Traffic light real-time management system based on big data

Abstract

The invention discloses a traffic light real-time management system based on big data, relates to the technical field of traffic management, and solves the problems that an existing traffic light real-time management system is difficult to accurately calculate a crowd passing density value and a vehicle passing density value according to traffic data and comprehensively calculate the passing density; safety risks are analyzed deeply according to the behavior data, and safety risk indexes are calculated; and meanwhile, the influence coefficient is difficult to comprehensively calculate by effectively utilizing various data and early warning information is difficult to generate to adjust and control the traffic light in real time. The system comprises the following modules: a data acquisition module, a data processing module, a data analysis module and an early warning management module. And further calculating the passing density of the pedestrian crosswalk by calculating the crowd passing density value and the vehicle passing density value. And calculating a security risk index according to the analyzed security risk. And finally, influence coefficients are calculated by integrating various data, early warning information is generated, and traffic lights are adjusted and controlled in real time according to the early warning information.

Description

Technical Field

[0001] The present invention belongs to the technical field of traffic management, and in particular is a real-time traffic light management system based on big data. Background Art

[0002] With the development of big data technology and the need for traffic management, traditional traffic signal management systems have been unable to cope with increasingly complex traffic conditions. The present invention pre-processes and analyzes traffic data on crosswalks, calculates the density of people and vehicles, and further analyzes the density of traffic through crosswalks. Based on the analysis of multiple safety risks, the safety risk index is calculated. Finally, the impact coefficient is calculated by combining the traffic density and the safety risk index, and early warning information is generated, and the traffic lights are adjusted and controlled in real time accordingly.

[0003] The existing real-time traffic light management system makes it difficult to calculate crowd density and vehicle density based on traffic data; it is difficult to comprehensively calculate traffic density based on crowd density and vehicle density; it is difficult to deeply analyze the safety risks of behavioral data and calculate the safety risk index; at the same time, it is difficult to effectively use various data to comprehensively calculate the impact coefficient and generate early warning information to adjust and control traffic lights in real time. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a real-time traffic light management system based on big data, which is used to solve the technical problems that the existing real-time traffic light management system is difficult to accurately calculate the crowd density value and the vehicle density value based on traffic data and comprehensively calculate the traffic density; it is difficult to deeply analyze the safety risks and calculate the safety risk index based on the behavior data; at the same time, it is difficult to effectively use various data to comprehensively calculate the influence coefficient and generate early warning information to adjust and control the traffic lights in real time.

[0005] To solve the above problems, the first aspect of the present invention provides a real-time traffic light management system based on big data, comprising the following modules:

[0006] Data collection module: collects video image data on the crosswalk, including traffic data and behavior data;

[0007] Data processing module: pre-process the collected video image data, including: denoising and image enhancement; pre-process the traffic data collected from the video image data, including: removing outliers, filling missing values, data type conversion, unified standards and normalization;

[0008] Data analysis module: analyze the traffic data in the pre-processed video image data; calculate the crowd density value and vehicle density value when passing through the crosswalk based on the analysis results; comprehensively analyze and calculate the traffic density of the crosswalk; analyze the safety risk when passing through the crosswalk based on the behavior data in the pre-processed video image data; calculate the potential value of the safety risk based on the analysis results; comprehensively analyze and calculate the safety risk index of passing through the crosswalk;

[0009] Early warning management module: Based on comprehensive analysis of traffic density and safety risk index, calculate the impact coefficient and generate early warning information to adjust and control traffic lights in real time.

[0010] As a further solution of the present invention: analyzing the traffic data in the pre-processed video image data comprises the following steps:

[0011] According to the pre-processed video images, the traffic data in the video images are analyzed by a deep learning model;

[0012] Traffic data for marking crosswalks include: the position of pedestrians and non-motor vehicles at the crosswalks, the speed of pedestrians and non-motor vehicles, the longitudinal spacing and occupied width of pedestrians and non-motor vehicles when passing;

[0013] Input the labeled crosswalk traffic data into the deep learning model for model training;

[0014] The real-time traffic light management system collects and processes video images in real time, measures and outputs the positions of pedestrians and non-motor vehicles at crosswalks, the speeds of pedestrians and non-motor vehicles, and the longitudinal spacing and occupied width of pedestrians and non-motor vehicles when passing.

[0015] As a further solution of the present invention, calculating the density value of people and vehicles passing through the crosswalk according to the analysis results includes the following steps:

[0016] By analyzing the formula:

[0017] The density value P of the crowd passing through the crosswalk is obtained; VP represents the walking speed of pedestrians, S P Indicates the longitudinal distance between pedestrians when walking, b P Indicates the lateral width occupied by pedestrians, T 绿 Indicates the duration of the green light, K P represents the correction factor;

[0018] By analyzing the formula:

[0019] Get the vehicle traffic density value C passing through the crosswalk; where V C represents the speed of non-motor vehicles, SC Indicates the longitudinal spacing of non-motor vehicles when passing, b C Indicates the lateral width occupied by non-motor vehicles, T 绿 Indicates the duration of the green light, K C Indicates the correction factor.

[0020] As a further solution of the present invention: a comprehensive analysis and calculation of the traffic density through the pedestrian crossing includes the following steps:

[0021] By analyzing the formula: F=P+C

[0022] The traffic density F passing through the crosswalk is obtained, where P and C represent the traffic density value of people and the traffic density value of vehicles respectively.

[0023] As a further solution of the present invention: analyzing the safety risk of crossing a crosswalk according to the behavior data in the preprocessed video image data, including the following steps:

[0024] Based on the behavioral data in the preprocessed video image data, the safety risks of the pedestrian crossings in the video images are analyzed through the deep learning model;

[0025] Mark the crosswalk i security risks; where i∈{1, 2, ...n};

[0026] Mark the first i The security risks are input into the convolutional neural network of the deep learning model for model training;

[0027] The real-time traffic light management system collects and processes video images in real time to identify the first pedestrian crossing. i A security risk.

[0028] As a further solution of the present invention: marking the ath i The security risks include:

[0029] a1: running a red light; a2: driving on a non-pedestrian crossing; a3: pedestrians selling items in the crosswalk; a4: children and pets driving alone; a5: riding a non-motor vehicle without a helmet; a6: pedestrians walking while looking at their phones; a7: pedestrians chasing others in the crosswalk; a8: vehicles not giving way to pedestrians; a9: vehicles occupying the crosswalk; a10: pedestrians driving on the sidewalk without a helmet; a11: pedestrians driving on the sidewalk without a helmet; a12: pedestrians driving on the sidewalk without a helmet; a13: pedestrians driving on the sidewalk without a helmet; a14: pedestrians driving on the sidewalk without a helmet; a15: pedestrians driving on the sidewalk without a helmet; a16: pedestrians driving on the sidewalk while looking at their phones; a17: pedestrians driving on the sidewalk without a helmet; a18: pedestrians driving on the sidewalk without a helmet; a19: pedestrians driving on the sidewalk without a helmet; a 10 Vehicle speeding.

[0030] As a further solution of the present invention: calculating the potential value of the security risk according to the analysis result includes the following steps:

[0031] By analyzing the formula:

[0032] Get the first i The potential value of security risk α i ; Where A represents the ath i The total number of times a safety risk occurs; T represents the total number of days in a year, and F represents the traffic density.

[0033] As a further solution of the present invention: comprehensively analyzing and calculating the safety risk index of passing through a crosswalk, comprising the following steps:

[0034] By analyzing the formula:

[0035] Get the security risk index ω; where α i Indicates the first i The potential value of security risk, W i Indicates the first i The weight of the security risk.

[0036] As a further solution of the present invention: the influence coefficient is calculated based on the comprehensive analysis of traffic density and safety risk index and warning information is generated to adjust and control the traffic lights in real time, including the following steps:

[0037] By analyzing the formula:

[0038] The influence coefficient δ is obtained; where F represents the traffic density and ω represents the safety risk index;

[0039] According to the value of the impact coefficient δ, different levels of thresholds are set, and the warning is divided into three levels: high-level warning, medium-level warning and low-level warning:

[0040] When δ≥60%, the warning is classified as a high-level warning, a red light is set, a serious warning signal is issued, and staff are summoned to deal with the safety risk;

[0041] When 30%<δ<60%, the warning is classified as a medium warning, and the green light time on the crosswalk is increased;

[0042] When δ≤30%, the warning is classified as a low-level warning, maintaining the normal traffic light cycle time, and continuing to detect traffic density and safety risk index in real time.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention discloses a real-time traffic light management system based on big data, which can obtain the crowd density value and vehicle density value when passing through the crosswalk by collecting traffic data of pre-processed video image data on the crosswalk; accurately calculate the traffic density according to the crowd density value and vehicle density value; analyze the safety risk according to the behavior data of the pre-processed video image data, realize the real-time collection of processed video images by the real-time traffic light management system, and identify the a of the crosswalk. i The invention realizes real-time adjustment of traffic lights according to the real-time calculation of the influence coefficient, effectively alleviates traffic congestion and improves traffic efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] 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.

[0046] Figure 1 It is a schematic diagram of the system framework of the present invention;

[0047] Figure 2 The figure is a flow chart of the system method of the present invention. DETAILED DESCRIPTION

[0048] The technical scheme of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0049] See also Figure 1 The first aspect of the present invention provides a real-time traffic light management system based on big data, comprising the following modules:

[0050] Data collection module: collects video image data on the crosswalk, including traffic data and behavior data;

[0051] Data processing module: pre-process the collected video image data, including: denoising and image enhancement; pre-process the traffic data collected from the video image data, including: removing outliers, filling missing values, data type conversion, unified standards and normalization;

[0052] Data analysis module: analyze the traffic data in the pre-processed video image data; calculate the crowd density value and vehicle density value when passing through the crosswalk based on the analysis results; comprehensively analyze and calculate the traffic density of the crosswalk; analyze the safety risk when passing through the crosswalk based on the behavior data in the pre-processed video image data; calculate the potential value of the safety risk based on the analysis results; comprehensively analyze and calculate the safety risk index of passing through the crosswalk;

[0053] Early warning management module: Based on comprehensive analysis of traffic density and safety risk index, calculate the impact coefficient and generate early warning information to adjust and control traffic lights in real time.

[0054] Specifically, the present invention includes the following modules: data acquisition module, data processing module, data analysis module and early warning management module. Preprocessing the collected image data includes: denoising, enhancing the contrast, brightness, smoothness and edge information of the image data; preprocessing the traffic data of the collected video image data includes: removing abnormal values ​​caused by equipment failure and human error, filling missing value data sets due to equipment failure and data transmission interruption, data type conversion unified standards and normalization processing including time format unification, converting continuous numerical variables into a unified numerical type; further analyzing the traffic density through the crosswalk by calculating the crowd traffic density value and vehicle traffic density value when passing through the crosswalk. Analyze the safety risk through the behavioral data of the preprocessed video image data. Calculate the safety risk index based on multiple safety risks. Finally, the influence coefficient is calculated by combining the traffic density and the safety risk index, and the early warning information is generated, and the traffic lights are adjusted and controlled in real time accordingly.

[0055] In one embodiment of the present invention, analyzing the traffic data in the pre-processed video image data includes the following steps:

[0056] According to the pre-processed video images, the traffic data in the video images are analyzed by a deep learning model;

[0057] Traffic data for marking crosswalks include: the position of pedestrians and non-motor vehicles at the crosswalks, the speed of pedestrians and non-motor vehicles, the longitudinal spacing and occupied width of pedestrians and non-motor vehicles when passing;

[0058] Input the labeled crosswalk traffic data into the deep learning model for model training;

[0059] The real-time traffic light management system collects and processes video images in real time, measures and outputs the positions of pedestrians and non-motor vehicles at crosswalks, the speeds of pedestrians and non-motor vehicles, and the longitudinal spacing and occupied width of pedestrians and non-motor vehicles when passing.

[0060] Specifically, the labeled traffic data is input into the deep learning model for training using a deep learning model; the positions of pedestrians and non-motor vehicles at the crosswalk are marked using bounding boxes, the speed of pedestrians and non-motor vehicles is calculated through the displacement of consecutive frames, and the longitudinal spacing and occupied width of pedestrians and non-motor vehicles when passing are directly obtained from the bounding box. The real-time collected and pre-processed video images are input into the trained deep learning model, and the model can measure the positions of pedestrians and non-motor vehicles at the crosswalk, the speed of pedestrians and non-motor vehicles, the longitudinal spacing and occupied width of pedestrians and non-motor vehicles when passing, and output the results.

[0061] In one embodiment of the present invention, calculating the crowd density value and vehicle density value when passing through a crosswalk according to the analysis results includes the following steps:

[0062] By analyzing the formula:

[0063] Get the crowd density value P passing through the crosswalk; where V P represents the walking speed of pedestrians, S P Indicates the longitudinal distance between pedestrians when walking, b P Indicates the lateral width occupied by pedestrians, T 绿 Indicates the duration of the green light, K P represents the correction factor;

[0064] By analyzing the formula:

[0065] Get the vehicle traffic density value C passing through the crosswalk; where V C represents the speed of non-motor vehicles, S C Indicates the longitudinal spacing of non-motor vehicles when passing, b C Indicates the lateral width occupied by non-motor vehicles, T 绿 Indicates the duration of the green light, K C Indicates the correction factor.

[0066] Specifically, the crowd density value P is obtained by multiplying the ratio of pedestrian walking speed to the longitudinal distance between pedestrians when walking, the ratio of green light duration to the horizontal width occupied by pedestrians, and finally multiplying it by the correction coefficient, where the correction coefficient K is P The value is 0.5; the vehicle density value C is obtained by multiplying the ratio of the non-motor vehicle speed to the longitudinal spacing when the non-motor vehicle passes by the ratio of the green light duration to the transverse width occupied by the non-motor vehicle and finally multiplying it by the correction coefficient, where the correction coefficient K C The value is 0.6; when the width of the crosswalk is narrower, b P and b C The smaller the value.

[0067] In one embodiment of the present invention, the comprehensive analysis and calculation of the traffic density through the crosswalk includes the following steps:

[0068] By analyzing the formula: F=P+C

[0069] The traffic density F passing through the crosswalk is obtained, where P and C represent the traffic density value of people and the traffic density value of vehicles respectively.

[0070] Specifically, the traffic density F passing through the crosswalk is equal to the sum of the crowd traffic density value P and the vehicle traffic density value C. The traffic density obtained by calculating the crowd traffic density value and the vehicle traffic density value can be used to obtain the traffic capacity on the crosswalk and possible congestion factors.

[0071] See also Figure 2 In one embodiment of the present invention, analyzing the safety risk of crossing a crosswalk according to the behavior data in the preprocessed video image data includes the following steps:

[0072] Based on the behavioral data in the preprocessed video image data, the safety risks of the pedestrian crossings in the video images are analyzed through the deep learning model;

[0073] Mark the crosswalk i security risks; where i∈{1,2,...n};

[0074] Mark the first i The security risks are input into the convolutional neural network of the deep learning model for model training;

[0075] The real-time traffic light management system collects and processes video images in real time to identify the first pedestrian crossing. i A security risk.

[0076] Specifically, the convolutional neural network of the deep learning model is used to label the first i The safety risks are input into the deep learning model for training; the real-time collected and pre-processed video images are input into the trained deep learning model, and the model can identify the a i security risk, and output the ath in the identified video image i A security risk.

[0077] In one embodiment of the present invention, the ath i The security risks include:

[0078] a1: running a red light; a2: driving on a non-pedestrian crossing; a3: pedestrians selling items in the crosswalk; a4: children and pets driving alone; a5: riding a non-motor vehicle without a helmet; a6: pedestrians walking while looking at their phones; a7: pedestrians chasing others in the crosswalk; a8: vehicles not giving way to pedestrians; a9: vehicles occupying the crosswalk; a10: pedestrians driving on the sidewalk without a helmet; a11: pedestrians driving on the sidewalk without a helmet; a12: pedestrians driving on the sidewalk without a helmet; a13: pedestrians driving on the sidewalk without a helmet; a14: pedestrians driving on the sidewalk without a helmet; a15: pedestrians driving on the sidewalk without a helmet; a16: pedestrians driving on the sidewalk while looking at their phones; a17: pedestrians driving on the sidewalk without a helmet; a18: pedestrians driving on the sidewalk without a helmet; a19: pedestrians driving on the sidewalk without a helmet; a 10 Vehicle speeding.

[0079] Specifically, a1 pedestrians and non-motor vehicles run red lights when crossing the crosswalk; a2 pedestrians and non-motor vehicles do not cross the zebra crossing of the crosswalk; a3 pedestrians sell items in the crosswalk; a4 children and pets cross the crosswalk alone; a5 non-motor vehicle riders cross the crosswalk without wearing helmets; a6 pedestrians walk on the crosswalk while looking at their mobile phones; a7 pedestrians chase others and run around in the crosswalk; a8 vehicles do not slow down and stop to give way to pedestrians; a9 vehicles occupy the crosswalk; a10 10 The vehicle was speeding through the pedestrian crossing.

[0080] In one embodiment of the present invention, calculating the potential value of the security risk according to the analysis result includes the following steps:

[0081] By analyzing the formula:

[0082] Get the first i The potential value of security risk α i ; Where A represents the ath i The total number of times a safety risk occurs; T represents the total number of days in a year, and F represents the traffic density.

[0083] Specifically, i The security risk events are independent of each other, and the average number of security risk events per day in a year is calculated. i The product of the probability of the occurrence of the safety risk and the traffic density is the a i The potential value of security risk α i . Identify the first i The security risks are statistically analyzed to obtain the a i The total number of occurrences of the security risk is A.

[0084] In one embodiment of the present invention, the comprehensive analysis and calculation of the safety risk index of passing through the crosswalk includes the following steps:

[0085] By analyzing the formula:

[0086] Get the security risk index ω; where α i Indicates the firsti The potential value of security risk, W i Indicates the first i The weight of the security risk.

[0087] Specifically, according to a i The weight of the security risk is multiplied by the a i The exponential sum of the potential values ​​of the security risks is divided by the square root of the sum of the squares of the weights to obtain the security risk index ω; i The weight of each safety risk is determined according to the i The severity of the security risk is assigned; the square root of the sum of the squares of the weights in the denominator will adjust the size of the total accordingly, ensuring that even if a certain weight is large, it will not dominate the total. This avoids excessive influence of extreme values ​​on the results and accurately calculates the security risk index ω.

[0088] In one embodiment of the present invention, the influence coefficient is calculated based on the comprehensive analysis of traffic density and safety risk index and warning information is generated to adjust and control the traffic lights in real time, including the following steps:

[0089] By analyzing the formula:

[0090] The influence coefficient δ is obtained; where F represents the traffic density and ω represents the safety risk index;

[0091] According to the value of the impact coefficient δ, different levels of thresholds are set, and the warning is divided into three levels: high-level warning, medium-level warning and low-level warning:

[0092] When δ≥60%, the warning is classified as a high-level warning, a red light is set, a serious warning signal is issued, and staff are summoned to deal with the safety risk;

[0093] When 30%<δ<60%, the warning is classified as a medium warning, and the green light time on the crosswalk is increased;

[0094] When δ≤30%, the warning is classified as a low-level warning, maintaining the normal traffic light cycle time, and continuing to detect traffic density and safety risk index in real time.

[0095] Specifically, the product of traffic density and safety risk index is normalized by dividing it by the square root of their sum of squares, and the final result is multiplied by a percentage so that the final result is calculated in the form of a percentage to obtain the influence coefficient δ. Different thresholds are set according to the value of the influence coefficient δ to divide it into three levels: high-level warning, medium-level warning and low-level warning. According to the image data collected and processed in real time, the influence coefficient δ is finally obtained through various analysis and calculation, and the warning information is obtained by comparing the range of the influence coefficient. When δ≥60%, the warning is divided into a high-level warning, the traffic signal on the crosswalk is set to a red light to prohibit passage, a serious warning signal is issued, and staff are summoned to deal with safety risks; when 30%<δ<60%, the warning is divided into a medium-level warning, the green light time on the crosswalk is increased, and the crosswalk is guaranteed to be unobstructed and traffic jams are reduced; when δ≤30%, the warning is divided into a low-level warning, the normal traffic light cycle time is maintained, and the traffic density and safety risk index are continuously detected in real time. According to the real-time adjustment of traffic lights, traffic congestion is alleviated.

[0096] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A real-time traffic light management system based on big data, characterized in that: Includes the following modules: Data collection module: collects video image data on the crosswalk, including traffic data and behavior data; Data processing module: pre-process the collected video image data, including: denoising and image enhancement; pre-process the traffic data collected from the video image data, including: removing outliers, filling missing values, data type conversion, unified standards and normalization; Data analysis module: analyze the traffic data in the pre-processed video image data; calculate the crowd density value and vehicle density value when passing through the crosswalk based on the analysis results; comprehensively analyze and calculate the traffic density of the crosswalk; analyze the safety risk when passing through the crosswalk based on the behavior data in the pre-processed video image data; calculate the potential value of the safety risk based on the analysis results; comprehensively analyze and calculate the safety risk index of passing through the crosswalk; Early warning management module: Based on comprehensive analysis of traffic density and safety risk index, calculate the impact coefficient and generate early warning information to adjust and control traffic lights in real time.

2. The real-time traffic light management system based on big data according to claim 1 is characterized in that: Analyzing the traffic data in the preprocessed video image data includes the following steps: According to the pre-processed video images, the traffic data in the video images are analyzed by a deep learning model; Traffic data for marking crosswalks include: the position of pedestrians and non-motor vehicles at the crosswalks, the speed of pedestrians and non-motor vehicles, the longitudinal spacing and occupied width of pedestrians and non-motor vehicles when passing; Input the labeled crosswalk traffic data into the deep learning model for model training; The real-time traffic light management system collects and processes video images in real time, measures and outputs the positions of pedestrians and non-motor vehicles at crosswalks, the speeds of pedestrians and non-motor vehicles, and the longitudinal spacing and occupied width of pedestrians and non-motor vehicles when passing.

3. The real-time traffic light management system based on big data according to claim 2 is characterized in that: According to the analysis results, the density of people and vehicles passing through the crosswalk is calculated, including the following steps: By analyzing the formula: Get the crowd density value P passing through the crosswalk; where V P represents the walking speed of pedestrians, S P Indicates the longitudinal distance between pedestrians when walking, b P represents the lateral width occupied by pedestrians, T 绿 Indicates the duration of the green light, K P represents the correction factor; By analyzing the formula: Get the vehicle traffic density value C passing through the crosswalk; where V C represents the speed of non-motor vehicles, S C Indicates the longitudinal spacing of non-motor vehicles when passing, b C Indicates the lateral width occupied by non-motor vehicles, T 绿 Indicates the duration of the green light, K C Indicates the correction factor.

4. The real-time traffic light management system based on big data according to claim 3 is characterized in that: Comprehensive analysis and calculation of the traffic density through the crosswalk includes the following steps: By analyzing the formula: F = P + C The traffic density F passing through the crosswalk is obtained, where P and C represent the traffic density value of people and the traffic density value of vehicles respectively.

5. The real-time traffic light management system based on big data according to claim 1 is characterized in that: According to the behavior data in the preprocessed video image data, the safety risk of crossing the crosswalk is analyzed, including the following steps: Based on the behavioral data in the preprocessed video image data, the safety risks of the pedestrian crossings in the video images are analyzed through the deep learning model; Mark the crosswalk i security risks; where i∈{1, 2, ...n}; Mark the first i The security risks are input into the convolutional neural network of the deep learning model for model training; The real-time traffic light management system collects and processes video images in real time to identify the first pedestrian crossing. i A security risk.

6. The real-time traffic light management system based on big data according to claim 5 is characterized in that: Mark the crosswalk i The security risks include: a1: running a red light; a2: driving on a non-pedestrian crossing; a3: pedestrians selling items in the crosswalk; a4: children and pets driving alone; a5: riding a non-motor vehicle without a helmet; a6: pedestrians walking while looking at their phones; a7: pedestrians chasing others in the crosswalk; a8: vehicles not giving way to pedestrians; a9: vehicles occupying the crosswalk; a10: pedestrians driving on the sidewalk without a helmet; a11: pedestrians driving on the sidewalk without a helmet; a12: pedestrians driving on the sidewalk without a helmet; a13: pedestrians driving on the sidewalk without a helmet; a14: pedestrians driving on the sidewalk without a helmet; a15: pedestrians driving on the sidewalk without a helmet; a16: pedestrians driving on the sidewalk while looking at their phones; a17: pedestrians driving on the sidewalk without a helmet; a18: pedestrians driving on the sidewalk without a helmet; a19: pedestrians driving on the sidewalk without a helmet; a10: pedestrians driving on the sidewalk without a helmet; a11: pedestrians driving on the sidewalk without a helmet; a12: pedestrians driving on the sidewalk without a helmet; a13: pedestrians driving on the sidewalk without a helmet; a14: pedestrians driving on the sidewalk without a helmet; a15: pedestrians driving on the sidewalk without a helmet; a16: pedestrians driving on the sidewalk without a helmet; a17: pedestrians driving on the sidewalk without a helmet; a18: pedestrians driving on the sidewalk without a helmet; a19: pedestrians driving on the sidewalk without a helmet 10 Vehicle speeding.

7. The real-time traffic light management system based on big data according to claim 6 is characterized in that: Calculating the potential value of security risk based on the analysis results includes the following steps: By analyzing the formula: Get the first i The potential value of security risk α i ; Where A represents the ath i The total number of times a safety risk occurs; T represents the total number of days in a year, and F represents the traffic density.

8. The real-time traffic light management system based on big data according to claim 7 is characterized in that: Comprehensive analysis and calculation of the safety risk index of pedestrian crossings includes the following steps: By analyzing the formula: Get the security risk index ω; where α i Indicates the first i The potential value of security risk, W i Indicates the first i The weight of the security risk.

9. The real-time traffic light management system based on big data according to claim 1, characterized in that: The impact coefficient is calculated based on comprehensive analysis of traffic density and safety risk index, and warning information is generated to adjust and control traffic lights in real time, including the following steps: By analyzing the formula: The influence coefficient δ is obtained; where F represents the traffic density and ω represents the safety risk index; According to the value of the impact coefficient δ, different levels of thresholds are set, and the warning is divided into three levels: high-level warning, medium-level warning and low-level warning: When 6 ≥ 60%, the warning is classified as a high-level warning, a red light is set, a serious warning signal is issued, and staff are summoned to deal with safety risks; When 30%<δ<60%, the warning is classified as medium warning, and the green light time on the crosswalk is increased: When 6≤30%, the warning is classified as a low-level warning, maintaining the normal traffic light cycle time, and continuing to detect traffic density and safety risk index in real time.

Images