Traffic violation personnel management and control early warning method

Abstract

The invention relates to the technical field of big data processing, in particular to a traffic violation personnel management and control early warning method, which comprises the steps of collecting multi-dimensional data for a to-be-assessed person, and classifying the data to obtain an event theme; a behavior fact table is obtained according to event subject classification; generating a figure portrait corresponding to the to-be-evaluated person according to the behavior fact table; and calculating a risk score of the to-be-assessed person according to the figure portrait, and then generating early warning information according to the risk score. Aiming at the problem that continuous illegal behaviors cannot be effectively distinguished in a big data analysis process in the prior art, a process of acquiring all associated multi-dimensional data of a person to be evaluated, then classifying the data according to an event theme and storing the data into a corresponding behavior fact table is introduced; according to the method, effective analysis and screening of the multi-dimensional data of the to-be-evaluated person before and after the illegal event are realized, finally, the figure portrait is generated according to the behavior fact table, the risk score is calculated, and effective screening of continuous illegal persons is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of big data processing, and in particular to a method for early warning of traffic violators. Background Art

[0002] With the increasing development of social economy, the total number of motor vehicles in cities is increasing, and the number of motor vehicle-related Internet businesses is also increasing (online car-hailing, intra-city freight, intra-city moving, etc.). Therefore, making full use of urban traffic big data analysis technology to warn and control urban road hazards is an indispensable task in the current transportation field.

[0003] In the prior art, there are technical solutions for assisting the analysis of traffic violations based on big data analysis. For example, Chinese patent CN202110275271.8 discloses a method for detecting traffic violations of non-motor vehicles and drivers and a system for executing the method. The method collects multiple consecutive video frames in the image to be detected; and determines the driving area of ​​the non-motor vehicle in the video frame and the vehicle information of the non-motor vehicle in the area and the biometric information of the driver, and determines the driving trajectory of the non-motor vehicle and the identity information of the driver based on the information of the non-motor vehicle and the driver; and further determines whether the non-motor vehicle and the driver have committed illegal acts. By using artificial intelligence, big data and other technologies to analyze and process the acquired high-position video images, non-motor vehicles can be correctly detected and identified in road traffic monitoring videos, and data support for violations can be provided to traffic control departments, thereby improving the supervision of non-motor vehicles, and assisting in the control of non-motor vehicle travel to ensure traffic safety.

[0004] However, during the actual implementation process, the inventors found that this type of technical solution can usually only punish a single violation during implementation, such as not wearing a helmet, running a red light, etc., and it is difficult to effectively analyze violations over a period of time, such as repeat offenders, scalpers deducting points, etc. Summary of the invention

[0005] In view of the above problems existing in the prior art, a method for early warning of traffic violators is now provided.

[0006] The specific technical solutions are as follows:

[0007] A traffic violator control and early warning method, comprising:

[0008] Step S1: For the person to be evaluated, collect multi-dimensional data and classify them to obtain event themes;

[0009] Step S2: Obtaining a behavior fact table according to the event subject classification;

[0010] Step S3: generating a character portrait corresponding to the person to be evaluated according to the behavior fact table;

[0011] Step S4: Calculate the risk score of the person to be assessed according to the character portrait, and then generate warning information according to the risk score.

[0012] On the other hand, the step S1 comprises:

[0013] Step S11: Determine the person to be evaluated, and collect the multi-dimensional data from an external data platform;

[0014] The multi-dimensional data includes basic information data of driver's license personnel, motor vehicle violation data, vehicle trajectory data, and motor vehicle registration information data;

[0015] Step S12: Divide the multi-dimensional data into traffic violation event sub-topics and traffic violation post-event sub-topics as the event topics.

[0016] On the other hand, in step S11, the multi-dimensional data is stored in the data lake;

[0017] In the step S12, in the data lake, the multi-dimensional data is sequentially divided into traffic violation topics under the traffic topic domain;

[0018] The traffic violation topic includes the traffic violation event sub-topic and the traffic violation event post-event sub-topic.

[0019] On the other hand, the step S2 comprises:

[0020] Step S21: creating a fact table entity according to the personnel to be evaluated;

[0021] Step S22: preprocessing the multi-dimensional data to obtain preprocessed data;

[0022] Step S23: In the fact table entity, the pre-processed data is extracted according to the traffic violation event sub-topic and the traffic violation event post-sub-topic and filled into the traffic violation event offender fact table and the traffic violation event post-offender fact table.

[0023] On the other hand, the step S23 comprises:

[0024] Step S231: processing the preprocessed data respectively to generate a primary label corresponding to each group of the preprocessed data;

[0025] Step S232: generating a secondary label for the pre-processed data according to the primary label, the time period and the first business dimension;

[0026] Step S233: generating the pre-processed third-level label according to the second-level label, the time period and the second business dimension;

[0027] Step S234: in the fact table entity, generating the traffic violation offender fact table during the traffic violation and the traffic violation offender fact table after the traffic violation according to the traffic violation sub-topic and the traffic violation sub-topic after the traffic violation respectively;

[0028] Step S234: Fill in the pre-processed data and the corresponding first-level label, second-level label and third-level label in the fact table of offenders during the traffic violation or the fact table of offenders after the traffic violation.

[0029] On the other hand, the step S3 comprises:

[0030] Step S31: creating the character portrait for the person to be evaluated;

[0031] The character portraits include portraits during the event and portraits after the event;

[0032] Step S32: Read the fact table of traffic violation offenders during the incident and the fact table of traffic violation offenders after the incident for the incident portrait and the post-incident portrait respectively, so as to obtain the corresponding first-level label, the second-level label and the third-level label for filling.

[0033] On the other hand, the step S4 comprises:

[0034] Step S41: obtaining a plurality of portrait tags corresponding to the character portrait;

[0035] Step S42: Calculating label weights for each of the plurality of portrait labels;

[0036] Step S43: weighting the image tags according to the tag weights to obtain the risk score;

[0037] Step S44: Generate the warning information according to the risk score and the risk score threshold.

[0038] On the other hand, in step S42, the method for generating the tag weight includes:

[0039]

[0040] Where N i (i=1, 2, 3) respectively correspond to the number of the first-level label, the second-level label and the third-level label; W i (i=1, 2, 3) correspond to the label weights of the primary label, the secondary label and the tertiary label respectively.

[0041] On the other hand, in step S43, the method for calculating the risk score includes:

[0042]

[0043] In the formula, Score m Score the risk, X n is the image label, W i (i=1, 2, 3) is the label weight.

[0044] The above technical solution has the following advantages or beneficial effects:

[0045] In view of the problem that the big data analysis process in the existing technology is usually only performed on single illegal behaviors and cannot effectively identify continuous illegal behaviors, this solution introduces a process of obtaining all relevant multi-dimensional data of the person to be evaluated, and then classifying them according to the subject of the event and storing them in the corresponding behavior fact table. This realizes the effective analysis and screening of the multi-dimensional data of the person to be evaluated before and after the illegal event, and finally generates a character portrait based on the behavior fact table and calculates the risk score, thereby realizing the effective screening of continuous offenders. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The embodiments of the present invention will be described more fully with reference to the attached drawings, which are provided for illustration and description only and are not intended to limit the scope of the present invention.

[0047] Figure 1 It is an overall schematic diagram of an embodiment of the present invention;

[0048] Figure 2 This is a schematic diagram of step S1 in an embodiment of the present invention;

[0049] Figure 3 This is a schematic diagram of step S2 in an embodiment of the present invention;

[0050] Figure 4 This is a schematic diagram of step S23 in an embodiment of the present invention;

[0051] Figure 5 This is a schematic diagram of step S3 in an embodiment of the present invention;

[0052] Figure 6 Schematic diagram of step S4 in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0055] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0056] The present invention comprises:

[0057] A traffic violator control and early warning method, such as Figure 1 As shown, including:

[0058] Step S1: For the person to be evaluated, collect multi-dimensional data and classify them to obtain event themes;

[0059] Step S2: Obtain a behavior fact table according to event subject classification;

[0060] Step S3: Generate a character portrait corresponding to the person to be evaluated according to the behavior fact table;

[0061] Step S4: Calculate the risk score of the person to be assessed based on the character portrait, and then generate warning information based on the risk score.

[0062] Specifically, in view of the problem that the big data analysis process in the prior art is usually only performed on a single illegal behavior and is unable to effectively identify continuous illegal behaviors, in this embodiment, a process is introduced to obtain all relevant multi-dimensional data of the person to be evaluated, and then classify them according to the subject of the event and store them in the corresponding behavior fact table. This achieves effective analysis and screening of the multi-dimensional data of the person to be evaluated before and after the illegal event, and finally generates a character portrait based on the behavior fact table, and calculates the risk score, thereby achieving effective screening of continuous offenders.

[0063] In one embodiment, Figure 2 As shown, step S1 includes:

[0064] Step S11: determine the person to be evaluated and collect multi-dimensional data from an external data platform;

[0065] Multi-dimensional data includes basic information data of driving license personnel, motor vehicle violation data, vehicle trajectory data, and motor vehicle registration information data;

[0066] Step S12: Divide the multi-dimensional data into traffic violation event sub-topics and traffic violation post-event sub-topics as event topics.

[0067] Specifically, in view of the problem that the big data analysis process in the prior art is usually only performed on a single violation and cannot effectively distinguish continuous violations, in this embodiment, for the person to be evaluated, data from multiple sources are pulled from external data platforms, including traffic management platforms, vehicle management platforms, etc., around the person to be evaluated, including basic information data of driver's license personnel, motor vehicle violation data, vehicle trajectory data, and motor vehicle registration information data. Then, in order to facilitate subsequent continuous analysis of the same person to be evaluated, the multi-dimensional data is divided into traffic violation sub-themes and traffic violation post-event sub-themes as event themes according to the time when the behavior of the person to be evaluated occurred.

[0068] Among them, motor vehicle violation data is mainly obtained through multiple sets of violation penalty data collected by the traffic police platform. Through the associated motor vehicle license plate and driver information, the basic information data of the driver and the motor vehicle registration information data can be pulled. In some areas, the vehicle trajectory data can also be determined according to the monitored license plate in combination with the road network analysis platform.

[0069] In the subsequent analysis process, for each violation, the relevant information during the event and the relevant information after the event can be analyzed separately, which is convenient for effective screening of single violations and continuous violations.

[0070] In one embodiment, in step S11, the multi-dimensional data is stored in a data lake;

[0071] In step S12, in the data lake, the multi-dimensional data is divided into traffic violation topics under the traffic topic domain in turn;

[0072] The traffic violation theme includes the traffic violation event sub-theme and the traffic violation event sub-theme.

[0073] Specifically, in order to achieve better storage effects for large amounts of data, in this embodiment, a data lake is first constructed, which is embodied as a database system for storing big data. Subsequently, a traffic theme domain is created in the data lake for the traffic violation monitoring scenario. The traffic theme domain is more finely divided into traffic violation event sub-themes and traffic violation post-event sub-themes according to the business process, thereby achieving a more detailed storage process.

[0074] In one embodiment, Figure 3 As shown, step S2 includes:

[0075] Step S21: Create a fact table entity according to the person to be evaluated;

[0076] Step S22: preprocessing the multi-dimensional data to obtain preprocessed data;

[0077] Step S23: In the fact table entity, the pre-processed data is extracted according to the traffic violation event sub-topic and the traffic violation event post-sub-topic and filled into the traffic violation event offender fact table and the traffic violation event post-offender fact table.

[0078] Specifically, in order to achieve a better analysis effect on the collected multi-dimensional data, in this embodiment, for the person to be evaluated who currently needs to be evaluated, first create a fact table entity, then generate a traffic violation in-process offender fact table and a traffic violation in-process offender fact table for the traffic violation in-process sub-topic and the traffic violation in-process sub-topic respectively under the table entity, forming a traffic violation in-process table corresponding to the current behavior and a traffic violation in-process table associated with the same person's past violations. On this basis, the multi-dimensional data is pre-processed and filled, achieving a better association process.

[0079] The preprocessing process includes cleaning, standardization, filtering out useless data, unifying data dictionary specifications and name specifications, etc.

[0080] In one embodiment, Figure 4 As shown, step S23 includes:

[0081] Step S231: processing the preprocessed data separately to generate a primary label corresponding to each group of preprocessed data;

[0082] Step S232: Generate a secondary tag for the pre-processed data according to the primary tag, the time period and the first business dimension;

[0083] Step S233: Generate a pre-processed third-level label according to the second-level label, time period and the second business dimension;

[0084] Step S234: in the fact table entity, generating a traffic violation offender fact table during the traffic violation and a traffic violation offender fact table after the traffic violation according to the traffic violation sub-topic and the traffic violation sub-topic;

[0085] Step S235: Fill in the pre-processed data and the corresponding primary label, secondary label and tertiary label in the fact table of traffic violation offenders during the traffic violation or the fact table of traffic violation offenders after the traffic violation.

[0086] Specifically, in order to facilitate different divisions of data and the subsequent portrait generation process, in this embodiment, after the fact table entity is created, the fact table of offenders during the traffic violation and the fact table of offenders after the traffic violation are generated according to the sub-topics of the traffic violation during the traffic violation and the sub-topics of the traffic violation after the traffic violation. At the same time, for each behavior in the fact table of offenders during the traffic violation and the fact table of offenders after the traffic violation, multiple groups of label fields are configured to analyze the entry behavior, so as to achieve a better representation of a certain illegal behavior.

[0087] During the characterization process, in order to achieve better characterization effects, the label field is divided into primary fields, secondary fields, and tertiary fields of different granularities.

[0088] Among them, the first-level field, second-level field and third-level field are set for different granularity rules. For example, the first-level label refers to the label that can be directly judged from the original data. This label is the label that can be directly obtained from the original table; the second-level label refers to the label obtained through the first-level label, time period, and business dimension definition; the third-level label is based on the second-level label, and the business dimension is subdivided to the atomic level business dimension. The overall structure is the time period, the second-level label, and the atomic business dimension.

[0089] For example, the first-level label: whether there have been traffic violations; the second-level label: whether there have been traffic violations in the past 30 days; the third-level label: whether there have been traffic violations such as drunk driving in the past 30 days, whether there have been traffic violations such as running red lights in the past 30 days, whether there have been traffic violations such as driving with fake license plates in the past 30 days, etc.

[0090] Therefore, by analyzing the preprocessed data according to the above process to obtain different fields, it is possible to perform correlation analysis on behaviors according to different business rules and store them in the corresponding fact table.

[0091] In one embodiment, Figure 5 As shown, step S3 includes:

[0092] Step S31: Creating a character portrait for the person to be evaluated;

[0093] Character portraits include portraits during the event and portraits after the event;

[0094] Step S32: Read the fact table of traffic offenders during the incident and the fact table of traffic offenders after the incident for the in-incident portrait and the post-incident portrait respectively to obtain the corresponding first-level labels, second-level labels and third-level labels for filling.

[0095] Specifically, after creating the above-mentioned labels, during analysis, a character portrait can be directly created for the person to be evaluated, including an in-event portrait and a post-event portrait. Subsequently, the fact table of offenders during the traffic violation and the fact table of offenders after the traffic violation are read respectively to obtain the corresponding first-level labels, second-level labels, and third-level labels for filling, thereby forming a character portrait with multiple labels.

[0096] In one embodiment, Figure 6 As shown, step S4 includes:

[0097] Step S41: obtaining multiple portrait tags corresponding to the character portrait;

[0098] Step S42: Calculate label weights for each of the multiple portrait labels;

[0099] Step S43: weighting the image tags according to the tag weights to obtain a risk score;

[0100] Step S44: Generate warning information according to the risk score and the risk score threshold.

[0101] Specifically, after constructing the character portrait, in order to achieve a better early warning effect, we first obtain multiple portrait tags corresponding to the character portrait, and then dynamically calculate the average weight of each type of tag in the portrait tag set of each type of business process according to the number of first-level, second-level, and third-level tags. Subsequently, we weight the portrait tags according to the tag weights to obtain the risk score, and make a judgment based on the pre-calibrated risk score threshold, so as to effectively screen high-risk personnel to be evaluated.

[0102] In one embodiment, in step S42, the method for generating tag weights includes:

[0103]

[0104] Where N i (i=1,2,3) corresponds to the number of primary tags, secondary tags and tertiary tags respectively; W i (i=1,2,3) corresponds to the label weights of the first-level label, the second-level label and the third-level label respectively.

[0105] In one embodiment, in step S43, the method for calculating the risk score includes:

[0106]

[0107] In the formula, Score m For risk score, X n is the image label, W i (i=1,2,3) is the label weight.

[0108] The above are only preferred embodiments of the present invention, and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. A traffic violator control and early warning method, characterized in that: include: Step S1: For the person to be evaluated, collect multi-dimensional data and classify them to obtain event themes; Step S2: Obtaining a behavior fact table according to the event subject classification; Step S3: generating a character portrait corresponding to the person to be evaluated according to the behavior fact table; Step S4: Calculate the risk score of the person to be assessed according to the character portrait, and then generate warning information according to the risk score.

2. The traffic violator control and early warning method according to claim 1 is characterized in that: The step S1 comprises: Step S11: Determine the person to be evaluated, and collect the multi-dimensional data from an external data platform; The multi-dimensional data includes basic information data of driver's license personnel, motor vehicle violation data, vehicle trajectory data, and motor vehicle registration information data; Step S12: Divide the multi-dimensional data into traffic violation event sub-topics and traffic violation post-event sub-topics as the event topics.

3. The traffic violator control and early warning method according to claim 2 is characterized in that: In the step S11, the multi-dimensional data is stored in the data lake; In the step S12, in the data lake, the multi-dimensional data is sequentially divided into traffic violation topics under the traffic topic domain; The traffic violation topic includes the traffic violation event sub-topic and the traffic violation event post-event sub-topic.

4. The traffic violator control and early warning method according to claim 2 is characterized in that: The step S2 comprises: Step S21: creating a fact table entity according to the personnel to be evaluated; Step S22: preprocessing the multi-dimensional data to obtain preprocessed data; Step S23: In the fact table entity, the pre-processed data is extracted according to the traffic violation event sub-topic and the traffic violation event post-sub-topic and filled into the traffic violation event offender fact table and the traffic violation event post-offender fact table.

5. The traffic violator control and early warning method according to claim 4 is characterized in that: The step S23 comprises: Step S231: processing the preprocessed data respectively to generate a primary label corresponding to each group of the preprocessed data; Step S232: generating a secondary label for the pre-processed data according to the primary label, the time period and the first business dimension; Step S233: generating the pre-processed third-level label according to the second-level label, the time period and the second business dimension; Step S234: in the fact table entity, generating the traffic violation offender fact table during the traffic violation and the traffic violation offender fact table after the traffic violation according to the traffic violation sub-topic and the traffic violation sub-topic after the traffic violation respectively; Step S235: Fill in the pre-processed data and the corresponding first-level label, second-level label and third-level label in the fact table of offenders during the traffic violation or the fact table of offenders after the traffic violation.

6. The traffic violator control and early warning method according to claim 5 is characterized in that: The step S3 comprises: Step S31: creating the character portrait for the person to be evaluated; The character portraits include portraits during the event and portraits after the event; Step S32: Read the fact table of traffic violation offenders during the incident and the fact table of traffic violation offenders after the incident for the incident portrait and the post-incident portrait respectively, so as to obtain the corresponding first-level label, the second-level label and the third-level label for filling.

7. The traffic violator control and early warning method according to claim 1 is characterized in that: The step S4 comprises: Step S41: obtaining a plurality of portrait tags corresponding to the character portrait; Step S42: Calculating label weights for each of the plurality of portrait labels; Step S43: weighting the image tags according to the tag weights to obtain the risk score; Step S44: Generate the warning information according to the risk score and the risk score threshold.

8. The traffic violator control and early warning method according to claim 7 is characterized in that: In step S42, the method for generating the tag weight includes: Where N i (i=1, 2, 3) respectively correspond to the number of the first-level label, the second-level label and the third-level label; W i (i=1, 2, 3) correspond to the label weights of the primary label, the secondary label and the tertiary label respectively.

9. The traffic violator control and early warning method according to claim 8 is characterized in that: In step S43, the method for calculating the risk score includes: In the formula, Score m Score the risk, X n is the image label, W i (i=1, 2, 3) is the label weight.

Images