Light environment planning management method based on traffic and pedestrian perception information

Abstract

The application discloses a light environment planning management method based on traffic and pedestrian perception information, belongs to the technical field of data processing methods specially applicable to management purposes, collects intersection historical information, constructs a vehicle flow prediction model under weather influence and a pedestrian flow prediction model under weather influence, substitutes into a target time to obtain a vehicle flow prediction value and a pedestrian flow prediction value, calculates a traffic comprehensive congestion value according to a traffic comprehensive congestion calculation strategy, collects a light brightness value in an intersection environment, calculates a light environment brightness output value according to a light environment brightness output value calculation strategy, adjusts the brightness of a light environment module group according to the calculated light environment brightness output value, and adjusts the output brightness of the light environment module group according to the real-time situation of the intersection, which helps to improve the safety of intersection passing and the intelligentization of light environment adjustment.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of data processing methods specially applicable to management purposes, in particular to a light environment planning management method based on traffic and pedestrian perception information. BACKGROUND

[0002] Light environment refers to the light conditions and illumination intensity in the surrounding environment. Light environment has an important influence on people's life and work, including factors such as illuminance, light color, light straightness, and light reflection. Designing a reasonable light environment is of great significance to people's health and comfort. An inappropriate light environment at an intersection easily limits the line of sight and reduces visibility, which can easily lead to traffic accidents.

[0003] The prior art lacks a traffic flow prediction model constructed based on intersection historical information, and further lacks adjustment of intersection light environment brightness in combination with actual intersection light brightness values. For example, a kind of adjustable lighting system based on environmental light and traffic information is disclosed in Chinese patent application No. CN116782464A. The invention relates to a lighting system, in particular to a kind of adjustable lighting system based on environmental light and traffic information, which comprises a control unit installed in a distributed lighting device on the roadside. The control unit collects surrounding environment images through a first image acquisition module and detects environmental light intensity using surrounding environment images through an environmental light intensity detection module. The control unit detects the road surface through a road detection module and processes the echo signal obtained by the road detection module through an echo signal processing module. The control unit obtains distance information and speed information of multiple detection points based on the echo signal processing result through a detection point information acquisition module, and determines effective point traces based on the echo signal processing result through a point trace determination module.

[0004] In order to adjust the output brightness of the light environment module according to the real-time situation of the intersection, improve the safety of intersection traffic, and intelligentize the adjustment of the light environment, the present application designs a light environment planning management method based on traffic and pedestrian perception information. SUMMARY

[0005] In view of the shortcomings of the prior art, the present application proposes a light environment planning management method based on traffic and pedestrian perception information. The present application collects intersection historical information, constructs a vehicle flow prediction model under the influence of weather and a pedestrian flow prediction model under the influence of weather, substitutes the target time to obtain vehicle flow prediction values and pedestrian flow prediction values, calculates traffic comprehensive congestion values according to a traffic comprehensive congestion calculation strategy, collects light brightness values in the intersection environment, calculates light environment brightness output values according to a light environment brightness output value calculation strategy, adjusts the brightness of the intersection light environment module according to the calculated light environment brightness output values, adjusts the output brightness of the light environment module according to the real-time situation of the intersection, which helps to improve the safety of intersection traffic and intelligentize the adjustment of the light environment.

[0006] To achieve the above object, the light environment planning management method based on traffic and pedestrian perception information provides the following technical solutions:

[0007] The light environment planning management method based on traffic and pedestrian perception information includes the following specific steps:

[0008] Step one, collect intersection historical information, build a traffic volume prediction model under the influence of weather, and substitute the target time to get the traffic volume prediction value;

[0009] Step two, collect intersection historical information, build a pedestrian flow prediction model under the influence of weather, and substitute the target time to get the pedestrian flow prediction value;

[0010] Step three, substitute the traffic volume prediction value and pedestrian flow prediction value of the target time into the traffic comprehensive congestion calculation strategy to calculate the traffic comprehensive congestion value;

[0011] Step four, collect the light intensity value in the intersection environment, substitute the light intensity value in the intersection environment and the traffic comprehensive congestion value into the light environment brightness output value calculation strategy to calculate the light environment brightness output value, and adjust the brightness of the intersection light environment module according to the calculated light environment brightness output value.

[0012] Specifically, the step one includes the following specific steps:

[0013] S11, collect intersection historical information and store it in the vehicle information storage module, the intersection historical information includes weather information, time and traffic volume, the weather information includes temperature, precipitation and visibility, substitute the weather information into the weather severity value calculation strategy to calculate the weather severity value, the weather severity value calculation strategy includes a weather severity value calculation formula, the weather severity value calculation formula is: Wherein, is the collected intersection temperature, is the intersection temperature threshold, is the collected intersection precipitation, is the intersection precipitation threshold, is the collected intersection visibility, is the intersection visibility threshold, is the intersection temperature proportion coefficient, is the intersection precipitation proportion coefficient, is the intersection visibility proportion coefficient, and ;

[0014] It should be noted that the intersection temperature is monitored by a temperature sensor, the intersection precipitation is monitored by a rain and snow sensor, and the intersection visibility is monitored by an infrared sensor; The intersection temperature threshold, intersection precipitation threshold and intersection visibility threshold here are determined according to the needs of the scene;

[0015] S12, the intersection history information dataset is divided into 2 subsets, the first 85% is the training dataset, and the last 15% is the test dataset, the first 85% training dataset is input into the traffic flow neural network prediction model under the influence of weather for training, and the initial traffic flow neural network prediction model under the influence of weather is obtained, and then the last 15% test dataset is used to test the initial traffic flow neural network prediction model under the influence of weather, and the traffic flow neural network prediction model under the influence of weather with the highest accuracy of traffic flow judgment is output;

[0016] S13, the output strategy formula of the specific neuron in the traffic flow neural network prediction model under the influence of weather is: , wherein, is the output of the i-th neuron in the j-th layer of the traffic flow neural network prediction model under the influence of weather, is the output of the i-th neuron in the j-th layer of the traffic flow neural network prediction model under the influence of weather, is the connection weight of the i-th neuron in the j-th layer, is the output of the i-th neuron in the j-th layer of the traffic flow neural network prediction model under the influence of weather, is the linear relationship bias of the i-th neuron in the j-th layer, is the Sigmoid activation function. S14, the traffic flow neural network prediction model under the influence of weather with the highest accuracy of traffic flow judgment is judged by the traffic flow prediction error value, and the traffic flow prediction error value calculation formula is: , wherein,

[0017] represents the weight coefficient of the traffic flow prediction value in the i-th period under the influence of weather, , is the prediction period of the traffic flow, and the prediction period is half an hour, represents the traffic flow prediction value in the i-th period under the influence of weather, is the traffic flow in the i-th period under the influence of weather, is the prediction error of the traffic flow neural network prediction model under the influence of weather in the i-th period. ​​​​​​​​​​​​​​​

[0018] S15. Input the target time and the weather severity value at the target time into the traffic flow neural network prediction model under the influence of weather to obtain the traffic flow prediction value at the target time.

[0019] Specifically, step two includes the following steps:

[0020] S21. Collect historical information about the intersection and store it in the pedestrian information storage module. The historical information about the intersection includes weather information, time and pedestrian flow. Divide the historical information dataset into two subsets, with the first 85% being the training dataset and the last 15% being the test dataset. Input the first 85% of the training dataset into the pedestrian flow neural network prediction model under the influence of weather for training to obtain the initial pedestrian flow neural network prediction model under the influence of weather. Then use the last 15% of the test dataset to test the initial pedestrian flow neural network prediction model under the influence of weather and output the pedestrian flow neural network prediction model under the influence of weather with the highest accuracy in judging pedestrian flow.

[0021] S22. The output strategy formula for a specific neuron in the neural network prediction model for pedestrian flow under weather influence is as follows: ,in, The first neural network prediction model for pedestrian flow under the influence of weather layer The output of the term neuron, The first neural network prediction model for pedestrian flow under the influence of weather Layer neurons and layer The connection weights of the term neurons, The first neural network prediction model for pedestrian flow under the influence of weather Layer neurons The output, The first neural network prediction model for pedestrian flow under the influence of weather Layer neurons and layer Bias in the linear relationship of term neurons Use the Sigmoid activation function;

[0022] S23. The neural network prediction model for pedestrian flow under weather conditions, which has the highest accuracy in judging pedestrian flow, is determined by the pedestrian flow prediction error value. The formula for calculating the pedestrian flow prediction error value is as follows: ,in, Indicates the first under the influence of weather Weighting coefficients for the predicted pedestrian flow values ​​over a given period , The forecasting period for human traffic. a traffic flow prediction value of the target time under the weather influence, a traffic flow prediction value of the target time under the weather influence, a traffic flow prediction value of the target time under the weather influence, a traffic flow prediction value of the target time under the weather influence, a prediction error of the traffic flow neural network prediction model of the target time under the weather influence; a prediction error of the traffic flow neural network prediction model of the target time under the weather influence;

[0023] S24, inputting the target time and the weather severity value of the target time into the traffic flow neural network prediction model to obtain a traffic flow prediction value of the target time.

[0024] Specifically, the step three includes the following specific steps:

[0025] The traffic comprehensive congestion value is calculated by substituting the obtained traffic flow prediction value of the target time and the traffic flow prediction value of the target time into the traffic comprehensive congestion calculation strategy, wherein the traffic comprehensive congestion calculation strategy includes a traffic comprehensive congestion calculation formula, and the traffic comprehensive congestion calculation formula is: wherein, a traffic flow prediction value under the weather influence, a traffic flow prediction value under the weather influence, a proportion coefficient of the traffic flow under the weather influence, a proportion coefficient of the traffic flow under the weather influence, and .

[0026] Specifically, the step four includes the following specific steps:

[0027] S41, collecting the light brightness value in the intersection environment, and substituting the light brightness value in the intersection environment and the traffic comprehensive congestion value into the light environment brightness output value calculation strategy to calculate the light environment brightness output value, wherein the light environment brightness output value calculation strategy includes a light environment brightness output value calculation formula, and the light environment brightness output value calculation formula is: wherein, a light brightness value in the intersection, a set rated brightness;

[0028] It should be noted that the light brightness value in the intersection environment is monitored by an illuminance sensor;

[0029] S42, adjusting the brightness of the intersection light environment module according to the calculated light environment brightness output value.

[0030] It should be noted that, , , , The value of the set rated brightness is obtained by selecting 5000 sets of intersection historical information, obtaining weather, time, vehicle flow and pedestrian flow, calculating the influence of vehicle flow under the influence of weather and pedestrian flow under the influence of weather on the traffic comprehensive congestion value, substituting the traffic congestion value and the intersection light brightness value into the light environment brightness comprehensive value calculation formula to calculate the light environment brightness comprehensive value, inviting 500 experts in the field to score the light environment module brightness required by the intersection to obtain, and importing the light environment brightness comprehensive value and the scored brightness value into related fitting software to output a set of light environment brightness value with the highest accuracy 、 、 、 and the value of the set rated brightness.

[0031] The light environment planning and management system based on traffic and pedestrian perception information is realized based on the light environment planning and management method based on traffic and pedestrian perception information, and comprises:

[0032] An information collection module is configured to collect intersection historical information and intersection light brightness values, wherein the intersection historical information comprises weather, time, vehicle flow and pedestrian flow.

[0033] An information storage module is configured to store intersection historical information, and is divided into a vehicle information storage module and a pedestrian information storage module.

[0034] A vehicle flow neural network prediction module is configured to obtain intersection historical information, construct a vehicle flow prediction model under the influence of weather, and substitute a target time to obtain a vehicle flow prediction value.

[0035] A pedestrian flow neural network prediction module is configured to obtain intersection historical information, construct a pedestrian flow prediction model under the influence of weather, and substitute a target time to obtain a pedestrian flow prediction value.

[0036] A traffic comprehensive congestion value module is configured to substitute the vehicle flow prediction value at the target time and the pedestrian flow prediction value at the target time into a traffic comprehensive congestion calculation strategy to calculate a traffic comprehensive congestion value.

[0037] A light environment brightness comprehensive value calculation module is configured to obtain light brightness values in an intersection environment, substitute the light brightness values in the intersection environment and the traffic comprehensive congestion value into a light environment brightness output value calculation strategy to calculate a light environment brightness output value.

[0038] A control module is configured to control the operation of the information collection module, the information storage module, the vehicle flow neural network prediction module, the pedestrian flow neural network prediction module, the traffic comprehensive congestion value module and the light environment brightness comprehensive value calculation module.

[0039] An electronic device comprises a processor and a memory, wherein the memory stores a computer program invokable by the processor, and the processor executes the above-mentioned light environment planning management method based on traffic and pedestrian perception information by invoking the computer program stored in the memory.

[0040] A computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to execute the above-mentioned light environment planning management method based on traffic and pedestrian perception information.

[0041] The beneficial effects of the present application are: collecting intersection historical information, constructing a traffic flow prediction model under the influence of weather and a pedestrian flow prediction model under the influence of weather, substituting target time to obtain traffic flow prediction value and pedestrian flow prediction value, calculating traffic comprehensive congestion value according to traffic comprehensive congestion calculation strategy, collecting light brightness value in intersection environment, calculating light environment brightness output value according to light environment brightness output value calculation strategy, adjusting intersection light environment module brightness according to the calculated light environment brightness output value, adjusting the output brightness of the light environment module according to the real-time situation of the intersection, which helps to improve the safety of intersection passing and the intelligentization of light environment adjustment. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 The flowchart of the light environment planning management method based on traffic and pedestrian perception information of the present application is shown in the figure.

[0043] Figure 2 The overall framework diagram of the light environment planning management system based on traffic and pedestrian perception information of the present application is shown in the figure. DETAILED DESCRIPTION

[0044] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations may, however, be implemented in many different forms and should not be construed as limited to the implementations set forth herein; rather, these implementations are provided so that this disclosure will be thorough and complete, and will fully convey the inventive aspects to those skilled in the art. Features described in the description, examples, or claims can be combined in any suitable manner in one or more implementations.

[0045] Example 1

[0046] Referring to Figure 1 An embodiment provided by the present application is a light environment planning management method based on traffic and pedestrian perception information, which comprises the following specific steps:

[0047] Step one, collect intersection historical information, construct a traffic flow prediction model under the influence of weather, and substitute target time to obtain traffic flow prediction value;

[0048] In this embodiment, step one comprises the following specific steps:

[0049] S11, collect intersection historical information and store in vehicle information storage module, the intersection historical information includes weather information, time and traffic flow, the weather information includes temperature, precipitation and visibility, the weather information is substituted into the weather bad value calculation strategy to calculate the weather bad value, the weather bad value calculation strategy includes weather bad value calculation formula, the weather bad value calculation formula is: Wherein, It is the collected intersection temperature, It is the intersection temperature threshold, It is the collected intersection precipitation, It is the intersection precipitation threshold, It is the collected intersection visibility, It is the intersection visibility threshold, It is the intersection temperature proportion coefficient, It is the intersection precipitation proportion coefficient, It is the intersection visibility proportion coefficient, And ;

[0050] It should be noted that the intersection temperature is monitored by a temperature sensor, the intersection precipitation is monitored by a rain and snow sensor, and the intersection visibility is monitored by an infrared sensor;The intersection temperature threshold, the intersection precipitation threshold and the intersection visibility threshold are determined according to the needs of the scene, the change of temperature will affect the condition of the road, such as freezing at low temperature, increasing the risk of vehicle driving;High temperature may cause the road to be soft, affecting the stability of the vehicle;The increase of precipitation will reduce the friction of the road, increase the possibility of vehicle skidding, especially in rainy and snowy weather, which will increase the probability of traffic accidents;The reduction of visibility will affect the driver's vision, increase the possibility of rear-end accidents and other accidents.

[0051] S12, the intersection historical information data set is divided into two subsets, the first 85% is the training data set, and the last 15% is the test data set, the first 85% of the training data set is input into the weather influence under the traffic flow neural network prediction model for training, and the initial weather influence under the traffic flow neural network prediction model is obtained, then the last 15% of the test data set is used to test the initial weather influence under the traffic flow neural network prediction model, and the weather influence under the traffic flow neural network prediction model with the highest accuracy of traffic flow judgment is output;

[0052] S13, the output strategy formula of the specific neuron in the weather influence under the traffic flow neural network prediction model is: Wherein, It is the output of the first Layer term neuron of the weather influence under the traffic flow neural network prediction model, the first layer of the weather-affected traffic flow neural network prediction model layer neurons and layer connection weights of the layer neurons, the first layer of the weather-affected traffic flow neural network prediction model layer neurons output, the first layer of the weather-affected traffic flow neural network prediction model layer neurons and layer bias of the linear relationship of the layer neurons, is a Sigmoid activation function;

[0053] S14, the weather-affected traffic flow neural network prediction model with the highest traffic flow judgment accuracy is determined by a traffic flow prediction error value, and the traffic flow prediction error value calculation formula is: wherein, represents a weight coefficient of the traffic flow prediction value in the weather-affected period, , is a prediction period of the traffic flow, and the prediction period is half an hour, represents the traffic flow prediction value in the weather-affected period, is the traffic flow in the weather-affected period is the prediction error of the weather-affected traffic flow neural network prediction model in the period; S15, input the target time and the weather severity value of the target time into the weather-affected traffic flow neural network prediction model to obtain the traffic flow prediction value of the target time. Step two, collect intersection historical information, construct a weather-affected traffic flow prediction model, and substitute the target time to obtain the traffic flow prediction value;

[0054] In this embodiment, step two includes the following specific steps:

[0055] Step two, collect intersection historical information, construct a weather-affected traffic flow prediction model, and substitute the target time to obtain the traffic flow prediction value;

[0056] In this embodiment, step two includes the following specific steps:

[0057] ​​S21, collect intersection historical information and store in pedestrian information storage module, intersection historical information includes weather information, time and passenger flow, divide intersection historical information data set into 2 subsets, wherein the first 85% is training data set, the last 15% is test data set, input the first 85% training data set into weather influence under the passenger flow neural network prediction model for training, obtain the initial weather influence under the passenger flow neural network prediction model, then test the initial weather influence under the passenger flow neural network prediction model with the last 15% test data set, output the weather influence under the passenger flow neural network prediction model with the highest passenger flow judgment accuracy;

[0058] S22, the output strategy formula of specific neurons in the weather influence under the passenger flow neural network prediction model is: , wherein, is the output of the first layer term neuron of the weather influence under the passenger flow neural network prediction model, is the output of the first layer neuron of the weather influence under the passenger flow neural network prediction model, is the connection weight of the first layer term neuron of the weather influence under the passenger flow neural network prediction model, is the output of the first layer neuron of the weather influence under the passenger flow neural network prediction model, is the bias of the linear relationship of the first layer neuron and the first layer term neuron of the weather influence under the passenger flow neural network prediction model,

[0059] S23, the weather influence under the passenger flow neural network prediction model with the highest passenger flow judgment accuracy is judged through the passenger flow prediction error value, and the passenger flow prediction error value calculation formula is: , wherein, represents the weight coefficient of the passenger flow prediction value in the first period under the weather influence, , is the prediction period of the passenger flow, represents the passenger flow prediction value in the first period under the weather influence, is the passenger flow in the first period under the weather influence is the prediction error of the weather influence under the passenger flow neural network prediction model in period.

[0060] S24, input the target time and the weather bad value of the target time into the pedestrian flow neural network prediction model under the influence of weather to obtain a pedestrian flow prediction value of the target time.

[0061] Step three, substituting the obtained traffic flow prediction value of the target time and the pedestrian flow prediction value of the target time into the traffic comprehensive congestion calculation strategy to calculate a traffic comprehensive congestion value;

[0062] In this embodiment, step three includes the following specific steps:

[0063] Substituting the obtained traffic flow prediction value of the target time and the pedestrian flow prediction value of the target time into the traffic comprehensive congestion calculation strategy to calculate a traffic comprehensive congestion value, the traffic comprehensive congestion calculation strategy includes a traffic comprehensive congestion calculation formula, and the traffic comprehensive congestion calculation formula is: , wherein, is the traffic flow prediction value under the influence of weather, is the pedestrian flow prediction value under the influence of weather, is the proportion coefficient of traffic flow under the influence of weather, is the proportion coefficient of pedestrian flow under the influence of weather, and .

[0064] Step four, collecting the light brightness value in the intersection environment, substituting the light brightness value in the intersection environment and the traffic comprehensive congestion value into the light environment brightness output value calculation strategy to calculate the light environment brightness output value, and adjusting the brightness of the intersection light environment module according to the calculated light environment brightness output value.

[0065] It should be noted that the light brightness value in the intersection environment is monitored by an illuminance sensor. In the intersection environment, the light brightness value will affect the visual ability and judgment ability of the driver, thereby affecting the safety and smoothness of the traffic. When the light brightness is too low, the visual ability of the driver will be limited, and it is difficult to clearly see the signs, traffic lights, other vehicles and pedestrians on the road, and traffic accidents are likely to occur. When the light brightness is too high, it will cause glare, affect the driver's vision, cause eye fatigue and distraction, and also increase the risk of traffic accidents. Therefore, when designing and managing the light environment of the intersection, it is necessary to reasonably control the brightness value.

[0066] In this embodiment, step four includes the following specific steps:

[0067] S41, collecting the light brightness value in the intersection environment, substituting the light brightness value in the intersection environment and the traffic comprehensive congestion value into the light environment brightness output value calculation strategy to calculate the light environment brightness output value, the light environment brightness output value calculation strategy includes a light environment brightness output value calculation formula, and the light environment brightness output value calculation formula is: wherein, is the intersection light brightness value, is the set rated brightness;

[0068] S42, adjusting the intersection light environment module brightness according to the calculated light environment brightness output value.

[0069] It should be noted that, , , , and the set rated brightness is obtained by: selecting 5000 sets of intersection historical information, obtaining weather, time, traffic flow and pedestrian flow, calculating the influence of traffic flow under the influence of weather and pedestrian flow under the influence of weather on traffic comprehensive congestion value, substituting traffic congestion value and intersection light brightness value into light environment brightness comprehensive value calculation formula to calculate light environment brightness comprehensive value, hiring 500 experts in the field to score the light environment module brightness required by the intersection, and importing the light environment brightness comprehensive value and the scored brightness value into the related fitting software to output the set of , , , and the set rated brightness.

[0070] Example 2

[0071] Please refer to Figure 2 , the light environment planning and management system based on traffic and pedestrian perception information, which is based on the above light environment planning and management method based on traffic and pedestrian perception information, which includes:

[0072] The information collection module is used for collecting intersection historical information and intersection light brightness value, and the intersection historical information includes weather, time, traffic flow and pedestrian flow.

[0073] The information storage module is used for storing intersection historical information, which is divided into vehicle information storage module and pedestrian information storage module.

[0074] The traffic flow neural network prediction module is used for obtaining intersection historical information, constructing a traffic flow prediction model under the influence of weather, and substituting the target time to obtain the traffic flow prediction value.

[0075] The pedestrian flow neural network prediction module is used for obtaining intersection historical information, constructing a pedestrian flow prediction model under the influence of weather, and substituting the target time to obtain the pedestrian flow prediction value.

[0076] The traffic comprehensive congestion value module is used for substituting the obtained traffic flow prediction value at the target time and the pedestrian flow prediction value at the target time into the traffic comprehensive congestion calculation strategy to calculate the traffic comprehensive congestion value.

[0077] The light environment brightness comprehensive value calculation module is configured to obtain the light brightness value in the intersection environment, and calculate the light environment brightness output value by substituting the light brightness value in the intersection environment and the traffic comprehensive congestion value into the light environment brightness output value calculation strategy.

[0078] The control module is configured to control the operation of the information acquisition module, the information storage module, the traffic flow neural network prediction module, the passenger flow neural network prediction module, the traffic comprehensive congestion value module, and the light environment brightness comprehensive value calculation module.

[0079] Embodiment 3

[0080] The embodiment provides an electronic device, which comprises a processor and a memory, wherein the memory stores a computer program that can be invoked by the processor, and the processor executes the light environment planning management method based on traffic and pedestrian perception information by invoking the computer program stored in the memory.

[0081] The electronic device can have great differences due to different configurations or performances, and can comprise one or more processors (Central Processing Units, CPUs) and one or more memories, wherein the memory stores at least one computer program, the computer program is loaded and executed by the processor to implement the light environment planning management method based on traffic and pedestrian perception information provided by the above method embodiment, which comprises the following specific steps: collecting intersection historical information, constructing a traffic flow prediction model under the influence of weather and a passenger flow prediction model under the influence of weather, substituting target time to obtain traffic flow prediction value and passenger flow prediction value, calculating traffic comprehensive congestion value according to the traffic comprehensive congestion calculation strategy, collecting light brightness value in the intersection environment, calculating light environment brightness output value according to the light environment brightness output value calculation strategy, and adjusting the brightness of the intersection light environment module according to the calculated light environment brightness output value.

[0082] Embodiment 4

[0083] The embodiment provides a computer readable storage medium, which stores instructions, and when a computer program runs on a computer device, the computer device executes the light environment planning management method based on traffic and pedestrian perception information.

[0084] For example, the computer readable storage medium can be a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a read-only compact disc (Compact Disc Read-Only Memory, CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device.

[0085] It should be understood that the size of the serial number of the above processes does not mean the order of execution in various embodiments of the present application, and the execution order of the processes should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0086] It should be understood that determining B according to A does not mean that B is determined only according to A, but B can also be determined according to A and / or other information.

[0087] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present application can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

Claims

1. A method for managing light environment planning based on traffic and pedestrian perception information, characterized in that, It comprises the following specific steps: Step one, collect intersection historical information, build a traffic flow prediction model under the influence of weather, and substitute the target time to get the traffic flow prediction value; Step two, collect intersection historical information, build a pedestrian flow prediction model under the influence of weather, and substitute the target time to get the pedestrian flow prediction value; Step three, substitute the obtained traffic flow prediction value and pedestrian flow prediction value of the target time into the traffic comprehensive congestion calculation strategy to calculate the traffic comprehensive congestion value; Step four, collect the light brightness value in the intersection environment, substitute the light brightness value in the intersection environment and the traffic comprehensive congestion value into the light environment brightness output value calculation strategy to calculate the light environment brightness output value, and adjust the brightness of the intersection light environment module according to the calculated light environment brightness output value; The step one comprises the following specific steps: S11, collecting intersection historical information and storing in the vehicle information storage module, the intersection historical information includes weather information, time and traffic flow, the weather information includes temperature, precipitation and visibility, substituting the weather information into a weather bad value calculation strategy to calculate a weather bad value, the weather bad value calculation strategy includes a weather bad value calculation formula, the weather bad value calculation formula is: Wherein, is the collected intersection temperature, is an intersection temperature threshold value, is the collected intersection precipitation, is an intersection precipitation threshold value, is the collected intersection visibility, is an intersection visibility threshold value, is an intersection temperature proportionality coefficient, is an intersection precipitation proportionality coefficient, is an intersection visibility proportionality coefficient, and ; S12, divide the intersection historical information dataset into two subsets, of which the first 85% is the training dataset and the last 15% is the test dataset, input the first 85% training dataset into the traffic flow neural network prediction model under the influence of weather for training, obtain the initial traffic flow neural network prediction model under the influence of weather, and then test the initial traffic flow neural network prediction model under the influence of weather with the last 15% test dataset, output the traffic flow neural network prediction model under the influence of weather with the highest accuracy of traffic flow judgment; S13, the first in the neural network prediction model for traffic flow under weather influence. layer The output strategy formula for the term neuron is: ,in, The first neural network prediction model for traffic flow under weather influence layer The output of the term neuron, The first neural network prediction model for traffic flow under weather influence Layer neurons and layer The connection weights of the term neurons, The first neural network prediction model for traffic flow under weather influence Layer neurons The output, The first neural network prediction model for traffic flow under weather influence Layer neurons and layer Bias in the linear relationship of term neurons Use the Sigmoid activation function; S14. The neural network prediction model for traffic flow under weather conditions, which has the highest accuracy in judging traffic flow, is judged by the traffic flow prediction error value. The formula for calculating the traffic flow prediction error value is as follows: ,in, Indicates the first under the influence of weather The weighting coefficients of the traffic flow forecast values ​​for each period. , For the traffic flow forecast period, Indicates the first under the influence of weather Traffic flow forecast for each period, Traffic flow affected by weather in the first The actual value for each period, A neural network prediction model for traffic flow under weather influences. The prediction error over each period is considered, with the smallest error value corresponding to the highest accuracy. S15, input the target time and the weather severity value of the target time into the traffic flow neural network prediction model under the influence of weather to obtain the traffic flow prediction value of the target time; The step two comprises the following specific steps: S21, collect intersection historical information and store it in the pedestrian information storage module, the intersection historical information includes weather information, time and pedestrian flow, divide the intersection historical information dataset into two subsets, of which the first 85% is the training dataset and the last 15% is the test dataset, input the first 85% training dataset into the pedestrian flow neural network prediction model under the influence of weather for training, obtain the initial pedestrian flow neural network prediction model under the influence of weather, and then test the initial pedestrian flow neural network prediction model under the influence of weather with the last 15% test dataset, output the pedestrian flow neural network prediction model under the influence of weather with the highest accuracy of pedestrian flow judgment; S22, the output of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is layer The output strategy formula of the i-th neuron is: , wherein The output of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is layer The output of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is The connection weight of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is layer The output of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is The connection weight of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is layer The output of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is The connection weight of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is The output of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is The connection weight of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is The connection weight of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is The connection weight of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is The connection weight of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is The connection weight of the i-th neuron in the j-th layer of the human flow neural network prediction model under the influence of the weather is The S23. The neural network prediction model for pedestrian flow under weather conditions, which has the highest accuracy in judging pedestrian flow, is determined by the pedestrian flow prediction error value. The formula for calculating the pedestrian flow prediction error value is as follows: ,in, Indicates the first under the influence of weather Weighting coefficients for the predicted pedestrian flow values ​​over a given period , The forecasting period for human traffic. Indicates the first under the influence of weather Forecasted pedestrian traffic values ​​for each period The number of people affected by the weather in the first One cycle Neural network prediction model for pedestrian flow under weather influence The prediction error over one cycle; S24, input the target time and the weather severity value of the target time into the pedestrian flow neural network prediction model under the influence of weather to obtain the pedestrian flow prediction value of the target time.

2. The light environment planning management method based on traffic and pedestrian perception information according to claim 1, wherein, The step three comprises the following specific steps: The traffic flow prediction value of the target time and the people flow prediction value of the target time are substituted into a traffic comprehensive congestion calculation strategy to calculate a traffic comprehensive congestion value, the traffic comprehensive congestion calculation strategy comprising a traffic comprehensive congestion calculation formula, the traffic comprehensive congestion calculation formula being: wherein, is a traffic flow prediction value under weather influence, is a people flow prediction value under weather influence, is a proportionality coefficient of traffic flow under weather influence, is a proportionality coefficient of people flow under weather influence, and .

3. The light environment planning management method based on traffic and pedestrian perception information according to claim 2, wherein, The step four comprises the following specific steps: S41, collect the light brightness value in the intersection environment, put the light brightness value in the intersection environment and the traffic comprehensive congestion value into the light environment brightness output value calculation strategy to calculate the light environment brightness output value, the light environment brightness output value calculation strategy includes a light environment brightness output value calculation formula, the light environment brightness output value calculation formula is: Wherein, is the light brightness value of the intersection, is the set rated brightness; S42, adjust the brightness of the intersection light environment module according to the calculated light environment brightness output value.

4. A light environment planning management system based on traffic and pedestrian perception information, which implements the light environment planning management method based on traffic and pedestrian perception information according to any one of claims 1-3, characterized in that, It comprises: An information collection module for collecting intersection historical information and intersection light brightness value, the intersection historical information including weather, time, traffic flow and pedestrian flow; An information storage module for storing intersection historical information, divided into a vehicle information storage module and a pedestrian information storage module; A traffic flow neural network prediction module for obtaining intersection historical information, building a traffic flow prediction model under the influence of weather, and substituting the target time to get the traffic flow prediction value; The human flow neural network prediction module is configured to obtain intersection historical information, construct a human flow prediction model under the influence of weather, and obtain a human flow prediction value by substituting a target time. The traffic comprehensive congestion value module is configured to substitute the vehicle flow prediction value at the target time and the human flow prediction value at the target time into a traffic comprehensive congestion calculation strategy to calculate a traffic comprehensive congestion value. The light environment brightness comprehensive value calculation module is configured to obtain a light brightness value in an intersection environment, substitute the light brightness value in the intersection environment and the traffic comprehensive congestion value into a light environment brightness output value calculation strategy, and calculate a light environment brightness output value. The control module is configured to control the operation of the information acquisition module, the information storage module, the vehicle flow neural network prediction module, the human flow neural network prediction module, the traffic comprehensive congestion value module, and the light environment brightness comprehensive value calculation module.

5. An electronic device, comprising: The memory and the processor, wherein the memory stores a computer program that can be called by the processor, and the processor executes the light environment planning management method based on traffic and pedestrian perception information according to any one of claims 1-3 by calling the computer program stored in the memory. The instructions are stored in the computer, and when the instructions are run on the computer, the computer executes the light environment planning management method based on traffic and pedestrian perception information according to any one of claims 1-3.

6. A computer-readable storage medium, characterized in that: ​

Images