A road traffic management system based on smart street lights

Abstract

The present invention relates to the technical field of intelligent transportation management, and specifically provides a road traffic management system based on smart street lights. The system includes a data perception module, a traffic flow assessment module, a signal adjustment module, a sudden event recognition module, and a response optimization module. In the present invention, through the fusion of geomagnetic and camera data, high-frequency and accurate perception of vehicle dynamic behaviors is achieved, improving the timeliness and integrity of data. By jointly analyzing traffic density and time periods, the saturation determination becomes more detailed. By comparing the queue length with the number of passing vehicles, traffic overload can be identified in advance. By jointly judging the signal skewness through the signal cycle ratio, the regional signal timing coordination is enhanced. By jointly recognizing dense vehicle shadows and sudden speed changes, the accuracy of anomaly detection is improved. By comparing the response delay and dynamically adjusting the priority, the signal precise regulation of key areas is strengthened. Overall, a leap from static control to active closed-loop regulation is achieved, improving traffic operation efficiency and adaptability.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent traffic management, and in particular to a road traffic management system based on smart street lamps. Background Art

[0002] The field of intelligent traffic management technology includes traffic flow management, road condition monitoring, traffic signal control, road monitoring, public transportation management and other related technologies. It aims to improve the operating efficiency and safety of the transportation system through intelligent means. The core content of this technology mainly involves traffic data collection and analysis, vehicle behavior prediction, real-time monitoring and dynamic scheduling. With the continuous development of information technology and Internet of Things technology, intelligent traffic management has gradually realized real-time perception and feedback of traffic conditions, making traffic management more efficient and accurate.

[0003] Among them, the road traffic management system refers to the use of smart street lights as a traffic management platform, which monitors road traffic conditions in real time through the integration of sensors, cameras, wireless communication technologies, etc., collects traffic data and analyzes and processes it, and provides solutions for optimizing traffic signal control and traffic flow management. It solves problems such as information islands and response lags in road traffic management, collects information such as vehicle flow and speed through the deployment of smart street lights, and realizes intelligent adjustment of traffic lights in combination with intelligent decision-making mechanisms.

[0004] Existing technologies primarily rely on basic data collection and static rule enforcement. Their traffic data collection methods are insufficiently reflective of temporal continuity and dynamic behavior, making them unable to accurately track instantaneous vehicle state changes, leading to biased traffic state identification. Regarding signal control, existing systems generally employ fixed-cycle or empirically defined methods, lacking sensitivity to real-time traffic loads. This results in signal timing lags during peak hours, such as morning and evening rush hours, leading to extended queues and intersection congestion. Signal timing between adjacent intersections lacks coordination, resulting in local adjustments leading to regional conduction blockages. For emergency state identification, traditional camera recognition relies on a single dimension, unable to accurately locate dense, abnormal areas and the synchronization of sudden speed changes, resulting in delayed responses to accidents or unusual congestion. Existing systems lack reaction time analysis and strategic feedback after signal interventions, preventing the original control effects from feeding back into current decisions. This leads to repeated and inefficient control behaviors, directly impacting road traffic efficiency and overall traffic safety. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a road traffic management system based on smart street lamps.

[0006] In order to achieve the above objectives, the present invention adopts the following technical solutions: A road traffic management system based on smart street lights includes:

[0007] The data perception module measures data based on the geomagnetic sensors of smart streetlights and traffic camera components, including the number of lane occupations, passing vehicle speeds, and the number of vehicles passing through within a signal cycle. Combining the number of occupations per unit time and the passing time section, it analyzes the average passing saturation of the intersection to obtain the passing saturation value of the intersection.

[0008] Based on the passing saturation value of the intersection, the traffic flow evaluation module identifies the queue length and the total number of vehicles passing through during the green light in the signal control cycle. By comparing the ratio of the two, it determines whether the current traffic flow reaches the adjustment threshold to obtain the signal load prediction index.

[0009] According to the signal load prediction index, the signal adjustment module determines whether it deviates from the green light priority interval by comparing the cycle ratio of the number of vehicles passing through the green light and the red light waiting time, and conducts a cycle deviation determination for adjacent intersections to obtain the signal control synchronization adjustment instruction.

[0010] The sudden event identification module calls the signal control synchronization adjustment instruction. By analyzing the vehicle density distribution map and speed mutation points obtained from the front-end image acquisition component, it identifies the concentrated appearance of the dense vehicle shadow area and the speed sudden drop points, screens the abnormal sections and associates the adjustment instruction range to obtain the abnormal concentrated response block identifier.

[0011] As a further solution of the present invention, the passing saturation value of the intersection includes passing density distribution, vehicle speed change range, and signal cycle passing rate. The signal load prediction index includes queue growth ratio, green light passing saturation, and cycle utilization efficiency. The signal control synchronization adjustment instruction includes adjacent intersection timing deviation, green light priority interval determination value, and signal cycle correction factor. The abnormal concentrated response block identifier includes abnormal density section number, speed mutation identification label, and image aggregation area identifier.

[0012] As a further solution of the present invention, the data perception module includes:

[0013] The occupancy detection sub-module measures data based on the geomagnetic sensors of smart streetlights and traffic camera components, extracts vehicle induction data and lane image frames of the camera component, analyzes the matching relationship between vehicle trajectories and induction point signals, counts the number of vehicle occupations on each lane, and generates a signal cycle lane occupancy sequence table.

[0014] Based on the signal cycle lane occupancy sequence table, the passing efficiency sub-module extracts the start and end coordinates and timestamps of the trajectories in the traffic camera component, identifies the passing distance and time interval of continuous trajectories, summarizes the average passing speed of the lane and the number of vehicles passing through within the signal cycle, and generates an intersection passing speed and traffic flow index set.

[0015] The saturation ratio evaluation sub-module, based on the intersection passing speed and the traffic flow index set, identifies the passing time and the passing quantity data of each lane, and numerically calculates the lane saturation state according to the ratio of the lane passing demand to the available time period per unit time, so as to obtain the intersection passing saturation value.

[0016] As a further solution of the present invention, the traffic flow evaluation module includes:

[0017] The passing saturation value calculation sub-module, based on the intersection passing saturation value, extracts the number of vehicles passing through per unit time and the upper limit of the lane passing capacity, and identifies the ratio of the traffic flow passing volume to the theoretical passing volume per cycle, so as to obtain the passing saturation value;

[0018] The queue proportion determination sub-module calls the passing saturation value, counts the number of queuing vehicles and the number of vehicles passing through in real time during the green light period of each cycle, analyzes the ratio difference between the two, and compares it with a fixed passing reference ratio to obtain the queue occupancy ratio;

[0019] The signal load identification sub-module, according to the queue occupancy ratio, combines the original queue trend and the total number of vehicles passing through in the current signal cycle, and uses the formula:

[0020] ;

[0021] Identify the signal cycle load degree, and compare it with the set signal load adjustment threshold to obtain the signal load prediction index;

[0022] Wherein, represents the signal load prediction index, represents the current number of queuing vehicles, represents the total number of vehicles passing through the green light, represents the passing saturation value, represents the number of queuing vehicles in the cycle, represents the duration of the green light in the cycle, represents the number of vehicles passing through during the green light in the cycle, represents the queue occupancy ratio, represents the average value of the original queue trend value, represents the selected number of cycles.

[0023] As a further solution of the present invention, the signal adjustment module includes:

[0024] The priority interval judgment sub-module determines whether the signal cycle deviates from the priority control strategy and obtains the traffic control deviation status by judging according to the signal load prediction index, combining the number of vehicles passing through the green light and the cumulative waiting time of the red light collected by the intelligent street lamp platform, and comparing with the set green light priority threshold;

[0025] The signal timing correction sub-module extracts the green light passing rate and green light duration in the traffic direction from the intelligent street lamp regulation library according to the traffic control deviation status, combines the average headway and the density of queuing vehicles, aggregates the traffic efficiency and traffic intensity differences in each direction, and uses the formula:

[0026] ;

[0027] Calculate the signal timing deviation level, and adjust the order of the signal timing scheme according to the level range to construct a light control rhythm rearrangement scheme;

[0028] Among them, represents the signal timing deviation level, is the offset of the green light passing rate, is the number of vehicles passing through the green light, is the number of vehicles waiting for the red light, is the direction queue length, is the direction green light time, is the direction traffic volume, is the direction queue density, is the total number of traffic directions;

[0029] The light control synchronization instruction acquisition sub-module calls the light control rhythm rearrangement scheme, collects the signal start time and synchronization deviation reference value of the current intersection and adjacent intersections in the intelligent street lamp control gateway, and judges whether the rearrangement scheme causes cycle misalignment and overlap to obtain the signal control synchronization adjustment instruction.

[0030] As a further solution of the present invention, the sudden change recognition module includes:

[0031] The vehicle shadow density recognition sub-module calls the signal control synchronization adjustment instruction, extracts the vehicle density distribution map, analyzes the pixel aggregation, boundary overlap and lane traffic space parameters of the image frame, and compares the spacing fluctuation between vehicle shadows in the area to obtain the vehicle shadow density coefficient value;

[0032] The speed mutation judgment sub-module identifies the original vehicle speed information according to the vehicle shadow density coefficient value, extracts the speed jump amplitude and frequency, and analyzes the deviation from the median value of the speed fluctuation interval to obtain the speed jump comparison value;

[0033] The abnormal section screening sub-module extracts the block number, response time series, and boundary coverage ratio involved in the signal control adjustment instruction according to the speed jump comparison value, and uses the formula:

[0034] ;

[0035] Identify the abnormal intensity, map it to the signal control number, and obtain the abnormal concentrated response block identifier;

[0036] Among them, represents the abnormal concentrated response block identifier, is the vehicle shadow overlap degree of the sub-block in the area , is the minimum vehicle speed of the sub-block in the area , is the response time series equilibrium value of the area , is the boundary coverage ratio of the area , represents the total number of areas.

[0037] As a further solution of the present invention, the system further includes a response optimization module:

[0038] Based on the abnormal concentrated response block identifier, the response optimization module extracts the time difference between the signal intervention delay and the evacuation result by calling the original signal adjustment record, compares the current block number with the original response duration, adjusts the trigger priority sequence, and obtains the signal regulation instruction for the key traffic area;

[0039] The signal regulation instruction for the key traffic area includes the signal response priority, the original intervention delay data, and the regulation block number matching value.

[0040] As a further solution of the present invention, the response optimization module includes:

[0041] Based on the abnormal concentrated response block identifier, the block identifier extraction sub-module extracts the signal response time and the abnormal event number from the intelligent street lamp log, identifies the corresponding response time period of the number, and maps and combines it with the road section number to generate an intelligent street lamp response block mark set;

[0042] The signal response delay calculation sub-module calls the intelligent street lamp response block mark set, extracts the signal intervention start time and the evacuation end time of the corresponding block, analyzes the time difference between the two and matches it to the block number, identifies the corresponding data of the block number and the response delay, and obtains the road section response delay list;

[0043] The signal regulation instruction generation sub-module rearranges the signal priority sequence according to the delay degree by comparing with the original response duration based on the road section response delay list, and allocates the adjusted traffic light cycle parameter values in combination with the road section numbers to obtain the traffic key area signal regulation instructions.

[0044] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0045] In the present invention, by jointly collecting the lane occupancy times, passing vehicle speeds, and the number of vehicles passing through the signal cycle by the geomagnetic sensor and the traffic camera component, a high-frequency perception of vehicle dynamic behaviors is formed, improving the data accuracy and timeliness, and effectively compensating for the limitation that static detection means cannot continuously track. Based on the comprehensive analysis means of traffic density and traffic time period, the determination of the average traffic saturation degree is made more fine-grained, providing a stable basis for subsequent traffic state judgment. By comparing the queue length with the number of vehicles passing through the green light, the traffic flow carrying pressure is accurately characterized, and the forward response under the overload situation is realized. Linking the determination of the traffic deviation interval with the signal cycle ratio realizes the agile identification of signal skewness, and introduces the cycle coordination judgment between adjacent intersections to improve the regional traffic time efficiency adaptation ability. Jointly extracting the dense vehicle shadow area and speed mutation points in the front-end image improves the accuracy and response granularity of abnormal area identification. By comparing the original signal intervention delay with the current response duration, the trigger priority is dynamically adjusted, optimizing the immediate response of the control signal to local sudden states, avoiding the aggravation of congestion caused by signal control lag, and strengthening the accurate coverage ability of signal regulation for key areas. The overall solution breaks through the limitation of static rule regulation, forms a full-chain closed-loop control logic in aspects such as acquisition dimension, response granularity, regional linkage, and sudden adaptability, enabling the traffic control to leap from passive response to active adjustment. Description of the Drawings

[0046] Figure 1 is the system flow chart of the present invention;

[0047] Figure 2 is the flow chart of the data perception module in the present invention;

[0048] Figure 3 is the flow chart of the traffic flow evaluation module in the present invention;

[0049] Figure 4 is the flow chart of the signal adjustment module in the present invention;

[0050] Figure 5 is the flow chart of the sudden identification module in the present invention;

[0051] Figure 6 is the flow chart of the response optimization module in the present invention. Detailed Embodiments

[0052] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0053] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention. In addition, in the description of the present invention, the meaning of "a plurality" is two or more unless otherwise specifically defined.

[0054] Please refer to Figure 1 , a road traffic management system based on smart street lights includes:

[0055] The data perception module measures data based on the geomagnetic sensors and traffic camera components of the smart street lights, including the number of lane occupations, the passing vehicle speed, and the number of vehicles passing through within the signal cycle. Combining the number of occupations per unit time and the passing time period, it analyzes the average traffic saturation of the intersection to obtain the traffic saturation value of the intersection;

[0056] The traffic flow assessment module identifies the queue length and the total number of vehicles passing through the green light during the signal control cycle based on the traffic saturation value of the intersection. By comparing the ratio of the two, it determines whether the current traffic flow reaches the adjustment threshold to obtain the signal load prediction index;

[0057] The signal adjustment module determines whether it deviates from the green light priority interval by comparing the cycle ratio of the number of vehicles passing through the green light and the red light waiting time according to the signal load prediction index. If it deviates, it adjusts the signal timing sequence and determines the cycle deviation of adjacent intersections to obtain the signal control synchronization adjustment instruction;

[0058] The sudden event identification module calls the signal control synchronization adjustment instruction. By analyzing the vehicle density distribution map and speed mutation points obtained from the front-end image acquisition component, it identifies the concentrated appearance of the dense vehicle shadow area and the speed sudden drop points, screens the abnormal sections and associates the adjustment instruction range to obtain the abnormal concentration response block identifier;

[0059] The response optimization module adjusts the trigger priority sequence based on the abnormal concentration response block identifier by calling the original signal adjustment record, extracting the time difference between the signal intervention delay and the evacuation result, and comparing the current block number with the original response duration to obtain the signal regulation instruction for key traffic areas.

[0060] The intersection traffic saturation values include traffic density distribution, vehicle speed change range, and signal cycle passing rate. The signal load prediction indicators include queue growth ratio, green light passing saturation, and cycle utilization efficiency. The signal control synchronization adjustment instructions include the timing deviation between adjacent intersections, the determination value of the green light priority interval, and the signal cycle correction factor. The abnormal concentrated response block identification includes the abnormal density section number, speed mutation identification label, and image aggregation area identification. The traffic key area signal regulation instructions include signal response priority, original intervention delay data, and the matching value of the regulation block number.

[0061] Please refer to Figure 2 , the data perception module includes:

[0062] The occupancy detection sub-module measures data based on the intelligent street lamp geomagnetic sensor and the traffic camera component, extracts vehicle induction data and the lane image frames of the camera component, analyzes the matching relationship between the vehicle trajectory and the induction point signal, counts the vehicle occupancy times of each lane, and generates a signal cycle lane occupancy sequence table;

[0063] First, extract the vehicle induction data and obtain the lane image frames of the traffic camera component. The geomagnetic sensor converts the signal of the vehicle passing by into induction data, which includes the timestamp of the vehicle passing by and the position of the induction point. The traffic camera component captures the image frames of each lane, and the image includes the moving trajectory of the vehicle. After extracting the data, through the matching analysis of the vehicle trajectory and the induction point signal, identify the moment when the vehicle passes through a specific position and its travel path in the lane. The matching process is realized through the relative relationship between the timestamp and the image frame, involving setting a time window to ensure that the vehicle passes through a certain induction point within this time. If the vehicle passes through a certain induction point within this window, it is determined that the vehicle passes through this induction point. According to the data of multiple induction points, count the vehicle occupancy times of each lane for subsequent traffic flow and efficiency analysis. For example, if the induction point signal of lane 1 shows that the vehicle occupancy time is 10 seconds and the vehicle passes through this area within 10 seconds, record the occupancy situation of this lane as one occupancy, and generate a signal cycle lane occupancy sequence table to show the occupancy situation of each lane at different time periods.

[0064] The traffic efficiency sub-module extracts the start and end coordinates and timestamps of the trajectories in the traffic camera component based on the signal cycle lane occupancy sequence table, identifies the passing distance and time interval of the continuous trajectories, summarizes the average passing speed of the lane and the number of vehicles passing through within the signal cycle, and generates an intersection passing speed and traffic flow index set;

[0065] Extract the start and end coordinates and timestamps of vehicle trajectories in the traffic camera component, extract the trajectories of each vehicle from the image frames, determine the coordinates of their starting and ending points, and combine the timestamps of each point to calculate the passing time and the distance traveled by the vehicle. The distance traveled by the vehicle is achieved by calculating the physical distance between the start and end coordinates, using a geographic coordinate conversion algorithm or directly converting the pixel coordinates in the image to the actual distance. The time interval is obtained by calculating the difference between the start and end timestamps. By analyzing the data, identify the passing distance and time interval of continuous trajectories. Calculate the average passing speed of the vehicle within the signal cycle, which can be calculated by the formula: speed = distance / time, to obtain the passing speed of each vehicle. Then, based on the statistical results, summarize the lane average passing speed and the number of vehicles passing through within the signal cycle for each lane, so as to obtain the passing capacity of each lane. For example, if a vehicle passes through a 200-meter lane in 20 seconds, the passing speed of the vehicle is 10 meters per second. Summarize the data to generate the intersection passing speed and traffic flow index set.

[0066] Based on the intersection passing speed and traffic flow index set, the saturation ratio evaluation sub-module identifies the passing time and passing quantity data of each lane, and performs a numerical operation on the lane saturation state according to the ratio of the lane passing demand per unit time to the available time section, to obtain the intersection passing saturation value;

[0067] Identify the passing time and passing quantity data of each lane, and calculate the saturation of the lane. By analyzing the passing speed and traffic flow, obtain the number of passing vehicles and the time required for passing in each lane within one signal cycle. The passing time of the lane can be obtained by summing up the passing time of each vehicle, and the passing quantity is the total number of passing vehicles. According to the data, calculate the passing demand per unit time of each lane, that is, the number of vehicles expected to pass through each lane per unit time. By comparing the passing demand per unit time with the available time section of the lane, calculate the saturation ratio of the lane. For example, if the passing demand of the lane is 10 vehicles per minute, and the signal cycle of the lane is 60 seconds (i.e., 1 minute), then its saturation ratio is 10 / 10 = 1, indicating that the lane reaches full load. In this way, obtain the intersection passing saturation value, so as to evaluate the passing pressure of the intersection.

[0068] Please refer to Figure 3 , the traffic flow evaluation module includes:

[0069] Based on the intersection passing saturation value, the passing saturation value calculation sub-module extracts the number of passing vehicles per unit time and the upper limit of the lane passing capacity, and identifies the ratio of the traffic flow passing volume to the theoretical passing volume per cycle, to obtain the passing saturation value;

[0070] In intelligent transportation, for example, in the traffic management of an intersection, by real-time monitoring of the traffic flow and lane capacity within each signal cycle, calculating the traffic flow through-put and the theoretical passing capacity for each signal cycle, which is based on the data collected by road sensors and traffic monitoring cameras. For example, the number of vehicles passing through in one cycle is 1000, while the theoretical passing capacity of the road design is 1200 vehicles per hour. Based on the data, the passing saturation value for each cycle is calculated. The calculation of the passing saturation value is by dividing the actual number of vehicles passing through each cycle by the theoretical maximum passing capacity. For example, 1000 / 1200 = 0.83, which indicates that the passing saturation value for this cycle is 83%. This value can intuitively reflect the passing efficiency of the traffic flow, and further provide data support for traffic signal adjustment to obtain the passing saturation value.

[0071] The queuing ratio determination sub-module calls the passing saturation value, counts the number of queuing vehicles and the number of vehicles passing through in real-time during the green light period of each cycle, analyzes the ratio difference between the two, and compares it with the fixed passing reference ratio to obtain the queuing occupancy ratio;

[0072] In intelligent traffic signal control, using the passing saturation value, further obtain the queuing length formed during the red light period and the number of vehicles actually passing through during the green light period in each signal cycle. For example, in a busy signal cycle, the number of queuing vehicles reaches 150, while the number of vehicles actually passing through during the green light period is only 100. By calculating the ratio of these two values, such as 150 / 100 = 1.5, that is, the queuing length is 1.5 times that of the passing vehicles. Compare it with the set passing ratio benchmark. If the set benchmark is 1.2 times, it means that the current queuing ratio exceeds the ideal state and the signal settings need to be adjusted. In this way, evaluate the efficiency of traffic signals and adjust them in real-time to reduce queuing, thereby improving the passing speed of the traffic flow to obtain the queuing occupancy ratio.

[0073] The signal load identification sub-module, based on the queuing occupancy ratio, combines the original queuing trend and the total number of vehicles passing through in the current signal cycle, using the formula:

[0074] ;

[0075] Identify the signal cycle load degree, and compare it with the set signal load adjustment threshold to obtain the signal load prediction index;

[0076] Among them, represents the signal load prediction index, represents the current number of queuing vehicles, represents the total number of vehicles passing through during the green light, represents the passing saturation value, represents the number of queuing vehicles in the cycle, represents the Duration of the green light in a cycle, represents the number of vehicles passing through during the green light in a cycle, represents the queuing occupancy ratio, represents the average value of the original queuing trend value, represents the number of selected cycles;

[0077] Combined with the queuing occupancy ratio and the traffic saturation value, further identify the signal cycle load situation, determine whether the signal flow reaches the adjustment threshold through numerical calculation, and obtain the load measurement level under the current signal control state. First, obtain the numerical values of each parameter in the formula. Among them, the current number of queuing vehicles is obtained from the measurement of the queuing length at 1 second before the end of the red light in a certain approach of the intersection during the current signal cycle. For example, if it is detected by a radar sensor that there are 75 vehicles queuing in this approach, then , the total number of vehicles passing through the green light can be detected by a loop inductive coil to count the number of vehicles leaving during the green light period, and the detection result is vehicles, and the traffic saturation value is the result calculated previously, set as , the original queuing trend value is the average value of the queuing vehicles in 3 consecutive signal cycles. For example, if the 3 cycles are 78, 72, and 81 respectively, then , the queuing occupancy ratio is the result obtained in the previous sub-module, set as , when calculating the summation term , it is assumed that original cycle data are taken for judgment, and the data for each cycle are respectively:

[0078] The 1st cycle: vehicles, the green light time seconds, the number of passing vehicles ;

[0079] The 2nd cycle: vehicles, seconds, ;

[0080] The 3rd cycle: vehicles, seconds, ;

[0081] Substitute them into the calculation respectively to get:

[0082] ;

[0083] ;

[0084] ;

[0085] Sum up the above values to get ;

[0086] Then, calculate the first term of the numerator part: ;

[0087] The final value of the numerator is: ;

[0088] The denominator part is: ;

[0089] Finally, substitute into the formula to get: ;

[0090] This result shows that the signal load prediction index is 4.53, which is significantly higher than the standing adjustment threshold value (set to 1.0), indicating that there is a serious load pressure in the current signal cycle. It is necessary to trigger the dynamic adjustment mechanism of the signal cycle. In terms of dimension, it is unified in the form of vehicle number / dimensionless coefficient, so that the index has the ability of standardization, which is convenient for subsequent horizontal comparison in multi-signal intersections;

[0091] Among them, represents the number of queuing vehicles (vehicles) at the end of the red light in the current cycle, is the total number of vehicles passing through during the green light (vehicles), is the traffic saturation value (dimensionless, value range 0-1), is the number of queuing vehicles (vehicles) in the th original cycle, is the duration of the green light (seconds) in the th cycle, is the actual number of passing vehicles (vehicles) during the green light in the th cycle, is the number of original cycles for value taking, is the current queuing occupancy ratio (dimensionless), is the average value of the original queuing vehicle number (vehicles), is the finally calculated signal load prediction index (dimensionless);

[0092] By introducing multiple dynamic elements such as the comparison of the current and original queuing quantities, the deviation of the green light passing efficiency, and the weighting of the original saturation value, a complex determination mechanism for the signal cycle load is realized, which has scalability and real-time performance in the system-level traffic flow regulation and effectively enhances the basis of traffic signal dynamic regulation.

[0093] Please refer to Figure 4 , the signal adjustment module includes:

[0094] The priority interval judgment sub-module judges whether the signal cycle deviates from the priority control strategy according to the signal load prediction index, combines the number of vehicles passing through the green light and the cumulative waiting time of the red light collected by the intelligent street lamp platform, and compares with the set green light priority threshold to obtain the traffic control deviation status;

[0095] In the management of intelligent street lamps, through real-time data analysis, the vehicle passing data during green and red lights is monitored. This data is obtained from the sensors of intelligent street lamps, including vehicle counters and time recorders. The data of the number of vehicles passing through during the green light and the waiting time of vehicles during the red light quantifies the traffic flow. Based on this data, the actual passing ratio is compared with the preset priority passing threshold. For example, within a traffic cycle, 100 vehicles are allowed to pass through the green light and the red light waiting time is 120 seconds. The passing ratio is calculated as 100 vehicles / 120 seconds and compared with the set upper limit of the green light priority interval of 1.0 vehicle / second to judge whether traffic control needs to be adjusted to adapt to traffic flow changes. In this way, it is confirmed whether the optimal traffic control strategy is deviated. If the current passing ratio is lower than the preset threshold, it means that the green light period is not fully utilized, and the green light duration needs to be increased or the traffic signal control strategy needs to be adjusted to optimize the traffic flow and achieve a more efficient intersection passing capacity. Finally, the traffic control deviation status is obtained, which indicates the deviation between the current signal setting and the optimal flow control.

[0096] The signal timing correction sub-module extracts the green light passing rate and green light duration in the intelligent street lamp regulation library according to the traffic control deviation status, combines the mean headway and queue vehicle density, aggregates the passing efficiency and passing intensity differences in each direction, and uses the formula:

[0097] ;

[0098] Calculate the signal timing deviation level, and adjust the order of the signal timing plan according to the level range to construct a traffic light control rhythm rearrangement plan;

[0099] Among them, represents the signal timing deviation level, is the deviation of the green light passing rate, is the number of vehicles passing through the green light, is the number of vehicles waiting for the red light, is the direction queue length, is the direction green light time, is the direction passing vehicle number, is the direction queue density, is the total number of passing directions;

[0100] In the intelligent street light traffic management, signal timing correction is based on the traffic control offset state. Its core goal is to generate adjustable timing structure adjustment suggestions through quantitative analysis of green light efficiency, queue density and directional traffic intensity, and obtain the number of green light vehicles. Number of vehicles waiting at red lights , which is recorded in real time by the embedded vehicle detector and lane coil, and obtained through periodic cumulative sampling, assuming that the number of vehicles passing during the green light period recorded by the counter in this cycle is 60, that is , the number of vehicles queuing before the red light ends is 90, that is , the equipment uniformly uses a monitoring segment with a cycle length of 90 seconds to capture parameters, and the green light pass rate offset The number of vehicles passing through this period is compared with the average number of vehicles passing through the past five periods. The original average is 75 vehicles. , reflecting the decline in efficiency of this cycle;

[0101] Then collect the traffic direction data and the queue length of each traffic direction , green light time , number of vehicles passing ,density The data is collected jointly by the video analysis module, traffic light timing controller, flow counter, radar probe or laser distance sensor in the smart street light system;

[0102] In direction 1, the camera detects the head position sequence of vehicles in the queue area and estimates that the total number of vehicles is 35. , the traffic light control records the green light time as 30 seconds, , the vehicle passing counter records 40 vehicles, The density is calculated by the ranging module in sections. The total queue length is 32 meters and the average density is Vehicles / meter, approximately 1.1;

[0103] The data collection process for direction 2 is the same as that for direction 3, and the following results are obtained: , , , the queue length is 35 meters, then ; , , , the queue length is 37 meters, then ;

[0104] Traffic efficiency offset term: ;

[0105] Offset items in each direction:

[0106] Direction 1: ;

[0107] Direction 2: ;

[0108] Direction 3: ;

[0109] The summation term is: ;

[0110] Total formula calculation: ;

[0111] Here, all parameters have been automatically collected and pre - processed by the embedded sensors or edge computing nodes in the smart streetlights, without relying on external data sources, forming a closed - loop monitoring link. The units are uniformly processed into dimensionless. Each participating term is normalized in the following way: Time, the number of vehicles, density, etc. form a unit - elimination ratio. The seconds, vehicles, and meters in are eliminated into proportional terms. Finally, all parameters are combined and calculated into the signal timing deviation level. The result value is 23.031, which belongs to the middle interval of the deviation level (the preset reference interval is 0 - 15 for stability, 15 - 30 for the adjustment transition zone, > 30 for the reconstruction trigger zone). Therefore, it should enter the rhythm structure optimization process, and thus obtain the lamp control rhythm rearrangement plan.

[0112] The lamp control synchronization instruction acquisition sub - module calls the lamp control rhythm rearrangement plan, collects the signal start time and the synchronization deviation reference value of the current intersection and adjacent intersections in the smart streetlight control gateway, judges whether the rearrangement plan causes cycle misalignment and overlap, and obtains the signal control synchronization adjustment instruction;

[0113] Controlling signal synchronization is the key to ensuring smooth road traffic. By real - time monitoring the signal timings of adjacent intersections and the signal timing of the current intersection. For example, if the signal timing of an adjacent intersection is advanced by 30 seconds due to a special event, and the current intersection is delayed by 20 seconds due to the lamp control rhythm rearrangement plan, this deviation causes the traffic signals at the two intersections to be out of sync, thus affecting the coherence and efficiency of the traffic flow. By comparing the adjusted time with the synchronization deviation reference value, it can be judged whether further adjustment is needed to maintain signal synchronization. If it is detected that the time deviation caused by the rearrangement plan exceeds the allowable error range, the smart streetlight control will automatically generate a calibration instruction to adjust the signal start time to ensure that the signal lights at all intersections operate synchronously, thus maintaining the smoothness of the traffic flow. Finally, a signal synchronization control instruction is generated, and this instruction ensures the coordination of signal timings among intersections and optimizes the traffic flow in the entire area.

[0114] Please refer to Figure 5 , the sudden - event recognition module includes:

[0115] The vehicle shadow density recognition sub-module calls the signal control synchronization adjustment instruction, extracts the vehicle density distribution map, analyzes the pixel aggregation, boundary overlap and lane passing space parameters of the image frame, compares the spacing fluctuations between vehicle shadows in the area, and obtains the vehicle shadow density coefficient value;

[0116] Through high-resolution cameras deployed by the roadside, real-time images of the road are collected. Through image processing technology, the distribution density of vehicles can be analyzed. For example, in one instance, the vehicle density area of a certain main road during the evening rush hour is monitored. Then, the pixel aggregation parameters of the image frame are extracted from the vehicle density map, and the vehicle density is judged by analyzing the color depth of each image pixel point. The area with a higher vehicle density is darker in color. The vehicle shadow boundary overlap parameter is calculated by detecting the overlap degree of vehicle contours. The data helps to judge the continuity of the traffic flow and congestion points. The lane passing space parameter is obtained by measuring the remaining visible space in the lane, which can reflect the usage efficiency of the lane. If the space is less than a certain threshold, it indicates an upcoming or existing traffic congestion. The minimum spacing fluctuation between vehicle shadows in the area is compared, which is achieved by calculating the position changes of the same vehicle or different vehicles in consecutive frames. If a continuously decreasing small spacing is detected, an early warning will be triggered to prevent potential traffic accidents. The vehicle shadow density coefficient value is obtained. This coefficient value is a comprehensive index calculated through the above parameters, reflecting the vehicle density in a certain area. Through real-time data display, such as during the evening rush hour, this index increases significantly, and the traffic management center will adjust the signal timing or issue traffic control information accordingly.

[0117] The speed mutation judgment sub-module identifies the original vehicle speed information according to the vehicle shadow density coefficient value, extracts the speed jump amplitude and frequency, analyzes the deviation from the median value of the speed fluctuation interval, and obtains the speed jump comparison value;

[0118] The original vehicle speed information is called. In intelligent transportation, based on the comparative analysis of the original data and real-time monitoring data, such as recording the vehicle speed data in the same period in the past, it can be used to predict and compare the current vehicle speed trend. The speed jump amplitude and minimum frequency are extracted, and the speed change of each vehicle is analyzed through an algorithm to identify vehicles with a speed reduction exceeding the set threshold. The threshold is set based on long-term data statistics and traffic flow theory. For example, if the vehicle speed drops suddenly by more than 20% within three consecutive image frames, it is considered that a speed mutation has occurred. Calculate its deviation from the median value of the speed fluctuation interval. The calculation is achieved by comparing the real-time vehicle speed with the median speed of the interval. If the deviation exceeds the preset safety range, an alarm will be generated or the working mode of the signal light will be adjusted to relieve the traffic congestion or accident risk. The speed jump comparison value is obtained. This comparison value is a numerical value that can intuitively reflect the stability of the vehicle speed during a certain period. If the value is large, it indicates that the speed changes violently and measures need to be taken.

[0119] The abnormal section screening sub-module extracts the block number, response time series, and boundary coverage ratio involved in the signal control adjustment instruction according to the speed jump comparison value, using the formula:

[0120] ;

[0121] Identify the abnormal intensity and map it to the signal control number to obtain the abnormal concentrated response block identifier;

[0122] Among them, represents the abnormal concentrated response block identifier, is the vehicle shadow overlap degree of the sub-block in the area , is the minimum vehicle speed of the sub-block in the area , is the response time series equilibrium value of the area , is the boundary coverage ratio of the area , represents the total number of areas;

[0123] Through the real-time records deployed in the signal control cabinets at each intersection in traffic control, within blocks with different numbers, record their signal light control cycles and their changes respectively. The block number is the logical index, and the response time series refers to the control cycle of the signal lights in each round of traffic state changes. For example, in the period from 8:00 to 9:00 in the morning in the block numbered B1, its response time series is [30s, 28s, 32s, 30s], reflecting the real-time fluctuations of the signal light intervals in this section. The boundary coverage ratio is identified by the proportion of the vehicle shadow boundary pixels detected by the image edge detection method, and the proportion of the edge pixels corresponding to the effective width of the lane in this area is extracted from the image frame. For example, the vehicle shadow boundary coverage ratio in the B1 section is 88.6%;

[0124] In terms of the vehicle shadow overlap degree , it is calculated by the proportion of the cumulative pixel area of the vehicle occlusion area in the image sequence, using the following method: For the a-th sub-block in the area b, count all the pixel points with vehicle shadows in the frame images within 5 seconds and multiply by the ratio of the total effective pixels of this sub-block. For example, if the cumulative vehicle-covered pixels in the 5-second image of a certain sub-block are 42000 and the total effective pixels are 60000, then ;

[0125] The minimum vehicle speed is obtained from the speed trajectory data generated by the license plate tracking module synchronized with the image frame, and select the minimum vehicle speed among all the vehicles recorded within the sampling period in the corresponding sub-block. For example, if the speed values in a certain sub-block are [24, 28, 20, 31] km / h, then km / h, the vehicle speed needs to be converted to , that is ;

[0126] Equilibrium value of response time series is the average signal cycle corresponding to the current block number. For example, taking the time series of this area [30s, 28s, 32s, 30s] and calculating the average value to get ;

[0127] Boundary coverage rate is the average boundary pixel coverage rate of the area directly calculated by the image processing module, standardized to a decimal value. For example, 88.6% is ;

[0128] Substitute the example data. Assume that the corresponding values of sub-blocks to are:

[0129] , ;

[0130] , ;

[0131] , ;

[0132] , ;

[0133] Calculate the numerator part of the above formula:

[0134] ;

[0135] ;

[0136] ;

[0137] ;

[0138] Numerator:

[0139] Calculate the denominator part:

[0140] ;

[0141] ;

[0142] ;

[0143] Final calculation: ;

[0144] The result shows that the response section numbered B1 exhibits an abnormally concentrated response block with an identification value of 191.24 during the current monitoring period, which is significantly higher than the empirical benchmark value of 120, indicating that this area needs to be included in the priority of synchronous signal adjustment;

[0145] represents the abnormal response intensity value of the representative area, is the vehicle shadow overlap degree (unitless, range 0 - 1) of the sub-block , is the minimum vehicle speed (unit: m / s) of the sub-block ; is the average signal response period (unit: s) of the area ; is the boundary pixel coverage rate (unitless);

[0146] The congestion degree is quantified by the product of the vehicle shadow overlap degree and the minimum vehicle speed, combined with the square amplification of the response time deviation, and adding a boundary interference term. Through the comprehensive compression of the abnormal interference data fluctuation by the square difference and square root structure, it effectively reveals the areas of sudden congestion or traffic anomalies, ensuring the accuracy and sensitivity of subsequent synchronous instruction adjustment.

[0147] Please refer to Figure 6 , the response optimization module includes:

[0148] The block identification extraction sub-module extracts the signal response time and abnormal event number from the intelligent street lamp log based on the abnormally concentrated response block identification, identifies the corresponding response time period of the number, and maps and combines it with the road section number to generate an intelligent street lamp response block mark set;

[0149] Extract the signal response time and abnormal event number, obtain the response data of the abnormal event from the intelligent street lamp log, distinguish it through the abnormally concentrated response block identification, then determine the signal response time period corresponding to the number, map each time period to the road section number, extract each abnormal event record through the abnormally concentrated response block identification, obtain the corresponding signal response time and event number from it, judge the start and end time of the event, divide each time period based on the response time range, and at the same time, according to the road section number, correspond each time period to the specific road section number one by one, and finally combine them into a block mark set. For example, assume that the response time of an intelligent street lamp is 20 seconds, the event number is 123, this event occurs in the area with road section number 5, the start point of the response time is 9 am, and the end point is 9:20 am, then the road section number and time period are mapped in the mark set, and finally the intelligent street lamp response block mark set is obtained. This mark set can provide data support for signal response delay calculation and instruction generation in subsequent processing.

[0150] The signal response delay calculation sub-module calls the intelligent street lamp response block marker set, extracts the signal intervention start time and the evacuation end time of the corresponding block, analyzes the time difference between the two and matches it to the block number, identifies the corresponding data between the block number and the response delay, and obtains the road section response delay list;

[0151] Extract the data in the intelligent street lamp response block marker set, extract the signal intervention start time and the evacuation end time, calculate the time difference between the two, that is, the response delay, and correspond this result to the block number. By traversing the intelligent street lamp response block marker set, obtain the signal intervention start time and the evacuation end time of each block therein, calculate the difference between these two time points. If the intervention start time is 9 o'clock and the evacuation end time is 9:15, then the delay is 15 minutes. Then associate the delay value with the block number to form a delay data. Perform the same processing for each block, and finally summarize and generate the road section response delay list, which records the signal response delay situation of each road section. The list provides data support for the generation of subsequent signal control instructions.

[0152] The signal control instruction generation sub-module, according to the road section response delay list, compares with the original reaction duration, rearranges the signal priority sequence according to the delay degree, and combines the road section number to allocate the adjusted traffic light cycle parameter value to obtain the signal control instruction for the key traffic area;

[0153] Rearrange the signal priority according to the delay degree. The operations in this process include sorting the delay data and sorting the signal priority from long to short according to the time delay. Then combine the road section number to allocate the adjusted traffic light cycle parameter value to each signal. For example, assume that the response delay of a certain road section is 30 seconds, while the delay of another road section is 10 seconds. Then give priority to adjusting the signal cycle of the road section with a longer delay to ensure smooth traffic. The ultimate goal of the control instruction is to optimize the traffic flow by adjusting the signal light cycle and alleviate the traffic congestion caused by the delay. Through the comparative analysis of the delay data, the control priority of the signal lights can be effectively allocated to obtain the signal control instruction for the key traffic area to ensure that the key area can receive more effective signal control.

[0154] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical content of the technical solution of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still belong to the protection scope of the technical solution of the present invention.

Claims

1. A road traffic management system based on intelligent street lights, characterized in that The system includes: The data perception module measures data based on the smart street lamp geomagnetic sensor and the traffic camera component, including the number of lane occupations, the passing vehicle speed, and the number of vehicles passing through within the signal cycle. By combining the number of occupations per unit time and the passing time period, it analyzes the average passing saturation degree of the intersection and obtains the passing saturation value of the intersection; The traffic flow evaluation module, based on the passing saturation value of the intersection, identifies the queue length and the total number of vehicles passing through during the green light in the signal control cycle. By comparing the ratio of the two, it determines whether the current traffic flow reaches the adjustment threshold and obtains the signal load prediction index; The signal adjustment module, according to the signal load prediction index, determines whether it deviates from the green light priority interval by comparing the cycle ratio of the number of vehicles passing through the green light and the red light waiting time, and conducts a cycle deviation determination for adjacent intersections to obtain the signal control synchronization adjustment instruction; The sudden event identification module calls the signal control synchronization adjustment instruction. By analyzing the vehicle density distribution map and the speed mutation points obtained from the front-end image acquisition component, it identifies the concentrated appearance of the dense vehicle shadow area and the speed sudden drop points, screens the abnormal sections and associates the adjustment instruction range to obtain the abnormal concentrated response block identifier; The sudden event identification module includes: The vehicle shadow density identification sub-module calls the signal control synchronization adjustment instruction, extracts the vehicle density distribution map, analyzes the pixel aggregation, boundary overlap, and lane passing space parameters of the image frame, and compares the spacing fluctuations between vehicle shadows in the area to obtain the vehicle shadow density coefficient value; The speed mutation judgment sub-module, according to the vehicle shadow density coefficient value, identifies the original vehicle speed information, extracts the speed jump amplitude and frequency, and analyzes the deviation from the median value of the speed fluctuation interval to obtain the speed jump comparison value; The abnormal section screening sub-module, according to the speed jump comparison value, extracts the block number, response time sequence, and boundary coverage rate involved in the signal control adjustment instruction, and uses the formula: ； Identifies the abnormal intensity and maps it to the signal control number to obtain the abnormal concentrated response block identifier; Among them, represents the abnormal concentration response block identifier, is the sub-block in the area of the vehicle shadow overlap degree, is the sub-block in the area of the minimum vehicle speed, is the response time series equilibrium value of the area is the boundary coverage rate of the area represents the total number of areas.​​ 2. The road traffic management system based on intelligent street lamps according to claim 1, characterized in that, The passing saturation value of the intersection includes the passing density distribution, the vehicle speed change amplitude, and the signal cycle passing rate. The signal load prediction index includes the queue growth ratio, the green light passing saturation degree, and the cycle utilization efficiency. The signal control synchronization adjustment instruction includes the timing deviation between adjacent intersections, the green light priority interval determination value, and the signal cycle correction factor. The abnormal concentrated response block identifier includes the abnormal density section number, the speed mutation identification label, and the image aggregation area identifier.

3. The road traffic management system based on smart street lamps according to claim 1, characterized in that, The data perception module includes: The occupancy detection sub-module measures data based on the smart street lamp geomagnetic sensor and the traffic camera component, extracts the vehicle induction data and the lane image frame of the camera component, analyzes the matching relationship between the vehicle trajectory and the induction point signal, counts the number of vehicle occupations on each lane, and generates a signal cycle lane occupancy sequence table; The passing efficiency sub-module, based on the signal cycle lane occupancy sequence table, extracts the start and end coordinates and timestamps of the trajectories in the traffic camera component, identifies the passing distance and time interval of the continuous trajectories, summarizes the average passing speed of the lanes and the number of vehicles passing through within the signal cycle, and generates an intersection passing speed and traffic flow index set; The saturation ratio evaluation sub-module, based on the intersection passing speed and the traffic flow index set, identifies the passing time and passing quantity data of each lane, and numerically calculates the lane saturation state according to the ratio of the lane passing demand to the available time period per unit time, obtaining the intersection passing saturation value.

4. The road traffic management system based on smart street lamps according to claim 3, characterized in that, The traffic flow evaluation module includes: The passing saturation value calculation sub-module, based on the intersection passing saturation value, extracts the number of vehicles passing through per unit time and the upper limit of the lane passing capacity, identifies the ratio of the vehicle flow passing volume to the theoretical passing volume per cycle, and obtains the passing saturation value; The queue proportion determination sub-module calls the passing saturation value, counts the number of queuing vehicles and the number of passing vehicles in real time during the green light period of each cycle, analyzes the proportion difference between the two, and compares it with a fixed passing benchmark ratio to obtain the queue occupancy ratio; The signal load identification sub-module, according to the queue occupancy ratio, combines the original queue trend and the total number of passing vehicles in the current signal cycle, and uses the formula: ； Identify the signal cycle load degree, and compare it with the set signal load adjustment threshold to obtain the signal load prediction index; Among them, represents the signal load prediction index, represents the current number of queuing vehicles, represents the total number of vehicles passing through the green light, represents the traffic saturation value, represents the number of queuing vehicles in the represents the duration of the green light in the represents the number of vehicles passing through during the green light in the represents the queuing occupancy ratio, represents the average value of the original queuing trend value, represents the selected number of cycles.

5. The road traffic management system based on intelligent street lights according to claim 4, wherein, The signal adjustment module includes: The priority interval judgment sub-module, according to the signal load prediction index, combines the number of vehicles passing through the green light and the cumulative waiting time of the red light collected by the intelligent street lamp platform, and compares it with the set green light priority threshold to judge whether the signal cycle deviates from the priority control strategy and obtain the passing control deviation state; The signal timing correction sub-module, according to the passing control deviation state, extracts the green light passing rate and the green light duration in the passing direction in the intelligent street lamp regulation library, combines the mean headway and the queue vehicle density, aggregates the passing efficiency and passing intensity differences in each direction, and uses the formula: ； Calculate the signal timing deviation level, and adjust the order of the signal timing scheme according to the level range to construct a light control rhythm rearrangement scheme; Among them, represents the signal timing deviation level, is the offset of the green light passing rate, is the number of vehicles passing through the green light, is the number of vehicles waiting at the red light, is the queue length in the direction, is the green light time in the direction, is the number of vehicles passing through in the direction, is the queue density in the direction; is the total number of passing directions; The light control synchronization instruction acquisition sub-module calls the light control rhythm rearrangement scheme, collects the signal start time and the synchronization deviation reference value of the current intersection and the adjacent intersection in the intelligent street lamp control gateway, and judges whether the rearrangement scheme causes cycle misalignment and overlap to obtain the signal control synchronization adjustment instruction.

6. The road traffic management system based on intelligent street lights according to claim 1, characterized in that, The system further includes a response optimization module: Based on the abnormal centralized response block identifier, the response optimization module extracts the time difference between the signal intervention delay and the evacuation result by calling the original signal adjustment record, compares the current block number with the original response duration, and adjusts the trigger priority sequence to obtain the signal regulation instruction for the key traffic area; The signal regulation instruction for the key traffic area includes the signal response priority, the original intervention delay data, and the regulation block number matching value.

7. The road traffic management system based on intelligent street lamps according to claim 6, wherein, The response optimization module includes: The block identifier extraction sub-module, based on the abnormal centralized response block identifier, extracts the signal response time and the abnormal event number in the intelligent street lamp log, identifies the corresponding response time period of the number, and maps and combines it with the road section number to generate an intelligent street lamp response block mark set; The signal response delay calculation sub-module calls the intelligent street lamp response block mark set, extracts the signal intervention start time and the evacuation end time of the corresponding block, analyzes the time difference between the two and matches it to the block number, identifies the corresponding data between the block number and the response delay, and obtains the road section response delay list; The signal regulation instruction generation sub-module, according to the road section response delay list, compares with the original reaction duration, rearranges the signal priority sequence according to the delay degree, and combines with the road section number to allocate the adjusted traffic light cycle parameter value, so as to obtain the signal regulation instruction for the key traffic area.

Images