Traffic violation data monitoring and early warning method and system based on intelligent identification

Abstract

The present invention discloses a traffic violation data monitoring and early warning method and system based on intelligent identification, and relates to the field of intelligent identification technology. The traffic violation data monitoring and early warning method based on intelligent identification specifically comprises the following steps: step one, identifying whether there is an illegal behavior in traffic data through an intelligent identification algorithm, and determining the illegal information in each time period; step two, calculating the illegal degree value of each time period based on the illegal information and road restriction attributes; step three, calculating the illegal degree value of historical traffic data, and determining the change of the illegal degree value; step four, realizing monitoring and early warning of traffic violation data according to the illegal degree value and the illegal degree change value; the present invention realizes monitoring and early warning of traffic violation data by intelligently identifying traffic violations, setting the illegal degree value and the illegal degree change value of each time period, and determining the correlation between traffic violations and traffic accidents.

Description

Technical Field

[0001] The present invention belongs to the technical field of traffic data monitoring, and in particular relates to a traffic violation data monitoring and early warning method and system based on intelligent identification. Background Art

[0002] The traditional traffic supervision model mainly relies on fixed surveillance cameras and traffic police patrols. However, due to the limited number and distribution of traffic police, as well as the fixed patrol routes of traffic police, there are monitoring blind spots. These blind spots are often high-incidence areas of traffic violations and traffic accidents. At the same time, manual response speed is slow, especially in complex situations such as traffic congestion and bad weather. Traffic police need to spend more time to arrive at the scene and make judgments, which further leads to the increase of traffic accidents. In addition, manual monitoring is prone to missing important events.

[0003] The current intelligent identification monitoring and early warning system, combined with AI algorithms, can achieve all-weather, all-round monitoring of traffic roads, accurately identify traffic violations such as running red lights, driving in the wrong direction, speeding, and driving in the wrong lane, thereby significantly improving the coverage and accuracy of monitoring.

[0004] How to analyze the correlation between traffic violations and traffic accidents through traffic violation data, so as to realize the monitoring and early warning of traffic violation data is an urgent problem that needs to be solved. Summary of the invention

[0005] The purpose of the present invention is to provide a traffic violation data monitoring and early warning method and system based on intelligent identification, which is used to solve the technical problem of how to analyze the correlation between traffic violations and traffic accidents through traffic violation data in the prior art.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] Traffic violation data monitoring and early warning method based on intelligent identification includes:

[0008] Step 1: Determine the monitoring area, monitor the monitoring area in real time through monitoring equipment, collect road restriction attributes in the monitoring area, extract key features from traffic data, identify whether there are illegal behaviors in the traffic data based on the key features, and determine the illegal information in each time period;

[0009] Step 2: Calculate the violation degree value for each time period based on the violation information and road restriction attributes;

[0010] Step 3: Divide the historical traffic data, and determine the change of the violation degree value by setting the violation degree change value calculation formula;

[0011] Step 4: Set up a dangerous trend formula, select the violation degree value threshold and the violation degree change value threshold, and monitor and warn traffic violation data by comparing the violation degree value, violation degree change value, violation degree value threshold, and violation degree change value threshold.

[0012] The specific contents of step one include: determining the monitoring area through the traffic monitoring platform, dividing the time period by time t, and implementing real-time monitoring of the traffic data of the monitoring area in each time period by deploying monitoring equipment in each section and area in the monitoring area, obtaining the traffic data in each time period, and collecting the road restriction attributes of each section and area in the monitoring area. The road restriction attributes include: space restriction attributes, behavior restriction attributes, and vehicle restriction attributes;

[0013] Among them, the spatial restriction attribute refers to the various restriction conditions set by departments or organizations on each road section and area within a certain space in the monitoring area, which is expressed as ,in represents the i-th restriction condition on the spatial restriction attribute of the monitoring area, and b represents a total of b restriction conditions in the monitoring area;

[0014] The behavior restriction attribute refers to the various restrictions imposed by departments or organizations on the behavior of pedestrians and drivers, which are expressed as ,in represents the i-th restriction condition on the behavior restriction attribute in the monitoring area, and c represents a total of c restriction conditions on the behavior restriction attribute;

[0015] Vehicle restriction attributes refer to various restrictions imposed on vehicles by departments or organizations, expressed as ,in represents the i-th restriction condition on vehicle restriction attribute in the monitoring area, and d represents a total of d restriction conditions on behavior restriction attribute;

[0016] The key features are extracted from the real-time monitored traffic data in order of time periods. The specific method is to set a time interval point every a seconds within a time period, collect the real-time monitored traffic data once after each time interval point, and then determine the key signs of the traffic data collected each time. The key features include: the speed, position and direction of travel of vehicles and pedestrians. The key features are matched with the restriction conditions by using the intelligent recognition algorithm combined with the road restriction attributes. When the key features do not meet the restriction conditions of the road restriction attributes, it is judged that there is an illegal behavior. Otherwise, it is judged that there is no illegal behavior, and the key features judged to have illegal behavior are marked as illegal information.

[0017] The specific contents of step 2 include: for the determined monitoring area, the corresponding relationship between the key features represented by the comprehensive violation information and the road restriction attributes is set, and the violation degree value in each time period is obtained by calculating the violation degree values ​​of all time points in each time period. The specific formula is as follows:

[0018] ;

[0019] Among them, p represents the degree of violation, n represents the nth time period, m represents the mth time point, and M represents that there are a total of M time points in a time period. Indicates the restriction degree value of the spatial restriction attribute at the mth time point in the nth time period. Indicates the restriction degree value of the behavior restriction attribute at the mth time point in the nth time period. Indicates the restriction degree value of the vehicle restriction attribute at the mth time point in the nth time period, represents the weight coefficient of the spatial restriction attribute, represents the weight coefficient of the behavior restriction attribute, Represents the weight coefficients regarding vehicle restriction attributes. These weight coefficients can be determined based on specific circumstances and needs, and are usually jointly developed and confirmed by professionals or relevant stakeholders.

[0020] Furthermore, the specific calculation formula for the restriction degree value of the space restriction attribute is as follows:

[0021] ;

[0022] in, Indicates that the restriction condition is not met at the mth time point in the nth time period. The number of times, Indicates sufficient constraints The weight coefficient of

[0023] Furthermore, the specific calculation formula for the restriction degree value of the behavior restriction attribute is as follows:

[0024] ;

[0025] in, Indicates that the restriction condition is not met at the mth time point in the nth time period. The number of times, Indicates restrictions The weight coefficient of

[0026] Furthermore, the specific calculation formula for the restriction degree value of the vehicle restriction attribute is as follows:

[0027] ;

[0028] in, Indicates that the restriction condition is not met at the mth time point in the nth time period. The number of times, Indicates restrictions The weight coefficient of .

[0029] in , and These weighting factors can be determined based on specific circumstances and needs, and are usually jointly developed and confirmed by professionals or relevant stakeholders.

[0030] The specific contents of step three include: collecting historical traffic data within the monitoring area within T time, dividing the historical traffic data into two parts, one part represents the traffic data without traffic accidents in the continuous time, recorded as safety data; the other part represents the traffic data with traffic accidents in the continuous time, recorded as accident data; wherein the continuous time of the safety data is composed of multiple adjacent time periods, the continuous time of the accident data is composed of j adjacent time periods, and a traffic accident occurred in the last time period of the adjacent time periods, calculating the violation degree values ​​of the safety data and the accident data in each time period, determining the change of the violation degree value according to the violation degree values ​​of the adjacent time periods, and setting the violation degree change value calculation formula, the specific formula is as follows:

[0031] ;

[0032] Among them, F represents the change value of the degree of violation, which is used to reflect the change of the degree of violation in different time periods. represents the nth time period, It represents the violation degree value in the nth time period, e represents the base of the logarithm of natural numbers, and r represents a constant.

[0033] The change in the violation degree value is determined by the size relationship of the violation degree change values ​​in adjacent time periods.

[0034] The specific contents of step 4 include: according to the changing trends of the violation degree value and the violation degree change value in the safety data and the accident data, setting the dangerous trend degree formula for screening the violation degree value threshold and the violation degree change value threshold. The dangerous trend degree formula for screening the violation degree value threshold is as follows:

[0035] ;

[0036] Where Q(p) represents the dangerous trend of the violation value p. It indicates the proportion of violation degree values ​​less than or equal to p in the security data to the total violation degree values. It represents the proportion of violation degree values ​​less than or equal to p in the accident data to the total violation degree values, and r represents a constant.

[0037] The formula for screening the dangerous trend degree threshold of the violation degree change value is as follows:

[0038] ;

[0039] Where Q(F) represents the dangerous trend of the violation degree change value F. It indicates the proportion of the violation degree change value less than or equal to F in the security data to the total violation degree change value. It represents the ratio of the violation degree change values ​​less than or equal to F in the accident data to the total violation degree change values, and r represents a constant.

[0040] Set the danger trend threshold QS. When Q(p) is less than or equal to QS, it means that the probability of a traffic accident occurring in the corresponding time period is small for violation level values ​​less than or equal to p. When Q(F) is less than or equal to QS, it means that the probability of a traffic accident occurring in the corresponding time period is small for violation level change values ​​less than or equal to F.

[0041] With the condition that Q(p) is less than or equal to QS, the maximum value of p that can meet the condition is screened, recorded as maxp; similarly, with the condition that Q(F) is less than or equal to QS, the maximum value of F that can meet the condition is screened, recorded as maxF;

[0042] Set maxp as the violation degree threshold, and set maxF as the violation degree change threshold. If the violation degree value of the current time period is greater than maxp, send a warning message to the traffic monitoring platform; if the violation degree change value between the current time period and the previous time period is greater than maxp, send a warning message to the traffic monitoring platform; if the violation degree value of the current time period and the violation degree change value of the previous time period are both less than or equal to the corresponding threshold, no warning message will be sent.

[0043] The present invention also provides a traffic violation data monitoring and early warning system based on intelligent identification, which mainly includes: a violation monitoring unit, a violation degree value calculation unit, a violation degree change value calculation unit and a traffic violation early warning unit;

[0044] Among them, the violation monitoring unit is used to determine the monitoring area, divide the time period by time t, monitor the traffic data of the monitoring area in real time in each time period, obtain the traffic data in each time period, collect the road restriction attributes of each section and area in the monitoring area, extract key features from the real-time monitored traffic data in the order of time periods, use intelligent recognition algorithms combined with road restriction attributes, match the key features with the road restriction attributes, and determine whether there is an illegal behavior;

[0045] The violation degree value calculation unit is used to comprehensively consider the correspondence between the key features represented by the violation information and the road restriction attributes, and set the violation degree value. The violation degree value in each time period is obtained by calculating the violation degree values ​​of all time points in each time period.

[0046] The violation degree change value calculation unit is used to divide the historical traffic data and calculate the violation degree change value according to the violation degree values ​​of adjacent time periods of the traffic data.

[0047] The traffic violation warning unit is used to screen the violation degree value threshold and the violation degree change value threshold according to the changing trends of the violation degree value and the violation degree change value in the safety data and accident data, compare the size relationship between the violation degree value, the violation degree change value and the corresponding threshold, and determine whether to send a warning message.

[0048] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0049] 1. The present invention defines road restriction attributes and extracts key features of traffic data to determine violation information, sets violation severity values ​​based on the violation information, and accurately locates time periods with high severity of violations, thus providing a scientific basis for correcting violations and improving the efficiency of resource allocation for correcting violations;

[0050] 2. The violation degree change value set by the present invention can intuitively reflect the increase and decrease trend of traffic violations by calculating the change of the violation degree values ​​in adjacent time periods, realize dynamic monitoring of traffic violations, and accurately predict the potential risks in traffic data, which is helpful to improve the road capacity and traffic management efficiency in the monitoring area;

[0051] 3. The present invention can quickly identify the degree of danger caused by traffic violations in different time periods by setting the danger trend degree and screening the thresholds of the violation degree value and the violation degree change value, thereby improving the targeted handling of traffic violations. By timely identifying and intervening in high-risk time periods caused by traffic violations, the incidence of traffic accidents can be significantly reduced, which helps to improve the safety of the road traffic environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1A step diagram of a traffic violation data monitoring and early warning method based on intelligent identification is shown;

[0054] Figure 2 A flow chart showing a method for implementing traffic violation data monitoring and early warning;

[0055] Figure 3 The block diagram of the traffic violation data monitoring and early warning system based on intelligent identification is shown. DETAILED DESCRIPTION

[0056] 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.

[0057] Embodiment 1, as Figure 1 , Figure 2 As shown, the traffic violation data monitoring and early warning method based on intelligent identification specifically includes the following steps:

[0058] Step 1: Identify whether there are illegal behaviors in the traffic data through intelligent recognition algorithms, and determine the illegal information in each time period; determine the monitoring area through the traffic monitoring platform, divide the time period by time t, and realize real-time monitoring of the traffic data in the monitoring area in each time period through the monitoring equipment deployed in each section and area in the monitoring area, obtain the traffic data in each time period, and collect the road restriction attributes of each section and area in the monitoring area. The road restriction attributes include: space restriction attributes, behavior restriction attributes, and vehicle restriction attributes;

[0059] Among them, the spatial restriction attribute refers to the various restriction conditions set by departments or organizations on each road section and area within a certain space in the monitoring area, which is expressed as ,in represents the i-th restriction condition on the spatial restriction attribute of the monitoring area, and b represents a total of b restriction conditions in the monitoring area;

[0060] The behavior restriction attribute refers to the various restrictions imposed by departments or organizations on the behavior of pedestrians and drivers, which are expressed as ,in represents the i-th restriction condition on the behavior restriction attribute in the monitoring area, and c represents a total of c restriction conditions on the behavior restriction attribute;

[0061] Vehicle restriction attributes refer to various restrictions imposed on vehicles by departments or organizations, expressed as ,in represents the i-th restriction condition on vehicle restriction attribute in the monitoring area, and d represents a total of d restriction conditions on behavior restriction attribute;

[0062] The key features are extracted from the real-time monitored traffic data in order of time periods. The specific method is to set a time interval point every a seconds within a time period, collect the real-time monitored traffic data once after each time interval point, and then determine the key signs of the traffic data collected each time. The key features include: the speed, position and direction of travel of vehicles and pedestrians. The key features are matched with the restriction conditions by using the intelligent recognition algorithm combined with the road restriction attributes. When the key features do not meet the restriction conditions of the road restriction attributes, it is judged that there is an illegal behavior. Otherwise, it is judged that there is no illegal behavior, and the key features judged to have illegal behavior are marked as illegal information.

[0063] Step 2: Based on the violation information and road restriction attributes, the violation degree value of each time period is calculated; for the determined monitoring area, the violation degree value is set based on the correspondence between the key features represented by the violation information and the road restriction attributes. The violation degree value in each time period is obtained by calculating the violation degree values ​​of all time points in each time period. The specific formula is as follows:

[0064] ;

[0065] Among them, p represents the degree of violation, n represents the nth time period, m represents the mth time point, and M represents that there are a total of M time points in a time period. Indicates the restriction degree value of the spatial restriction attribute at the mth time point in the nth time period. Indicates the restriction degree value of the behavior restriction attribute at the mth time point in the nth time period. Indicates the restriction degree value of the vehicle restriction attribute at the mth time point in the nth time period, represents the weight coefficient of the spatial restriction attribute, represents the weight coefficient of the behavior restriction attribute, Indicates the weight coefficients of the vehicle restriction attributes. These weight coefficients can be determined according to specific circumstances and needs, and are usually jointly formulated and confirmed by professionals or relevant stakeholders. In this embodiment, , , .

[0066] Furthermore, the specific calculation formula for the restriction degree value of the space restriction attribute is as follows:

[0067] ;

[0068] in, Indicates that the restriction condition is not met at the mth time point in the nth time period. The number of times, Indicates sufficient constraints The weight coefficient of

[0069] Furthermore, the specific calculation formula for the restriction degree value of the behavior restriction attribute is as follows:

[0070] ;

[0071] in, Indicates that the restriction condition is not met at the mth time point in the nth time period. The number of times, Indicates restrictions The weight coefficient of

[0072] Furthermore, the specific calculation formula for the restriction degree value of the vehicle restriction attribute is as follows:

[0073] ;

[0074] in, Indicates that the restriction condition is not met at the mth time point in the nth time period. The number of times, Indicates restrictions The weight coefficient of .

[0075] in , and These weighting factors can be determined based on specific circumstances and needs, and are usually jointly developed and confirmed by professionals or relevant stakeholders.

[0076] Step 3: Calculate the violation degree value of historical traffic data and determine the change of the violation degree value; collect historical traffic data within the monitoring area within T time, and divide the historical traffic data into two parts, one part represents traffic data without traffic accidents in the continuous time, recorded as safety data; the other part represents traffic data with traffic accidents in the continuous time, recorded as accident data; wherein the continuous time of safety data is composed of multiple adjacent time periods, the continuous time of accident data is composed of j adjacent time periods, and a traffic accident occurred in the last time period of the adjacent time periods, in this embodiment, j=2 is set, and the violation degree values ​​of safety data and accident data in each time period are calculated, and the change of the violation degree value is determined according to the violation degree values ​​of the adjacent time periods, and the calculation formula of the violation degree change value is set, and the specific formula is as follows:

[0077] ;

[0078] Among them, F represents the change value of the degree of violation, which is used to reflect the change of the degree of violation in different time periods. represents the nth time period, represents the violation degree value of the nth time period, e represents the base of the logarithm of a natural number, and r represents a constant. In this embodiment, r=1 is set.

[0079] The change in the violation degree value is determined by the size relationship of the violation degree change values ​​in adjacent time periods.

[0080] Step 4: Monitor and warn traffic violation data based on the violation degree value and the violation degree change value; set a dangerous trend degree formula based on the change trend of the violation degree value and the violation degree change value in the safety data and accident data to screen the violation degree value threshold and the violation degree change value threshold. The dangerous trend degree formula for screening the violation degree value threshold is as follows:

[0081] ;

[0082] Where Q(p) represents the dangerous trend of the violation value p. It indicates the proportion of violation degree values ​​less than or equal to p in the security data to the total violation degree values. It represents the proportion of violation degree values ​​less than or equal to p in the accident data to the total violation degree values, and r represents a constant.

[0083] The formula for screening the dangerous trend degree threshold of the violation degree change value is as follows:

[0084] ;

[0085] Where Q(F) represents the dangerous trend of the violation degree change value F. It indicates the proportion of the violation degree change value less than or equal to F in the security data to the total violation degree change value. It represents the ratio of the violation degree change values ​​less than or equal to F in the accident data to the total violation degree change values, and r represents a constant.

[0086] Set the danger trend threshold QS. When Q(p) is less than or equal to QS, it means that the probability of a traffic accident occurring in the corresponding time period is small for violation level values ​​less than or equal to p. When Q(F) is less than or equal to QS, it means that the probability of a traffic accident occurring in the corresponding time period is small for violation level change values ​​less than or equal to F.

[0087] With the condition that Q(p) is less than or equal to QS, the maximum value of p that can meet the condition is screened, recorded as maxp; similarly, with the condition that Q(F) is less than or equal to QS, the maximum value of F that can meet the condition is screened, recorded as maxF;

[0088] Set maxp as the violation degree threshold, and set maxF as the violation degree change threshold. If the violation degree value of the current time period is greater than maxp, send a warning message to the traffic monitoring platform; if the violation degree change value between the current time period and the previous time period is greater than maxp, send a warning message to the traffic monitoring platform; if the violation degree value of the current time period and the violation degree change value of the previous time period are both less than or equal to the corresponding threshold, no warning message will be sent.

[0089] Embodiment 2, as Figure 3 As shown in the figure, the traffic violation data monitoring and early warning system based on intelligent identification specifically includes the following contents:

[0090] Violation monitoring unit: through the traffic monitoring platform, determine the monitoring area, divide the time period by time t, and realize real-time monitoring of the traffic data in the monitoring area in each time period by deploying monitoring equipment in each section and area in the monitoring area, obtain traffic data in each time period, and collect road restriction attributes of each section and area in the monitoring area. Road restriction attributes include: space restriction attributes, behavior restriction attributes and vehicle restriction attributes;

[0091] Among them, the spatial restriction attribute refers to the various restrictions set by departments or organizations on the sections and areas within the monitoring area within a certain space, the behavior restriction attribute refers to the various restrictions set by departments or organizations on the behavior of pedestrians and drivers, and the vehicle restriction attribute refers to the various restrictions set by departments or organizations on moving vehicles;

[0092] The key features are extracted from the real-time monitored traffic data in order of time periods. The specific method is to set a time interval point every a seconds within a time period, collect the real-time monitored traffic data once after each time interval point, and then determine the key signs of the traffic data collected each time. The key features include: the speed, position and direction of travel of vehicles and pedestrians. The key features are matched with the restriction conditions by using the intelligent recognition algorithm combined with the road restriction attributes. When the key features do not meet the restriction conditions of the road restriction attributes, it is judged that there is an illegal behavior. Otherwise, it is judged that there is no illegal behavior, and the key features judged to have illegal behavior are marked as illegal information.

[0093] Violation degree value calculation unit: For the determined monitoring area, the corresponding relationship between the key features represented by the comprehensive violation information and the road restriction attributes is used to set the violation degree value. The violation degree value in each time period is obtained by calculating the violation degree values ​​at all time points in each time period. The specific formula is as follows:

[0094] ;

[0095] Among them, p represents the degree of violation, n represents the nth time period, m represents the mth time point, and M represents that there are a total of M time points in a time period. Indicates the restriction degree value of the spatial restriction attribute at the mth time point in the nth time period. Indicates the restriction degree value of the behavior restriction attribute at the mth time point in the nth time period. Indicates the restriction degree value of the vehicle restriction attribute at the mth time point in the nth time period, represents the weight coefficient of the spatial restriction attribute, represents the weight coefficient of the behavior restriction attribute, Represents the weight coefficient of the vehicle restriction attribute.

[0096] Violation degree change value calculation unit: collect historical traffic data within the monitoring area within T time, divide the historical traffic data into two parts, one part represents traffic data without traffic accidents in the continuous time, recorded as safety data; the other part represents traffic data with traffic accidents in the continuous time, recorded as accident data; wherein the continuous time of safety data is composed of multiple adjacent time periods, the continuous time of accident data is composed of j adjacent time periods, and a traffic accident occurred in the last time period of the adjacent time periods, in this embodiment, j=2 is set, the violation degree values ​​of safety data and accident data in each time period are calculated, the change of violation degree values ​​is determined according to the violation degree values ​​of adjacent time periods, and the violation degree change value calculation formula is set, and the specific formula is as follows:

[0097] ;

[0098] Among them, F represents the change value of the degree of violation, which is used to reflect the change of the degree of violation in different time periods. represents the nth time period, It represents the violation degree value in the nth time period, e represents the base of the logarithm of natural numbers, and r represents a constant.

[0099] Traffic violation warning unit: according to the changing trend of the violation degree value and the violation degree change value in the safety data and accident data, a dangerous trend degree formula is set to screen the violation degree value threshold and the violation degree change value threshold. The dangerous trend degree formula for screening the violation degree value threshold is as follows:

[0100] ;

[0101] Where Q(p) represents the dangerous trend of the violation value p. It indicates the proportion of violation degree values ​​less than or equal to p in the security data to the total violation degree values. It represents the proportion of violation degree values ​​less than or equal to p in the accident data to the total violation degree values, and r represents a constant.

[0102] The formula for screening the dangerous trend degree threshold of the violation degree change value is as follows:

[0103] ;

[0104] Where Q(F) represents the dangerous trend of the violation degree change value F. It indicates the proportion of the violation degree change value less than or equal to F in the security data to the total violation degree change value. It represents the ratio of the violation degree change values ​​less than or equal to F in the accident data to the total violation degree change values, and r represents a constant.

[0105] Set the danger trend threshold QS. When Q(p) is less than or equal to QS, it means that the probability of a traffic accident occurring in the corresponding time period is small for violation level values ​​less than or equal to p. When Q(F) is less than or equal to QS, it means that the probability of a traffic accident occurring in the corresponding time period is small for violation level change values ​​less than or equal to F.

[0106] With the condition that Q(p) is less than or equal to QS, the maximum value of p that can meet the condition is screened, recorded as maxp; similarly, with the condition that Q(F) is less than or equal to QS, the maximum value of F that can meet the condition is screened, recorded as maxF;

[0107] Set maxp as the violation degree threshold, and set maxF as the violation degree change threshold. If the violation degree value of the current time period is greater than maxp, send a warning message to the traffic monitoring platform; if the violation degree change value between the current time period and the previous time period is greater than maxp, send a warning message to the traffic monitoring platform; if the violation degree value of the current time period and the violation degree change value of the previous time period are both less than or equal to the corresponding threshold, no warning message will be sent.

[0108] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0109] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A traffic violation data monitoring and early warning method based on intelligent identification, characterized in that: include: Step 1: Determine the monitoring area, monitor the monitoring area in real time through monitoring equipment, collect road restriction attributes in the monitoring area, extract key features from traffic data, identify whether there are illegal behaviors in the traffic data based on the key features, and determine the illegal information in each time period; in Key features include: speed, location and direction of travel of vehicles and pedestrians; Step 2: Calculate the violation degree value for each time period based on the violation information and road restriction attributes; Step 3: Divide the historical traffic data, and determine the change of the violation degree value by setting the violation degree change value calculation formula; Using the formula Indicates the change value of the degree of violation, where F represents the change value of the degree of violation, which is used to reflect the change of the degree of violation in different time periods. represents the nth time period, represents the violation degree value of the nth time period, e represents the base of the logarithm of natural numbers, r represents a constant, and t represents the length of time for dividing the time period; Step 4. Set the dangerous trend degree formula and dangerous trend degree threshold. By comparing the dangerous trend degree of the violation degree value and the violation degree change value with the dangerous trend degree threshold, screen the violation degree value threshold and the violation degree change value threshold. By comparing the violation degree value, the violation degree change value with the violation degree value threshold and the violation degree change value threshold, realize the monitoring and early warning of traffic violation data.

2. The traffic violation data monitoring and early warning method based on intelligent identification according to claim 1 is characterized in that: Road restriction attributes, including space restriction attributes, behavior restriction attributes, and vehicle restriction attributes; Among them, the spatial restriction attribute refers to the various restriction conditions set by departments or organizations on each road section and area within a certain space in the monitoring area, which is expressed as ,in represents the i-th restriction condition on the spatial restriction attribute of the monitoring area, and b represents a total of b restriction conditions in the monitoring area; The behavior restriction attribute refers to the various restrictions imposed by departments or organizations on the behavior of pedestrians and drivers, which are expressed as ,in represents the i-th restriction condition on the behavior restriction attribute in the monitoring area, and c represents a total of c restriction conditions on the behavior restriction attribute; Vehicle restriction attributes refer to various restrictions imposed on vehicles by departments or organizations, expressed as ,in represents the i-th restriction condition on vehicle restriction attribute in the monitoring area, and d represents a total of d restriction conditions on behavior restriction attribute.

3. The traffic violation data monitoring and early warning method based on intelligent identification according to claim 1 is characterized in that: Based on the violation information and road restriction attributes, the violation degree value of each time period is calculated. The specific method is as follows: Using the formula represents the degree of violation, where p represents the degree of violation, n represents the nth time period, m represents the mth time point, M represents a total of M time points in a time period, Kj represents the degree of restriction value for the spatial restriction attribute, Xw represents the degree of restriction value for the behavior restriction attribute, Cl represents the degree of restriction value for the vehicle restriction attribute, represents the weight coefficient of the spatial restriction attribute, represents the weight coefficient of the behavior restriction attribute, Represents the weight coefficient of the vehicle restriction attribute.

4. The method for monitoring and early warning of traffic violation data based on intelligent identification according to claim 3 is characterized in that: Regarding the restriction degree values ​​of the spatial restriction attribute, the behavior restriction attribute and the vehicle restriction attribute, specifically include: Using the formula Determine the restriction degree value for the spatial restriction attribute, where Indicates that the restriction condition is not met at the mth time point in the nth time period. The number of times, Indicates sufficient constraints The weight coefficient of Using the formula Represents the restriction degree value of the behavior restriction attribute, where: Indicates that the restriction condition is not met at the mth time point in the nth time period. The number of times Indicates restrictions The weight coefficient of Using the formula Represents the restriction degree value of the vehicle restriction attribute, where: Indicates that the restriction condition is not met at the mth time point in the nth time period. The number of times, Indicates restrictions The weight coefficient of .

5. The method for monitoring and early warning of traffic violation data based on intelligent identification according to claim 1 is characterized in that: The historical traffic data is divided into the following methods: The historical traffic data is divided into two parts. One part represents the traffic data without traffic accidents in the continuous time, which is recorded as safety data; the other part represents the traffic time when traffic accidents occur in the continuous time, which is recorded as accident data. The continuous time of safety data is composed of multiple adjacent time periods without traffic accidents, and the continuous time of accident data is composed of j adjacent time periods, and a traffic accident occurs in the last time period of the adjacent time period.

6. The method for monitoring and early warning of traffic violation data based on intelligent identification according to claim 1 is characterized in that: Set the dangerous trend degree formula, the specific method is: According to the changing trend of the violation degree value and the violation degree change value in the safety data and accident data, the dangerous trend degree formula is set, and the formula is used represents the dangerous trend of the violation value, where Q(p) represents the dangerous trend of the violation value p, It indicates the proportion of violation degree values ​​less than or equal to p in the security data to the total violation degree values. It represents the proportion of the violation degree values ​​less than or equal to p in the accident data to the total violation degree values, and r represents a constant; Using the formula Indicates the dangerous trend of the violation degree change value, where Q(F) represents the dangerous trend of the violation degree change value F, It indicates the proportion of the violation degree change value less than or equal to F in the security data to the total violation degree change value. It represents the ratio of the violation degree change values ​​less than or equal to F in the accident data to the total violation degree change values, and r represents a constant.

7. The method for monitoring and early warning of traffic violation data based on intelligent identification according to claim 6 is characterized in that: The threshold of the degree of violation and the threshold of the degree of violation change are screened as follows: Set the danger trend threshold QS, and take Q(p) less than or equal to QS as the condition to screen the maximum value of p that can meet the condition, recorded as maxp; similarly, take Q(F) less than or equal to QS as the condition to screen the maximum value of F that can meet the condition, recorded as maxF; Set maxp as the violation degree threshold, and maxF as the violation degree change threshold. If the violation degree value in the current time period is greater than maxp, send an early warning message to the traffic monitoring platform; if the violation degree change value between the current time period and the previous time period is greater than maxp, send an early warning message to the traffic monitoring platform.

8. A traffic violation data monitoring and early warning system based on intelligent identification, applied to a traffic violation data monitoring and early warning method based on intelligent identification as claimed in any one of claims 1 to 7, characterized in that: include: The violation monitoring unit is used to determine the monitoring area, divide the time period by time t, monitor the traffic data of the monitoring area in real time in each time period, obtain the traffic data in each time period, collect the road restriction attributes of each section and area in the monitoring area, extract key features from the real-time monitored traffic data in the order of time periods, use intelligent recognition algorithms combined with road restriction attributes, match the key features with the road restriction attributes, and determine whether there is an illegal behavior; The violation degree value calculation unit is used to comprehensively consider the correspondence between the key features represented by the violation information and the road restriction attributes, and set the violation degree value. The violation degree value in each time period is obtained by calculating the violation degree values ​​of all time points in each time period. A violation degree change value calculation unit, used to divide the historical traffic data and calculate the violation degree change value according to the violation degree values ​​of adjacent time periods of the traffic data; Using the formula Indicates the change value of the degree of violation, where F represents the change value of the degree of violation, which is used to reflect the change of the degree of violation in different time periods. represents the nth time period, represents the violation degree value of the nth time period, e represents the base of the logarithm of natural numbers, r represents a constant, and t represents the length of time for dividing the time period; The traffic violation warning unit is used to screen the violation degree value threshold and the violation degree change value threshold according to the changing trends of the violation degree value and the violation degree change value in the safety data and accident data. The safety data refers to the traffic data without traffic accidents in a continuous time, and the accident data refers to the traffic data with traffic accidents in a continuous time. The violation degree value, the violation degree change value and the corresponding threshold are compared to determine whether to send a warning message.

Images