Method and system for intelligently adjusting traffic facilities based on optical fiber display

Abstract

The invention discloses a traffic facility intelligent adjustment method and system based on optical fiber display, and the method comprises the steps: obtaining traffic flow information through obtaining and preprocessing road surface data, determining the change time of a traffic facility, and generating a first control instruction; meanwhile, a brightness optimization model is obtained by performing preprocessing and deep learning model training on a traffic signal lamp image data set. And calculating an optimal brightness value by using the brightness optimization model in combination with the real-time environment and state data, and generating a second control instruction. Finally, the traffic facility is intelligently adjusted according to the two control instructions, self-adaptive adjustment and automatic dodging processing of the brightness of the signal lamp and the electronic roadside signboard are achieved, and controllability and stability of the traffic facility are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of traffic signal adaptive control, and in particular relates to a traffic facility intelligent adjustment method and system based on optical fiber display. Background Art

[0002] Existing traffic signs and signal lights have problems with unclear road signs at night, in fog and other conditions of low visibility. In severe weather conditions, traffic signs and signal lights may fail, and static traffic signs cannot provide real-time information updates, which has a great impact on traffic order and poses a safety hazard to road traffic.

[0003] Currently, many traffic signs rely on reflective materials to improve visibility, but these materials will gradually age over time, especially under the influence of sunlight and bad weather, their reflective effect at night is greatly reduced, making it more difficult for drivers to identify the signs. In addition, trees, buildings, billboards, etc. may block part of the traffic signs or signal lights, further reducing visibility, making it more difficult for drivers to see these key traffic information, which can easily cause traffic accidents.

[0004] Traditional traffic signs are usually static and cannot provide flexible feedback based on real-time road conditions. When a traffic accident, construction or other emergencies occur, traditional traffic signs cannot make corresponding changes, which affects the driver's judgment and makes it impossible for the driver to adjust the driving route or speed in time. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a method and system for intelligent adjustment of traffic facilities based on fiber optic display to solve the problems existing in the above-mentioned prior art.

[0006] To achieve the above object, the present invention provides a method for intelligently adjusting traffic facilities based on optical fiber display, comprising:

[0007] Acquire road surface data and preprocess it, and obtain traffic flow data based on the preprocessed road surface data; obtain traffic facility change time based on the traffic flow data; and generate a first control instruction based on the traffic facility change time;

[0008] Obtain a traffic light image dataset and preprocess the traffic light image dataset; train a deep learning model based on the preprocessed traffic light image dataset to obtain a brightness optimization model;

[0009] Acquire environmental data and status data in real time, obtain an optimal brightness value based on the environmental data, the status data and the brightness optimization model; and generate a second control instruction based on the optimal brightness value;

[0010] Intelligently adjust the traffic facilities based on the first control instruction and the second control instruction.

[0011] Optionally, the road surface data includes the length of the traffic flow waiting for the traffic light, the direction of the vehicle travel, and the vehicle flow; the process of obtaining the traffic flow data includes:

[0012] Obtaining surveillance videos from surveillance cameras installed near each traffic light, performing data processing on the surveillance videos, and obtaining the length of the traffic flow and the driving direction of the vehicles waiting for the traffic light;

[0013] The vehicle flow rate is obtained by detecting the magnetic field changes of vehicles passing by through the ring coil installed under the road surface;

[0014] The lane to which each vehicle belongs is determined based on the monitoring video, and the direction in which each vehicle will travel is determined based on the pre-acquired driving direction landmark arrow of each lane; the traffic flow length is the length of the vehicle queue.

[0015] Optionally, the process of obtaining the traffic facility change time based on the traffic flow data control includes:

[0016] Obtain the driving directions of all vehicles waiting for the traffic light, and if there is any vehicle driving direction whose traffic length is not within the preset traffic length range and is greater than the maximum value of the preset traffic length range, determine whether there is any other vehicle driving direction whose traffic length is not within the preset traffic length range and is less than the minimum value of the preset traffic length range;

[0017] If it exists, the driving direction of the vehicle whose traffic flow length is not within the preset traffic flow length range and is greater than the maximum value of the preset traffic flow length range is taken as the first driving direction; the driving direction of the vehicle whose traffic flow length is not within the preset traffic flow length range and is less than the minimum value of the preset traffic flow length range is taken as the second driving direction; wherein, the first driving direction and the second driving direction are not opposite directions;

[0018] Constructing a traffic light change time rule; and obtaining the traffic light change time corresponding to the first driving direction and the second driving direction based on the traffic light change time rule.

[0019] Optionally, a process of obtaining a traffic light image dataset and preprocessing the traffic light image dataset includes:

[0020] A public traffic light image dataset is obtained; traffic light images under different weather conditions and light intensities are collected on site to obtain a real-time dataset; the real-time dataset is standardized and the directional information of the images is cleared; the processed real-time dataset is annotated, and after data enhancement is performed on the annotated dataset, the annotated dataset is combined with the public traffic light image dataset to obtain a training dataset; wherein the annotated information includes weather conditions, ambient light intensity, and color status.

[0021] Optionally, a deep learning model is trained based on the preprocessed traffic light image dataset to obtain a brightness optimization model, including:

[0022] Two deep models, yolov8x and yolov8n, are selected, and the training data set is used to train yolov8x and yolov8n respectively; when training yolov8n, the output of yolov8x is introduced, and the distillation loss and the student loss are calculated in combination with the preset distillation parameters; yolov8n is optimized based on the distillation loss and the student loss.

[0023] Optionally, the environmental data includes weather conditions and ambient light intensity; and the status data is color status.

[0024] Optionally, the process of intelligently adjusting the traffic facility based on the second control instruction includes:

[0025] The light source configuration is obtained based on the second control instruction; the current display image of the traffic facility is obtained, and the current display image is converted based on the second control instruction to obtain an HSV image that meets the requirements of the second control instruction; the HSV image is corrected and homogenized, and then converted into an RGB image, which is transmitted to the display surface using fiber optic display technology.

[0026] The present invention also provides a traffic facility intelligent adjustment system based on optical fiber display, comprising:

[0027] A change time control instruction generation module is used to obtain and pre-process road surface data, obtain traffic flow data based on the pre-processed road surface data; obtain the traffic facility change time based on the traffic flow data; and generate a first control instruction based on the traffic facility change time;

[0028] A data sorting module, used to obtain a traffic light image dataset and pre-process the traffic light image dataset;

[0029] A model building module is used to train a deep learning model based on the preprocessed traffic light image dataset to obtain a brightness optimization model;

[0030] The brightness control instruction generation module acquires environment data and status data in real time, obtains an optimal brightness value based on the environment data, the status data and the brightness optimization model; and generates a second control instruction based on the optimal brightness value;

[0031] The integrated control module is used to intelligently adjust the traffic facilities based on the first control instruction and the second control instruction.

[0032] The present invention also provides a computer device, which includes a memory and a processor connected to the memory; the memory is used to store computer programs; the processor is used to run the computer programs stored in the memory to execute the steps of the above-mentioned intelligent adjustment method of traffic facilities based on fiber optic display.

[0033] The present invention also provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-mentioned intelligent adjustment method for traffic facilities based on optical fiber display.

[0034] Compared with the prior art, the present invention has the following advantages and technical effects:

[0035] The present invention first obtains and preprocesses road surface data, obtains traffic flow data based on the preprocessed road surface data; obtains the change time of traffic facilities based on the traffic flow data; generates a first control instruction based on the change time of traffic facilities; then obtains a traffic signal light image data set, preprocesses the traffic signal light image data set; trains a deep learning model based on the preprocessed traffic signal light image data set to obtain a brightness optimization model; then, obtains environmental data and state data in real time, obtains an optimal brightness value based on the environmental data, the state data and the brightness optimization model; generates a second control instruction based on the optimal brightness value; finally, intelligently adjusts traffic facilities based on the first control instruction and the second control instruction. The method provided by the present invention realizes adaptive adjustment of the brightness of signal lights and electronic variable roadside signs, and realizes automatic uniform light processing of signal lights and electronic variable roadside signs; realizes direct control of traffic lights by the center, and realizes adjustable signs and stable and controllable signal lights according to the road site conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0037] Figure 1 A schematic diagram of a method flow of an embodiment of the present invention;

[0038] Figure 2This is a principle diagram of a variable no-parking sign based on optical fiber display according to an embodiment of the present invention;

[0039] Figure 3 This is a principle diagram of a variable speed limit sign based on optical fiber display according to an embodiment of the present invention;

[0040] Figure 4 This is a schematic diagram of a fiber optic display traffic signal light according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0042] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0043] Embodiment 1

[0044] like Figure 1-4 As shown, this embodiment provides a method for intelligently adjusting traffic facilities based on optical fiber display, including:

[0045] Acquire road surface data and perform preprocessing, obtain traffic flow data based on the preprocessed road surface data; obtain traffic facility change time based on the traffic flow data; generate a first control instruction based on the traffic facility change time;

[0046] In some specific embodiments, the road surface data includes the length of the traffic flow waiting for the traffic light, the direction of the vehicle travel, and the vehicle flow rate; the process of obtaining the traffic flow data includes:

[0047] Obtain surveillance videos from surveillance cameras installed near each traffic light, perform data processing on the surveillance videos, and obtain the length of the traffic flow and the driving direction of the vehicles waiting for the traffic light;

[0048] The vehicle flow rate is obtained by detecting the magnetic field changes of vehicles passing by through the ring coil installed under the road surface;

[0049] The lane to which each vehicle belongs is determined based on the surveillance video, and the direction in which each vehicle will travel is determined based on the pre-acquired driving direction landmark arrow of each lane; the traffic flow length is the length of the vehicle queue.

[0050] In some specific embodiments, the process of obtaining the change time of traffic facilities based on traffic flow data includes:

[0051] Obtain the driving directions of all vehicles waiting for traffic lights. If the length of traffic flow in any driving direction of any vehicle is not within the preset traffic flow length range and is greater than the maximum value of the preset traffic flow length range, determine whether there are other driving directions of vehicles whose traffic flow length is not within the preset traffic flow length range and is less than the minimum value of the preset traffic flow length range;

[0052] If it exists, the driving direction of the vehicle whose traffic flow length is not within the preset traffic flow length range and is greater than the maximum value of the preset traffic flow length range is taken as the first driving direction; the driving direction of the vehicle whose traffic flow length is not within the preset traffic flow length range and is less than the minimum value of the preset traffic flow length range is taken as the second driving direction; wherein the first driving direction and the second driving direction are not opposite directions;

[0053] Construct a traffic light change time rule; and obtain the traffic light change time corresponding to the first driving direction and the second driving direction based on the traffic light change time rule.

[0054] Specifically, according to the change in the time of controlling the traffic lights based on the length of the traffic flow waiting for the traffic lights and the direction of vehicle travel, during the peak hours of urban traffic, there is often traffic congestion in a certain direction of travel, but not many vehicles in the other direction. The other direction of travel here does not include the direction of travel opposite to the direction of traffic congestion, resulting in low efficiency in the command of traffic lights. If traffic police are arranged to conduct on-site command, firstly, the time for traffic police to arrive at the scene is uncertain, resulting in the traffic congestion situation cannot be resolved in time, and secondly, there are many congested traffic intersections in the city, and arranging traffic police for on-site command at each traffic intersection requires a lot of human resources.

[0055] Exemplarily, it is determined whether the traffic flow length of waiting for traffic lights in any vehicle driving direction in the vehicle data is within a preset traffic flow length range, and the preset traffic flow length range is preferably a normal traffic flow length range, preferably 30m-90m; if there is any vehicle driving direction whose traffic flow length waiting for traffic lights is not within the preset traffic flow length range and is greater than the maximum value of the preset traffic flow length range, it is determined whether there are other vehicle driving directions in the vehicle data whose traffic flow length waiting for traffic lights is not within the preset traffic flow length range and is less than the minimum value of the preset traffic flow length range; the other vehicle driving directions do not include the driving direction opposite to the vehicle driving direction whose traffic flow length waiting for traffic lights is not within the preset traffic flow length range and is greater than the maximum value of the preset traffic flow length range; if there is any vehicle driving direction whose traffic flow length waiting for traffic lights is not within the preset traffic flow length range and is less than the minimum value of the preset traffic flow length range, the waiting direction of the vehicle driving direction is The vehicle driving direction in which the traffic flow length of the traffic light is not within the preset traffic flow length range and is greater than the maximum value of the preset traffic flow length range is taken as the first driving direction; the vehicle driving direction in which the traffic flow length waiting for the traffic light in the vehicle driving direction is not within the preset traffic flow length range and is less than the minimum value of the preset traffic flow length range is taken as the second driving direction; the second driving direction cannot be a driving direction opposite to the first driving direction; according to the first driving direction and the second driving direction, the traffic light change time corresponding to the first driving direction and the second driving direction is screened out from the corresponding rules of the traffic light change time to control the time change of the traffic light; that is, the passing time of the first driving direction is increased and the passing time of the second driving direction is reduced; for example, at a traffic intersection, the first driving direction is forward driving at the current intersection, and the second driving direction is forward driving at the left intersection, then the passing time of the forward driving at the left intersection is reduced, and the passing time of the forward driving at the current intersection is increased.

[0056] Obtain a traffic light image dataset and preprocess the traffic light image dataset; train a deep learning model based on the preprocessed traffic light image dataset to obtain a brightness optimization model;

[0057] In some specific embodiments, the process of obtaining a traffic light image dataset and preprocessing the traffic light image dataset includes:

[0058] Obtain a public traffic light image dataset; collect traffic light images under different weather conditions and light intensities on site to obtain a real-time dataset; standardize the real-time dataset and remove the directional information of the image; annotate the processed real-time dataset, perform data enhancement on the annotated dataset, and then combine it with the public traffic light image dataset to obtain a training dataset; wherein the annotated information includes weather conditions, ambient light intensity, and color status.

[0059] In some specific embodiments, the process of training a deep learning model based on the preprocessed traffic light image dataset to obtain a brightness optimization model includes:

[0060] Two deep models, yolov8x and yolov8n, are selected, and the training data sets are used to train yolov8x and yolov8n respectively; the output of yolov8x is introduced when training yolov8n, and the distillation loss and student loss are calculated based on the preset distillation parameters; yolov8n is optimized based on the distillation loss and student loss.

[0061] Acquire environmental data and status data in real time, and obtain the optimal brightness value based on the environmental data, status data and brightness optimization model; generate a second control instruction based on the optimal brightness value; wherein the environmental data includes weather conditions and ambient light intensity; and the status data is color status.

[0062] Intelligently adjust the traffic facilities based on the first control instruction and the second control instruction.

[0063] In some specific embodiments, the process of intelligently adjusting the traffic facilities based on the second control instruction includes:

[0064] The light source configuration is obtained based on the second control instruction; the current display image of the traffic facility is obtained, and the current display image is converted based on the second control instruction to obtain an HSV image that meets the requirements of the second control instruction; the HSV image is converted into an RGB image after correction and homogenization, and is transmitted to the display surface using fiber optic display technology.

[0065] Specifically, the RGB image composed of the pixels displayed by the optical fiber is converted into an HSV image. The conversion steps are as follows:

[0066] Normalize RGB values: If the RGB values ​​are given as integers (such as in the range 0-255), first convert them to floating point numbers between 0 and 1. This is because HSV calculations are usually based on values ​​in the range 0 to 1.

[0067] Calculate the maximum and minimum values: Find the maximum and minimum values ​​of the three channels, as well as the channel to which the maximum value corresponds.

[0068] Calculate Value: Value (V) is the value of the brightest color component.

[0069] Calculate Saturation: Saturation (S) reflects the purity of the color. If the maximum and minimum values ​​are the same (i.e. gray), the saturation is 0; otherwise, the saturation is calculated based on the maximum value.

[0070] Calculate Hue: Hue (H) indicates the type of color. It depends on which color channel has the largest value, and needs to calculate the relationship between the other two channels and the maximum value.

[0071] Specifically, the process of achieving uniform brightness of traffic light display includes:

[0072] Data collection to obtain the optimal brightness value: Collect data such as ambient light intensity, traffic flow and weather conditions through photosensitive sensors, traffic flow monitoring equipment and weather sensors.

[0073] Model calculation: Use a pre-trained model (such as a model based on BP neural network) to calculate the optimal brightness value in the current environment.

[0074] Command conversion: Convert the calculated optimal brightness value into specific control instructions, which may include PWM (pulse width modulation) settings or other dimming methods.

[0075] Control signal sending: Send control instructions to the control system of the signal light to adjust the brightness.

[0076] Light source configuration: Ensure that the light source configuration of the signal light is reasonable, such as using multiple light sources and fiber optic arrays to achieve more uniform light distribution.

[0077] Fiber optic display technology: Utilizing fiber optic display technology, the light from the light source is transmitted to the display surface of the signal light through optical fiber to ensure even distribution of light.

[0078] RGB to HSV conversion: Convert the display content of the signal light from RGB color space to HSV color space for better adjustment of brightness (brightness).

[0079] Gamma Correction: Perform gamma correction on the brightness value to adjust the brightness and contrast of the image to make it consistent with the visual characteristics of the human eye.

[0080] Homogenization: Apply grayscale stretching and average grayscale value adjustment to local areas to eliminate non-uniform lighting and color.

[0081] Real-time monitoring: Use sensors to monitor the brightness of traffic lights and changes in ambient light in real time.

[0082] Feedback mechanism: Dynamically adjust the brightness of traffic lights based on real-time monitoring data to maintain a uniform effect.

[0083] HSV to RGB conversion: Convert the rectified and homogenized HSV image back to an RGB image for display on a signal light.

[0084] Traffic light display: According to the adjusted RGB value, the light source of the traffic light is controlled to ensure uniform brightness and meet the optimal brightness value.

[0085] The display device used in this embodiment is a fiber display screen including a fiber array and an optical machine system that controls the light transmission of the optical fiber. Light is emitted from the laser light source of the optical machine (connecting many optical fibers to the white light source). The N pixel point optical fibers of the red light are twisted into a ball, and the beam is shaped by the lens. The shaped beam passes through the filter, and the light is injected into the corresponding optical fiber after passing through the filter. Then the light propagates in the optical fiber and finally emits light at the end of the optical fiber.

[0086] Each light source is connected to 50 optical fibers, which will cover every pixel point in the entire circle to emit light. In front of the light source, a lens will be added to shape the beam, and there is also a filter. The filter that allows red light to pass through is placed for red light, and the filter that allows green light to pass through is placed for green light.

[0087] The present invention adopts high-performance optical machine to connect low-loss polymer optical fiber, and realizes high-performance light beam through one or more polymer optical fibers and matching optical lenses to replace the original LED solution. Compared with the existing LED solution, it can greatly improve the optical performance, realize flexible and enhanced display, and the display content can be flexibly adjusted. It can also reduce the impact of lightning on traffic signs and signal lights, ensure the orderly flow of traffic, and reduce the occurrence of traffic accidents.

[0088] When the remote control command is sent to the cleaning power control box, the power control box controls the power supply of different light sources in the optical machine by powering on or off. There is a lens in front of a beam of light to collimate the light beam. The collimated light is shot into the optical fiber. The "O" and the "\" in the circle in the "No Long-term Parking of Vehicles" sign are displayed through two optical fibers. Both of these optical fibers pass through light source 1. The display fiber of the " / " in the "No Temporary or Long-term Parking of Vehicles" sign is a bundle of optical fibers connected to light source 2. Each circle is an optical fiber. One end of the optical fiber is connected to the display board to display the pattern, and the other end is tied into a bundle and connected to the light source to receive light. When the "No Long-term Parking of Vehicles" sign is on, light source 1 is powered on to work. When the "No Temporary or Long-term Parking of Vehicles" sign is working, both light source 1 and light source 2 are powered on to light up. There is also a lens at the end of the optical fiber on the sign display panel to shape the beam so that the light is emitted just right. Such as Figure 2 shown.

[0089] The operating principle of the variable speed limit sign based on fiber optic display is as follows Figure 3 As shown, a bundle of optical fibers displaying the number "4" passes through light source 1, a bundle of optical fibers displaying the number "6" passes through light source 2; a bundle of optical fibers displaying the number "8" passes through light source 3; and a bundle of optical fibers displaying the number "0" passes through light source 4. Different light sources are energized according to different speed limit requirements.

[0090] The operating principle of traffic lights based on fiber optic display is as follows Figure 4 As shown, a bundle of optical fibers showing a red light is connected to light source 1; a bundle of optical fibers showing a yellow light is connected to light source 2; and a bundle of optical fibers showing a green light is connected to light source 3. Different light sources are energized according to different lighting requirements.

[0091] The present invention also provides a traffic facility intelligent adjustment system based on optical fiber display, comprising:

[0092] A change time control instruction generation module is used to obtain and pre-process road surface data, obtain traffic flow data based on the pre-processed road surface data; obtain the traffic facility change time based on the traffic flow data; and generate a first control instruction based on the traffic facility change time;

[0093] A data sorting module is used to obtain a traffic light image dataset and pre-process the traffic light image dataset;

[0094] A model building module is used to train a deep learning model based on the preprocessed traffic light image dataset to obtain a brightness optimization model;

[0095] The brightness control instruction generation module acquires environmental data and status data in real time, obtains the optimal brightness value based on the environmental data, status data and the brightness optimization model; and generates a second control instruction based on the optimal brightness value;

[0096] The integrated control module is used to intelligently adjust the traffic facilities based on the first control instruction and the second control instruction.

[0097] The present invention also provides a computer device, which includes a memory and a processor connected to the memory; the memory is used to store computer programs; the processor is used to run the computer programs stored in the memory to execute the steps of the above-mentioned intelligent adjustment method of traffic facilities based on optical fiber display.

[0098] The present invention also provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-mentioned intelligent adjustment method for traffic facilities based on optical fiber display.

[0099] Existing LED road traffic lights are prone to unstable power supply. For example, the passing of large trucks will affect the working stability of the powered LED traffic lights. In severe weather conditions (such as fog, heavy rain, etc.), the visibility of LED road traffic lights may be affected, which in turn affects the driver's judgment. Traditional signal control systems are not intelligent enough and cannot automatically adjust the frequency and duration of signal light changes according to actual traffic conditions, making it difficult to adapt to changes in traffic demand in sudden situations. The maintenance of facilities requires regular inspection and replacement of damaged parts, which not only increases urban management costs, but may also cause failures due to untimely maintenance.

[0100] At present, the roadside signs are all painted or have fixed LED lights, which cannot make the display signs adjustable, cannot be integrated with modern intelligent transportation, and cannot communicate. When the road conditions change, the road signs need to be replaced in time, which consumes a lot of manpower, material resources and financial resources; some important traffic information may not be fully displayed, such as unclear directions to a destination, or lack of key safety tips. With the continuous development of the city, new roads are opened or old roads are renovated, and the original traffic signs may not be updated in time, misleading drivers; at night or in bad weather conditions, the reflective effect of some traffic signs is not good, and it is difficult for drivers to clearly identify them; some old or damaged signs are not repaired or replaced in time, which may affect the effectiveness of their information.

[0101] Through the intelligent adjustment method of traffic facilities based on optical fiber display provided by the present invention, adaptive adjustment of the brightness of signal lights and electronic variable roadside signs is achieved, and automatic uniform light processing of signal lights and electronic variable roadside signs is achieved; direct control of traffic lights by the center is achieved, and the signs can be adjusted according to the road site conditions, and the signal lights are stable and controllable.

[0102] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for intelligently adjusting traffic facilities based on optical fiber display, characterized in that: The following steps are involved: Acquire and preprocess road surface data, and obtain traffic flow data based on the preprocessed road surface data; Obtaining a traffic facility change time based on the traffic flow data; generating a first control instruction based on the traffic facility change time; Acquire a traffic light image dataset, and preprocess the traffic light image dataset; The deep learning model is trained based on the preprocessed traffic light image dataset to obtain a brightness optimization model; Acquire environmental data and status data in real time, obtain an optimal brightness value based on the environmental data, the status data and the brightness optimization model; and generate a second control instruction based on the optimal brightness value; Intelligently adjust the traffic facilities based on the first control instruction and the second control instruction.

2. The intelligent adjustment method of traffic facilities based on optical fiber display according to claim 1 is characterized in that: Road surface data includes the length of traffic waiting for traffic lights, the direction of vehicle travel, and vehicle flow; The process of obtaining traffic flow data includes: Obtaining surveillance videos from surveillance cameras installed near each traffic light, performing data processing on the surveillance videos, and obtaining the length of the traffic flow and the driving direction of the vehicles waiting for the traffic light; The vehicle flow rate is obtained by detecting the magnetic field changes of vehicles passing by through the ring coil installed under the road surface; The lane to which each vehicle belongs is determined based on the monitoring video, and the direction in which each vehicle will travel is determined based on the pre-acquired driving direction landmark arrow of each lane; the traffic flow length is the length of the vehicle queue.

3. The intelligent adjustment method of traffic facilities based on optical fiber display according to claim 2 is characterized in that: The process of obtaining the change time of traffic facilities based on the traffic flow data includes: Obtain the driving directions of all vehicles waiting for the traffic light, and if there is any vehicle driving direction whose traffic length is not within the preset traffic length range and is greater than the maximum value of the preset traffic length range, determine whether there is any other vehicle driving direction whose traffic length is not within the preset traffic length range and is less than the minimum value of the preset traffic length range; If it exists, the driving direction of the vehicle whose traffic flow length is not within the preset traffic flow length range and is greater than the maximum value of the preset traffic flow length range is taken as the first driving direction; the driving direction of the vehicle whose traffic flow length is not within the preset traffic flow length range and is less than the minimum value of the preset traffic flow length range is taken as the second driving direction; wherein, the first driving direction and the second driving direction are not opposite directions; Constructing a traffic light change time rule; and obtaining the traffic light change time corresponding to the first driving direction and the second driving direction based on the traffic light change time rule.

4. The method and system for intelligently adjusting traffic facilities based on optical fiber display according to claim 1, characterized in that: A traffic light image dataset is obtained, and a process of preprocessing the traffic light image dataset includes: A public traffic light image dataset is obtained; traffic light images under different weather conditions and light intensities are collected on site to obtain a real-time dataset; the real-time dataset is standardized and the directional information of the images is cleared; the processed real-time dataset is annotated, and after data enhancement is performed on the annotated dataset, the annotated dataset is combined with the public traffic light image dataset to obtain a training dataset; wherein the annotated information includes weather conditions, ambient light intensity, and color status.

5. The intelligent adjustment method of traffic facilities based on optical fiber display according to claim 1 is characterized in that: The process of training the deep learning model based on the preprocessed traffic light image dataset to obtain the brightness optimization model includes: Two deep models, yolov8x and yolov8n, are selected, and the training data set is used to train yolov8x and yolov8n respectively; when training yolov8n, the output of yolov8x is introduced, and the distillation loss and the student loss are calculated in combination with the preset distillation parameters; yolov8n is optimized based on the distillation loss and the student loss.

6. The intelligent adjustment method of traffic facilities based on optical fiber display according to claim 1 is characterized in that: The environmental data includes weather conditions and ambient light intensity; the status data is color status.

7. The intelligent adjustment method of traffic facilities based on optical fiber display according to claim 1 is characterized in that: The process of intelligently adjusting the traffic facilities based on the second control instruction includes: The light source configuration is obtained based on the second control instruction; the current display image of the traffic facility is obtained, and the current display image is converted based on the second control instruction to obtain an HSV image that meets the requirements of the second control instruction; the HSV image is corrected and homogenized, and then converted into an RGB image, which is transmitted to the display surface using fiber optic display technology.

8. An intelligent traffic facility adjustment system based on optical fiber display, characterized in that: include: A change time control instruction generation module is used to obtain and pre-process road surface data, and obtain traffic flow data based on the pre-processed road surface data; Obtaining a traffic facility change time based on the traffic flow data; generating a first control instruction based on the traffic facility change time; A data sorting module, used to obtain a traffic light image dataset and pre-process the traffic light image dataset; A model building module is used to train a deep learning model based on the preprocessed traffic light image dataset to obtain a brightness optimization model; The brightness control instruction generation module acquires environment data and status data in real time, obtains an optimal brightness value based on the environment data, the status data and the brightness optimization model; and generates a second control instruction based on the optimal brightness value; The integrated control module is used to intelligently adjust the traffic facilities based on the first control instruction and the second control instruction.

9. A computer device, characterized in that: The computer device includes a memory and a processor connected to the memory; the memory is used to store computer programs; the processor is used to run the computer programs stored in the memory to execute the steps of the intelligent adjustment method of traffic facilities based on fiber optic display as described in any one of claims 1-8.

10. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the intelligent adjustment method of traffic facilities based on optical fiber display as described in any one of claims 1 to 8 are implemented.

Images