminal numbers are continuous. If so, determine whether the terminal location information is in the same area. If so, determine whether the parsed identification codes belong to the same group of violation detection results of target image information.

Images