Visual recognition-based tunnel entrance lighting system
The tunnel entrance lighting system based on vision recognition solves the problem of poor adaptability of tunnel entrance lighting systems to dynamic environmental changes, realizes fine control of curve obstruction and real-time vehicle conditions, and improves driving safety and the utilization efficiency of lighting resources in the tunnel entrance section.
Patent Information
- Application Number
- CN202611576378.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-13
- Publication Date
- 2026-11-17
AI Technical Summary
Existing supplemental lighting systems at tunnel entrances are ill-equipped to cope with rapid changes in weather conditions, tunnel curves that obstruct light, and fluctuations in real-time traffic flow, resulting in poor lighting performance or energy waste.
A visual recognition-based supplementary lighting system is adopted for the tunnel entrance section. The feature analysis module assesses the amount and trend of environmental fluctuations, the tunnel impact module analyzes the impact of light shading on the curve section, the vehicle condition analysis module assesses the stability of vehicle conditions, and the supplementary lighting adjustment module performs fine control, including identifying densely trafficked sections and adjusting the supplementary lighting intensity.
It improves the adaptability of the supplementary lighting system to dynamic traffic environments, optimizes the utilization efficiency of supplementary lighting resources, reduces the visual risks to drivers caused by sudden environmental changes, and improves driving safety and the intelligence level of the supplementary lighting system at tunnel entrances.
Smart Images

Figure CN123124169A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of supplemental lighting at tunnel entrances, and more particularly to a supplemental lighting system for tunnel entrances based on visual recognition. Background Technology
[0002] The tunnel entrance section is a critical transition area from the high-brightness external environment to the low-brightness internal environment of the tunnel, where the driver's visual adaptation ability is affected. If the lighting at the entrance section is insufficient or the lighting strategy is inappropriate, it can easily trigger a black hole effect, causing the driver to be unable to clearly identify the road conditions, vehicles, or obstacles ahead the moment they enter the tunnel, thereby increasing the risk of traffic accidents such as rear-end collisions and side collisions.
[0003] Therefore, there is an urgent need for a tunnel entrance lighting system that can improve the dynamic traffic environment, tunnel geometry, and real-time vehicle condition adaptability.
[0004] For example, Chinese Patent Publication No. CN116963343A discloses a lighting control system for the entrance section of a highway tunnel to mitigate the "black hole effect." It includes a data acquisition module, a control center module, and a load circuit. The data acquisition module includes a visibility sensor outside the tunnel, a temperature sensor, a meteorological detection device, and a color temperature sensor inside the tunnel. The visibility sensor detects the external light intensity at the tunnel entrance, the temperature sensor detects the external temperature, the meteorological detection device detects the external weather conditions, and the color temperature sensor detects the color temperature inside the tunnel entrance section. Based on the real-time detection results of the external light environment, the data acquisition module transmits the collected information to the control center. This scheme uses color temperature as a design indicator for tunnel lighting, adjusting the color temperature of the tunnel entrance section lighting to adapt to changes in visibility outside the tunnel at different times and seasons, thus improving the driving environment and enhancing the safety and comfort of driving in the tunnel.
[0005] However, the following problems still exist in the existing technology. Most of the control strategies that use timed control or are based on single light intensity feedback are difficult to cope with dynamic factors such as rapid changes in weather conditions, tunnel curves that block light, and real-time fluctuations in traffic flow, resulting in poor supplementary lighting or energy waste. Summary of the Invention
[0006] To address these issues, the present invention provides a tunnel entrance section supplementary lighting system based on visual recognition, which overcomes the problems of existing supplementary lighting systems, such as poor adaptability to dynamic environmental changes, neglect of the impact of tunnel curves, disconnection between supplementary lighting strategies and real-time vehicle conditions, and inability to perform fine-grained control for localized dense traffic flows.
[0007] To achieve the above objectives, the present invention provides a tunnel entrance section supplementary lighting system based on visual recognition, comprising: The feature analysis module is used to collect driving environment data of the target tunnel, extract the environmental state characteristics of the corresponding entrance section, evaluate the amount of environmental fluctuation, analyze the trend of environmental fluctuation, and determine whether to adjust the supplementary lighting response speed. The tunnel impact module is used to analyze whether the curved section of the target tunnel has an impact on vehicle driving light obstruction, so as to determine whether to adjust the supplementary lighting distance. The vehicle condition analysis module is used to extract driving status characteristics within a predetermined road area, assess vehicle condition stability values, and classify supplementary lighting needs into categories. A supplementary lighting adjustment module, used to optimize the supplementary lighting of the entrance section based on the supplementary lighting requirement category, including, If the supplementary lighting requirement category is the strong supplementary lighting requirement category, then identify the densely trafficked sections, determine the supplementary lighting optimization requirement, and adjust the initial supplementary lighting intensity; If the supplemental lighting requirement category is weak supplemental lighting requirement category, then maintain the initial supplemental lighting intensity.
[0008] Furthermore, the feature analysis module is used to assess the amount of environmental fluctuation, including: Obtain the environmental state characteristics corresponding to several sampling times, and solve for the corresponding environmental fluctuations. The environmental fluctuation amount is determined based on the statistical variance of the environmental state characteristics; The environmental characteristics include the difference in light intensity inside and outside the entrance section and the meteorological optical visibility.
[0009] Furthermore, the feature analysis module is used to analyze environmental fluctuation trends and determine whether to adjust the supplementary lighting response speed, including: If the amount of environmental fluctuation is greater than or equal to the preset amount of environmental fluctuation, the environmental fluctuation trend will be identified as a high environmental fluctuation trend. If the environmental fluctuation trend is one of high fluctuation, then the supplemental lighting response speed will be adjusted.
[0010] Furthermore, the tunnel impact module is used to analyze whether the curved sections of the target tunnel have an impact on vehicle traffic due to light shading, including: If the distance between the curved section and the entrance section of the target tunnel is less than the preset distance between the curved section and the entrance section, and the radius of curvature of the curved section is less than the preset radius of curvature, then it is determined that the curved section has a light shading effect on vehicle driving.
[0011] Furthermore, the tunnel impact module is used to determine whether to adjust the supplementary lighting distance, including: If the curved section obstructs the light for vehicle travel, adjust the distance of the supplemental lighting.
[0012] Furthermore, the vehicle condition analysis module is used to evaluate the vehicle condition stability value, including: The vehicle stability value is determined based on driving condition characteristics; Calculate the average vehicle speed ratio and the dispersion ratio respectively; The sum of the average vehicle speed ratio adjusted according to the first contribution ratio and the dispersion ratio adjusted according to the second contribution ratio is determined as the vehicle condition stability value. The driving state characteristics include the dispersion of average vehicle speed and vehicle spacing.
[0013] Furthermore, the vehicle condition analysis module is used to classify supplementary lighting needs into categories, including: If the vehicle condition stability value is greater than or equal to the preset vehicle condition stability value, it is classified as a strong supplementary lighting requirement category. If the vehicle condition stability value is less than the preset vehicle condition stability value, it is classified as a weak supplementary lighting requirement category.
[0014] Furthermore, the supplementary lighting adjustment module is used to identify densely trafficked areas, including: The predetermined road area is divided into several path segments, and the corresponding road traffic condition images are retrieved. Mark the vehicles whose driving distance within each of the aforementioned path segments is less than the preset driving distance; Count the number of marked vehicles and calculate the ratio of the number of marked vehicles to the total number of vehicles; If the quantity ratio is greater than or equal to a preset quantity ratio, the corresponding path segment is identified as the densely trafficked segment.
[0015] Furthermore, the supplementary lighting adjustment module is used to determine the degree of supplementary lighting optimization requirement, including: The supplementary lighting optimization requirement is determined based on the deviation of the ratio of the light intensity in front of the vehicle to the light intensity behind the vehicle in the densely trafficked section, as well as the distance between the densely trafficked section and the entrance section. Calculate the deviation ratio and distance ratio separately; The sum of the deviation ratio adjusted according to the first influence ratio and the distance ratio adjusted according to the second influence ratio is determined as the supplementary lighting optimization requirement.
[0016] Furthermore, the supplementary lighting adjustment module is used to adjust the initial supplementary lighting intensity, including: The adjustment amount of the initial supplemental lighting intensity is positively correlated with the degree of supplemental lighting optimization requirement.
[0017] Compared with existing technologies, this invention incorporates a feature analysis module, a tunnel impact module, a vehicle condition analysis module, and a supplementary lighting adjustment module. By collecting driving environment data from the target tunnel, it extracts environmental state characteristics of the corresponding entrance section, assesses environmental fluctuations, analyzes environmental fluctuation trends, and determines whether to adjust the supplementary lighting response speed. It also analyzes whether the curves of the target tunnel obstruct vehicle illumination, determining whether to adjust the supplementary lighting range. Furthermore, it extracts driving state characteristics within a predetermined road area, assesses vehicle stability values, classifies supplementary lighting needs, and adaptively optimizes the supplementary lighting at the entrance section. This invention enhances the adaptability of the supplementary lighting system to dynamic traffic environments, improving driving safety at tunnel entrances while optimizing the utilization efficiency of supplementary lighting resources.
[0018] In particular, this invention analyzes the severity of environmental fluctuations using statistical methods, thereby adjusting the supplementary lighting response speed in advance to achieve proactive adaptive control based on dynamic environmental changes. On one hand, it quantifies the amplitude of brightness jumps between the tunnel's interior and exterior using the difference in light intensity between the entrance section and the exterior. A greater difference in light intensity requires a longer visual adaptation time for the driver, necessitating higher compensation brightness from the supplementary lighting system. Rapid fluctuations in this characteristic indicate an unstable environment, easily leading to visual fatigue or even temporary blindness. On the other hand, it quantifies atmospheric visibility using meteorological optical visibility, a feature directly affecting the driver's visual perception distance of the road ahead. Rapid fluctuations in this feature indicate unstable visibility conditions, requiring faster supplementary lighting adjustment to prevent momentary driver blindness. This invention calculates the statistical variance of these two features from two dimensions: light contrast and atmospheric transmission characteristics, comprehensively quantifying the visual environment quality and dynamic changes at the tunnel entrance section, providing data support for subsequent decisions on adjusting the supplementary lighting response speed. Through this mechanism, the supplementary lighting system's adaptability to sudden changes in light intensity can be improved, effectively reducing the visual risks to drivers from environmental changes.
[0019] In particular, this invention constructs an active supplementary lighting compensation mechanism based on tunnel alignment, introducing a joint determination of spatial distance and curvature of tunnel curves to achieve quantification and adaptive control of light path obstruction. On one hand, the spatial proximity between the curve and entrance sections of the target tunnel is quantified by the distance between the curve and entrance sections. In reality, the closer the curve is to the entrance, the greater the probability that the driver will face the curve immediately upon entering the tunnel. At this time, the supplementary lighting at the entrance section may not have fully covered the curve area, and the curvature of the curve will prevent light from propagating in a straight line along the tunnel axis to the inside of the curve. Therefore, the smaller the distance, the earlier the curve's blocking effect on the entrance section's light occurs, and the shorter the driver's visual preparation time before entering the curve. On the other hand, the curvature radius of the curve section is used to quantify the degree of curvature of the curve and the driver's visual extensibility. The smaller the curvature radius, the more abrupt the change in the tunnel axis direction, and the easier it is for light to be blocked by the tunnel walls on the outside or inside of the curve when propagating in a straight line. Simultaneously, the driver's visible path inside the curve at the beginning of the curve is shorter, creating a blind spot, requiring supplementary lighting to illuminate the inner wall of the curve and the road surface. Based on this, the present invention comprehensively quantifies the actual degree of obstruction of the tunnel geometry to the supplementary lighting effect from two dimensions: spatial location and geometric shape, to achieve structurally adaptive adjustment of the supplementary lighting distance. The supplementary lighting compensation mechanism constructed by the present invention accurately eliminates visual risks in the curve area without increasing energy waste, improves the path recognition ability of vehicles before the curve, and reduces the risk of side collisions or lane departures.
[0020] In particular, this invention employs vehicle-road cooperative control logic to match the supplementary lighting strategy with the actual safety risk level of traffic flow. On one hand, it quantifies the overall speed level of the traffic flow and its deviation from the reference speed by using average vehicle speed. The greater the deviation of the average vehicle speed from the reference speed, the worse the traffic flow stability, the higher the potential risk, and the stronger the supplementary lighting required to help drivers detect obstacles or curves ahead in advance. On the other hand, it quantifies the uniformity of vehicle spacing by using the dispersion of vehicle spacing. A greater dispersion indicates uneven vehicle spacing, which can easily lead to high-risk scenarios such as sudden braking by the vehicle in front and following too closely. In such cases, the traffic flow is unstable and prone to chain-reaction accidents, such as rear-end collisions. Simultaneously, drivers' field of vision and reaction time vary significantly, resulting in uneven reliance on supplementary lighting. Based on this, this invention assesses the stability of vehicle conditions from two dimensions: macroscopic stability and microscopic orderliness. It comprehensively quantifies the driving stability and potential safety risks within the predetermined road area at the tunnel entrance, providing data support for subsequent classification of supplementary lighting needs. Therefore, under unstable driving conditions, increasing the intensity of supplemental lighting improves the ability of drivers behind to recognize the position and brake lights of the vehicle in front, shortens reaction time, and thus reduces the probability of rear-end collisions and chain collisions; under stable driving conditions, excessive supplemental lighting is avoided, reducing energy consumption and light pollution, thereby achieving supplemental lighting on demand, taking into account both safety and energy conservation.
[0021] In particular, this invention further identifies densely trafficked sections and adjusts the intensity of supplemental lighting to address the specific needs of vehicles requiring strong supplemental lighting. This forms a complete chain from macroscopic vehicle stability to fine-tuned control of localized densely trafficked sections, avoiding the blind enhancement of overall lighting. On one hand, by measuring the deviation of the ratio of the light intensity in front of and behind vehicles in densely trafficked sections, the invention quantifies the severity of localized lighting differences caused by factors such as vehicle obstruction, uneven tunnel lighting layout, and differences in headlight brightness relative to the normal allowable range, as well as whether the unevenness of lighting in the driver's line of sight exceeds an acceptable level. For example, if a vehicle behind is obstructed by the vehicle in front, or the brake lights of the vehicle in front are not clearly visible; or if the front is too bright and the rear is too dark, the deviation will increase, leading to a greater severity of localized visual contrast and increasing the risk of rear-end collisions or glare. In such cases, supplemental lighting needs to be enhanced to improve the visibility of the vehicle in front to the following vehicle and improve localized visual contrast. On the other hand, by measuring the distance between the densely trafficked section and the tunnel entrance section, the invention quantifies the urgency of the risks posed by the spatial location of the densely trafficked section and the tunnel entrance, as well as the visual adaptation pressure on the driver. As a critical area of alternating light and shadow, the entrance section is particularly sensitive to changes in lighting, especially for densely trafficked areas where drivers, having just entered the tunnel, are still visually adapting. Therefore, this invention quantifies the need for supplemental lighting optimization from two dimensions: the degree of localized lighting imbalance and the urgency of spatial risks. This determines the urgency and extent to which additional supplemental lighting intensity needs to be increased at the entrance section, providing a reliable basis for subsequent adjustments to the initial supplemental lighting intensity. This invention achieves spatially differentiated, risk-quantified, and precise energy-saving supplemental lighting optimization at tunnel entrances while ensuring driving safety, especially in scenarios with dense following traffic, thus enhancing the intelligence level of the supplemental lighting system. Attached Figure Description
[0022] Figure 1 A functional block diagram of a tunnel entrance section supplementary lighting system based on visual recognition, as shown in an embodiment of the invention; Figure 2 A logic block diagram illustrating the analysis of environmental fluctuation trends in an embodiment of the invention; Figure 3 This is a logic block diagram for identifying densely trafficked sections according to an embodiment of the invention; Figure 4 This is a logic block diagram illustrating the optimization of supplemental lighting in the entrance section according to an embodiment of the invention. Detailed Implementation
[0023] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.
[0024] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.
[0025] Please see Figure 1 The diagram shown is a functional block diagram of a tunnel entrance section supplementary lighting system based on visual recognition according to an embodiment of the present invention. The tunnel entrance section supplementary lighting system based on visual recognition according to an embodiment of the present invention includes: The feature analysis module is used to collect driving environment data of the target tunnel, extract the environmental state characteristics of the corresponding entrance section, evaluate the amount of environmental fluctuation, analyze the trend of environmental fluctuation, and determine whether to adjust the supplementary lighting response speed. The tunnel impact module, which is connected to the feature analysis module, is used to analyze whether the curved section of the target tunnel has an impact on vehicle driving due to light shading, so as to determine whether to adjust the supplementary lighting distance. The vehicle condition analysis module, which is connected to the tunnel impact module, is used to extract driving status characteristics within a predetermined road range, assess vehicle condition stability values, and classify supplementary lighting needs into categories. A supplementary lighting adjustment module, connected to the vehicle condition analysis module, is used to optimize the supplementary lighting of the entrance section based on the supplementary lighting demand category, including: If the supplementary lighting requirement category is the strong supplementary lighting requirement category, then identify the densely trafficked sections, determine the supplementary lighting optimization requirement, and adjust the initial supplementary lighting intensity; If the supplemental lighting requirement category is weak supplemental lighting requirement category, then maintain the initial supplemental lighting intensity.
[0026] Specifically, the driving environment data includes environmental state characteristics, the distance between the curve section and the entrance section of the target tunnel, the radius of curvature of the curve section, driving state characteristics, the distance between vehicles within the path section, the deviation of the ratio of the light intensity in front of the vehicle to the light intensity behind the vehicle, and the distance between the densely trafficked section and the entrance section.
[0027] Specifically, there are no specific limitations on the method of collecting environmental state characteristics. It can be done by setting up an external brightness detector outside the tunnel to measure the intensity of external natural light; and by setting up multiple internal illuminance detection nodes along the driving direction inside the tunnel entrance section, using an anti-interference photoelectric sensor array to collect illuminance data, thereby determining the difference in light intensity inside and outside the entrance section. Meteorological optical visibility can be calculated using Beer-Lambert's law by deploying transmission-type visibility detectors or scattering-type visibility sensors near the tunnel entrance.
[0028] Specifically, the distance from the entrance section to the curve section and the radius of curvature of the curve section can be obtained from the tunnel engineering as-built drawings, tunnel structure BIM model or high-precision tunnel design drawings.
[0029] The distance between the curve segment and the entrance segment refers to the spatial arc length from the tunnel entrance along the centerline of the driving direction to the starting point of the first curve segment.
[0030] Specifically, there are no specific limitations on the method of collecting and acquiring driving status features. Road vehicle condition images can be collected by high-definition cameras, and vehicles in the image can be identified based on deep learning target detection algorithms. The speed of the target vehicle can be estimated by the pixel displacement change of the vehicle position between frames, and then the average vehicle speed can be obtained. By continuously sensing the longitudinal distance between vehicles in front and behind within a cross-section, and relying on millimeter-wave radar and lidar to output the distance values between vehicles in front and behind, the spatial distance is converted into the inter-vehicle distance. The standard deviation of the vehicle distance within a predetermined distance range is calculated to obtain the dispersion of the driving distance.
[0031] Specifically, the method for acquiring the deviation of the ratio of the light intensity in front of the vehicle to the light intensity behind the vehicle is not specifically limited. It can rely on traffic monitoring cameras deployed on both sides of the road outside the tunnel. By analyzing the pixel brightness of the corresponding areas in front of and behind the vehicle in the image, and based on the mapping relationship between pixel grayscale values and light intensity established through on-site calibration of the cameras using a standard illuminance meter, the actual light intensity can be obtained. Then, the ratio of the light intensity in front of the vehicle to the light intensity behind the vehicle can be calculated. By determining the relative deviation of this ratio—that is, the absolute value of the difference between this ratio and a reference ratio—the ratio of the absolute value of the difference to the reference ratio is obtained, thus yielding the deviation of the ratio of the light intensity in front of the vehicle to the light intensity behind the vehicle.
[0032] In this embodiment, the reference ratio is used to characterize the ideal ratio of the light intensity in front of the vehicle to the light intensity behind the vehicle in a densely trafficked road section. Preferably, it is set to 1.0. When the light intensity in front of the vehicle and the light intensity behind the vehicle are equal, the visual contrast is minimized when the driver's line of sight switches between different distances, which is beneficial for quickly identifying the brake lights and driving status of the vehicle in front in dense traffic scenarios, reducing the risk of rear-end collisions. Of course, those skilled in the art can also adaptively adjust the reference ratio within the range of 0.8 to 1.2 based on actual tunnel operation data (such as the statistical value of the front-to-rear light ratio under normal traffic flow).
[0033] Specifically, there are no specific limitations on the method for collecting the distance between the densely trafficked section and the entrance section. It can be based on each vehicle in the road traffic image and its corresponding longitudinal position, and automatically mapped to absolute mileage coordinates according to fixed reference objects in the image field of view, such as tunnel wall markers, mileage signs, or virtual sampling points of a pre-established high-precision tunnel map. The arc length along the centerline from the starting position of the path segment to the tunnel entrance is then obtained, which is the distance between the densely trafficked section and the entrance section.
[0034] Specifically, the predetermined road range refers to a road section of a certain length extending from the tunnel entrance to the outside of the tunnel. This range can be adjusted between 100 and 200 meters depending on the tunnel's design speed. Preferably, it is set to 150 meters. This range covers the core visual adaptation area for the driver before entering the tunnel, effectively captures high-risk vehicle conditions such as dense following and speed fluctuations, while also considering the sensor's actual detection capabilities and data processing efficiency.
[0035] It should be noted that the multiple functional modules involved in this application are only a logical division based on the functions implemented according to the present invention, and are not a strict limitation on the physical structure; in practical applications, the above functional modules can be implemented by one or more integrated circuits, a processor executing program code in memory, or a combination of the above devices.
[0036] Specifically, the feature analysis module is used to assess environmental fluctuations, including: Obtain the environmental state characteristics corresponding to several sampling times, and solve for the corresponding environmental fluctuations. The environmental fluctuation amount is determined based on the statistical variance of the environmental state characteristics; The environmental characteristics include the difference in light intensity inside and outside the entrance section and the meteorological optical visibility.
[0037] Specifically, please refer to Figure 2 As shown, this is a logic block diagram for analyzing environmental fluctuation trends according to an embodiment of the present invention. The feature analysis module is used to analyze environmental fluctuation trends and determine whether to adjust the supplementary lighting response speed, including: If the environmental fluctuation amount E is greater than or equal to the preset environmental fluctuation amount E0, the environmental fluctuation trend will be identified as a high environmental fluctuation trend. If the environmental fluctuation trend is one of high fluctuation, then the supplemental lighting response speed will be adjusted.
[0038] To determine the preset environmental fluctuation amount, the environmental fluctuation amount for each sampling period can be calculated from data of the target tunnel during several historical normal environmental periods, obtaining the mean and standard deviation of the sequence. Then, the product of the standard deviation and the sensitivity coefficient is calculated, and the sum of this product and the mean is determined as the preset environmental fluctuation amount.
[0039] The sensitivity coefficient can be selected within the range [1.5, 3.0] based on the tolerance for missed detection of environmental fluctuations, and is generally set to 2.0. If sufficient historical data is available, the 95th percentile can also be used as the preset environmental fluctuation value.
[0040] Furthermore, the preset environmental fluctuation amount should be dynamically updated periodically with seasonal or environmental changes to adapt to the drift of the environment around the tunnel. Simultaneously, it should be coordinated with sampling parameters such as light intensity difference and meteorological optical range, and iteratively optimized by verifying the misjudgment ratio of the normal window.
[0041] It is understandable that adjusting the supplemental lighting response speed refers to increasing the supplemental lighting response speed under the trend of high environmental fluctuations, and the supplemental lighting response speed is positively correlated with the amount of environmental fluctuations.
[0042] When the ambient fluctuation E is greater than or equal to the preset ambient fluctuation E0, the adjustment of the supplementary light response speed is triggered. In this embodiment, the supplementary lighting response speed is adjusted by changing the sampling and control cycle of the supplementary lighting control system. The default control cycle Tc0 is set to 500ms, and the adjusted control cycle Tc is determined as follows:
[0043] Where Tc represents the adjusted control period, Tcmin represents the minimum control period, Tc0 represents the default control period, E0 represents the preset environmental fluctuation amount, and E represents the environmental fluctuation amount. The minimum control period Tcmin is 100ms, that is, the control period is continuously shortened between 100ms and 500ms as the environmental fluctuation amount increases, thereby achieving positive correlation continuous adjustment where the more severe the environmental fluctuation, the faster the supplementary light response. At the same time, the lower limit protection ensures the stable operation of the system.
[0044] Specifically, the tunnel impact module is used to analyze whether the curved sections of the target tunnel have an impact on vehicle traffic due to light shading, including: If the distance between the curved section and the entrance section of the target tunnel is less than the preset distance between the curved section and the entrance section, and the radius of curvature of the curved section is less than the preset radius of curvature, then it is determined that the curved section has a light shading effect on vehicle driving.
[0045] It is understandable that when the starting point of a curve is within the lighting coverage area of the entrance section, the curve wall will block the supplementary lighting light from the entrance section, creating a blind spot. Therefore, the preset distance between the curve section and the entrance section should generally not exceed the length of the entrance section. In this embodiment, the preset distance between the curve section and the entrance section is set as the product of the entrance section length and the length deviation coefficient. The length deviation coefficient is selected within the range [0.3, 0.5], and is preferably 0.4 in practice. The length deviation coefficient can be adjusted according to the tunnel type and the corresponding design speed; the higher the design speed, the smaller the coefficient value, so as to trigger the supplementary lighting distance adjustment earlier.
[0046] In this embodiment, the purpose of setting a preset radius of curvature is to determine whether the tunnel alignment has reached the critical point where active supplemental lighting is required. Referring to the "Detailed Rules for Highway Tunnel Lighting Design" and the "Standards for Urban Road Lighting Design," a radius of curvature ≤ 1000m is used as the critical value for distinguishing between straight sections and curved sections requiring enhanced supplemental lighting. Furthermore, those skilled in the art can adjust the preset radius of curvature within the range of 800m to 1200m based on the tunnel's design speed and alignment characteristics.
[0047] Specifically, the tunnel impact module is used to determine whether to adjust the supplementary lighting distance, including: If the curved section obstructs the light for vehicle travel, adjust the distance of the supplemental lighting.
[0048] Understandably, when there is light obstruction in the curved section, the supplementary lighting system should increase the effective distance of the supplementary lighting at the entrance section so that the light can extend to the starting area of the curve or the inside of the curve to compensate for the insufficient lighting caused by the tunnel wall.
[0049] In this embodiment, by controlling the adjustable-angle LED lights installed at the entrance section, the vertical or horizontal projection angle of the lights is increased, causing the beam to point towards the curve and extending the effective lighting distance. The adjustment angle is related to the radius of curvature of the curve and the distance.
[0050] Where A represents the adjustment angle, A0 represents the initial projection angle, R represents the radius of curvature of the curve segment, and R0 represents the preset radius of curvature; The sharper the curve, the smaller the radius of curvature of the curve segment, the larger the adjustment angle, and the farther the effective distance.
[0051] Alternatively, while maintaining the overall lighting power at the entrance section, the luminous flux output of the lamps near the curve can be increased, allowing the light to travel further along the tunnel direction. The adjusted intensity I is related to the distance attenuation factor:
[0052] Where I0 represents the initial illuminance at the end of the entrance section, D represents the distance from the curve section to the entrance section, and H represents the length of the entrance section; The closer the curve is to the entrance, the greater the increase in intensity, to ensure that the supplementary lighting covers the curve area.
[0053] Specifically, the vehicle condition analysis module is used to evaluate the vehicle condition stability value, including: The vehicle stability value is determined based on driving condition characteristics; Calculate the average vehicle speed ratio and the dispersion ratio respectively; The sum of the average vehicle speed ratio adjusted according to the first contribution ratio and the dispersion ratio adjusted according to the second contribution ratio is determined as the vehicle condition stability value. The driving state characteristics include the dispersion of average vehicle speed and vehicle spacing.
[0054] The average vehicle speed ratio is the ratio of the average vehicle speed to the reference vehicle speed. The dispersion ratio is the ratio of the dispersion of the driving distance to the preset dispersion. Specifically, the sum of the first contribution ratio and the second contribution ratio is 1. Initially, the first and second contribution ratios can be set equally, and subsequently, they can be adaptively adjusted according to the sensitivity requirements of average vehicle speed or inter-vehicle distance divergence in actual scenarios. In addition, those skilled in the art can determine the specific values through on-site calibration or historical data fitting.
[0055] In this embodiment, the design speed corresponding to the road grade of the target tunnel is used as the reference speed. Alternatively, when the tunnel is already in operation and historical speed data is available, the average vehicle speed during normal traffic flow periods can also be collected and used as the reference speed; this will not be elaborated further.
[0056] In this embodiment, the purpose of setting a preset dispersion is to characterize a critical state where the uniformity of vehicle spacing distribution is low. By acquiring the dispersion data of vehicle spacing in the target tunnel during several historical normal traffic flow periods, the mean dispersion is calculated as a baseline value under normal conditions. Based on the purpose of setting the preset dispersion, the preset dispersion value is determined as the product of the mean dispersion and the dispersion deviation coefficient.
[0057] Since the dispersion of vehicle spacing in normal traffic flow exhibits reasonable random fluctuations, a coefficient slightly higher than the mean can effectively encompass normal variations, preventing frequent deviations in vehicle condition stability values due to minor fluctuations, thereby improving the reliability of supplementary lighting demand classification. Based on this, the dispersion deviation coefficient is selected within the interval [1.1, 1.3], with 1.2 being preferred in practice. This preferred value achieves a good balance between filtering normal fluctuations and accurately identifying abnormal discrete states. Furthermore, those skilled in the art can readjust the value interval or adjust the dispersion deviation coefficient within a given interval based on the actual traffic flow characteristics of the tunnel and the required sensitivity of the supplementary lighting response.
[0058] Please see Figure 4 As shown, it is a logic block diagram for optimizing the supplementary lighting of the entrance section according to an embodiment of the present invention.
[0059] Specifically, the vehicle condition analysis module is used to classify supplementary lighting needs into categories, including: If the vehicle condition stability value is greater than or equal to the preset vehicle condition stability value, it is classified as a strong supplementary lighting requirement category. If the vehicle condition stability value is less than the preset vehicle condition stability value, it is classified as a weak supplementary lighting requirement category.
[0060] To determine the preset vehicle condition stability value, the vehicle condition stability value for each time period can be calculated from data of the target tunnel during several historical normal traffic flow periods, obtaining the mean and standard deviation of the sequence. Then, the product of the standard deviation and the sensitivity coefficient is calculated, and the sum of this product and the mean is determined as the preset vehicle condition stability value.
[0061] The sensitivity coefficient can be selected within the range [1.5, 3.0] based on the tolerance for missed detection of high-risk vehicle conditions, and is generally set to 2.0. If sufficient historical data is available, the 95th percentile can also be used as the preset vehicle condition stability value.
[0062] Furthermore, the preset vehicle condition stability value should be dynamically updated periodically according to changes in traffic flow characteristics to adapt to drift in tunnel operation status. Simultaneously, it should maintain synergy with the contribution ratios of the average vehicle speed ratio and the dispersion ratio, and can be iteratively optimized by verifying the division ratio of normal samples.
[0063] Specifically, please refer to Figure 3 The diagram shown is a logic block diagram for identifying densely populated vehicle sections according to an embodiment of the present invention. The supplementary lighting adjustment module is used to identify densely populated vehicle sections and includes: The predetermined road area is divided into several path segments, and the corresponding road traffic condition images are retrieved. Mark the vehicles whose driving distance within each of the aforementioned path segments is less than the preset driving distance; Count the number of marked vehicles and calculate the ratio of the number of marked vehicles to the total number of vehicles; If the quantity ratio is greater than or equal to a preset quantity ratio, the corresponding path segment is identified as the densely trafficked segment.
[0064] Specifically, for each path segment, the driving distance between each vehicle and the adjacent vehicle in front is determined sequentially. If the driving distance is less than the preset driving distance, the vehicle behind is marked.
[0065] In this embodiment, the purpose of setting a preset driving distance is to characterize the lower limit of the safe distance between vehicles under normal driving conditions, and to determine whether vehicles are in a dense state of following too closely. By acquiring driving distance data of the target tunnel during several historical normal traffic flow periods, the average driving distance is calculated as the benchmark value under normal conditions. Based on the purpose of setting the preset driving distance, the preset driving distance is determined as the product of the average driving distance and the distance deviation coefficient.
[0066] Since vehicle spacing fluctuates reasonably during normal driving, a coefficient slightly higher than the average can effectively accommodate normal spacing variations, avoiding frequent misjudgments of dense traffic due to slight spacing fluctuations, thereby improving the reliability of identifying dense traffic sections. Based on this, the spacing deviation coefficient is selected within the range [1.1, 1.3], with 1.2 being preferred in practice. This preferred value achieves a good balance between filtering normal fluctuations and accurately identifying abnormally dense traffic conditions. Furthermore, those skilled in the art can readjust the value range or adjust the spacing deviation coefficient within a given range according to the actual traffic flow characteristics of the tunnel and the sensitivity requirements for dense traffic identification.
[0067] Understandably, when more than half of the vehicles are following too closely, it indicates that the traffic density on that section of road has reached a critical point of widespread congestion, and most drivers face the risk of rear-end collisions. In this situation, enhanced supplemental lighting can significantly improve driving safety. Based on this, setting the preset quantity ratio to 0.5 achieves a good balance between avoiding false triggering due to a few vehicles temporarily slowing down or changing lanes and responding promptly to the actual risk of traffic congestion. Furthermore, those skilled in the art can adjust the ratio within the range of 0.3 to 0.7 according to the actual traffic flow, design speed, and safety requirements of the tunnel.
[0068] Specifically, the supplementary lighting adjustment module is used to determine the degree of supplementary lighting optimization required, including: The supplementary lighting optimization requirement is determined based on the deviation of the ratio of the light intensity in front of the vehicle to the light intensity behind the vehicle in the densely trafficked section, as well as the distance between the densely trafficked section and the entrance section. Calculate the deviation ratio and distance ratio separately; The sum of the deviation ratio adjusted according to the first influence ratio and the distance ratio adjusted according to the second influence ratio is determined as the supplementary lighting optimization requirement.
[0069] The deviation ratio is the ratio of the ratio of the light intensity in front of the vehicle to the light intensity behind the vehicle in a densely trafficked section to the deviation from the reference ratio. The distance ratio is the ratio of a preset distance to the distance between the densely trafficked section and the entrance section.
[0070] Specifically, the sum of the first influence ratio and the second influence ratio is 1. In this embodiment, considering that the degree of uneven illumination directly affects the driver's ability to identify the vehicle in front, and the distance of the path segment from the entrance determines the urgency of visual adaptation, both have a significant impact on the demand for supplemental lighting. Based on engineering experience and theoretical analysis, focusing on compensation for uneven illumination, the initial first influence ratio is set to 0.6, and the second influence ratio is set to 0.4. In addition, those skilled in the art can adjust the first influence ratio within the range of 0.4 to 0.8 and the second influence ratio within the range of 0.2 to 0.6 based on actual tunnel operation data, such as accident statistics, traffic flow characteristics, or on-site calibration tests, while keeping the sum of the two at 1.
[0071] In this embodiment, a preset distance is used as a benchmark reference value for the distance between the densely trafficked section and the tunnel entrance. This is used to characterize situations where the spatial location between the densely trafficked section and the tunnel entrance presents a high degree of risk and urgency, resulting in significant visual adaptation pressure for the driver. By acquiring data on the distance between the densely trafficked section and the entrance section of the target tunnel during several historical normal operating periods, the average distance is calculated as the benchmark value under normal conditions. Based on the purpose of setting the preset distance, it is determined as the product of the average distance and the distance deviation coefficient.
[0072] Because the distance between dense path sections and the entrance varies reasonably under normal operating conditions, the coefficient can be adjusted near the mean to adapt to different tunnel alignments and safety requirements. Based on this, the distance deviation coefficient is selected within the range [0.8, 1.2], with 1.0 being preferred in practice. The preferred value uses the historical mean as a benchmark, effectively reflecting the distance characteristics under normal circumstances. Of course, those skilled in the art can adjust the distance deviation coefficient within a given range according to the actual risk sensitivity of the tunnel. For example, a smaller coefficient, such as 0.8, can be used for tunnels with higher risks to enhance the sensitivity of distance factors.
[0073] Specifically, the fill light adjustment module is used to adjust the initial fill light intensity, including: The adjustment amount of the initial supplemental lighting intensity is positively correlated with the degree of supplemental lighting optimization requirement.
[0074] It is understandable that when the demand for supplemental lighting optimization is higher, it indicates that the unevenness of illumination in densely trafficked sections is more severe, the section is closer to the entrance, and the driving risk is more urgent. Therefore, it is necessary to significantly increase the initial supplemental lighting intensity. When the demand for supplemental lighting optimization is lower, it indicates that the local illumination conditions are relatively balanced and the risk is lower. It is only necessary to slightly increase the initial supplemental lighting intensity or even keep it unchanged.
[0075] The initial supplemental lighting intensity can be dynamically adjusted based on actual traffic scenarios and tunnel operation needs. For example, the supplemental lighting system can reserve configurable adjustment curves or mapping tables, allowing for flexible setting of the correspondence between the supplemental lighting optimization demand and the adjustment amount based on historical accident data, traffic flow statistics, or on-site calibration tests. For instance, a linear mapping can be used, where the product of a baseline coefficient and the supplemental lighting optimization demand is used as the adjustment amount. The baseline coefficient can be determined based on factors such as historical tunnel traffic flow, accident rate, and design speed. Alternatively, a piecewise linear mapping can be used, dividing the supplemental lighting optimization demand into several intervals, each corresponding to a fixed adjustment step size. Logarithmic or exponential mappings can also be used to accommodate different sensitivity requirements.
[0076] Furthermore, to prevent excessively low or high supplemental lighting intensity from affecting driving safety, the adjusted supplemental lighting intensity is limited between a preset minimum and maximum safe intensity. The minimum safe intensity can be determined based on the brightness requirements of the entrance section in the "Detailed Rules for Lighting Design of Highway Tunnels," while the maximum safe intensity is capped at the level that does not produce glare or excessive energy waste.
[0077] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.
Claims
1. A tunnel entrance section supplementary lighting system based on visual recognition, characterized in that, include: The feature analysis module is used to collect driving environment data of the target tunnel, extract the environmental state characteristics of the corresponding entrance section, evaluate the amount of environmental fluctuation, analyze the trend of environmental fluctuation, and determine whether to adjust the supplementary lighting response speed. The tunnel impact module is used to analyze whether the curved section of the target tunnel has an impact on vehicle driving light obstruction, so as to determine whether to adjust the supplementary lighting distance. The vehicle condition analysis module is used to extract driving status characteristics within a predetermined road area, assess vehicle condition stability values, and classify supplementary lighting needs into categories. A supplementary lighting adjustment module, used to optimize the supplementary lighting of the entrance section based on the supplementary lighting requirement category, including, If the supplementary lighting requirement category is the strong supplementary lighting requirement category, then identify the densely trafficked sections, determine the supplementary lighting optimization requirement, and adjust the initial supplementary lighting intensity; If the supplemental lighting requirement category is weak supplemental lighting requirement category, then maintain the initial supplemental lighting intensity.
2. The tunnel entrance section supplementary lighting system based on visual recognition according to claim 1, characterized in that, The feature analysis module is used to assess the amount of environmental fluctuation, including: Obtain the environmental state characteristics corresponding to several sampling times, and solve for the corresponding environmental fluctuations. The environmental fluctuation amount is determined based on the statistical variance of the environmental state characteristics; The environmental characteristics include the difference in light intensity inside and outside the entrance section and the meteorological optical visibility.
3. The tunnel entrance section supplementary lighting system based on visual recognition according to claim 2, characterized in that, The feature analysis module is used to analyze environmental fluctuation trends and determine whether to adjust the supplementary lighting response speed, including: If the amount of environmental fluctuation is greater than or equal to the preset amount of environmental fluctuation, the environmental fluctuation trend will be identified as a high environmental fluctuation trend. If the environmental fluctuation trend is one of high fluctuation, then the supplemental lighting response speed will be adjusted.
4. The tunnel entrance section supplementary lighting system based on visual recognition according to claim 1, characterized in that, The tunnel impact module is used to analyze whether the curved sections of the target tunnel have an impact on vehicle traffic due to light shading, including: If the distance between the curved section and the entrance section of the target tunnel is less than the preset distance between the curved section and the entrance section, and the radius of curvature of the curved section is less than the preset radius of curvature, then it is determined that the curved section has a light shading effect on vehicle driving.
5. The tunnel entrance section supplementary lighting system based on visual recognition according to claim 4, characterized in that, The tunnel impact module is used to determine whether to adjust the supplementary lighting distance, including: If the curved section obstructs the light for vehicle travel, adjust the distance of the supplemental lighting.
6. The tunnel entrance section supplementary lighting system based on visual recognition according to claim 1, characterized in that, The vehicle condition analysis module is used to evaluate the vehicle condition stability value, including: The vehicle stability value is determined based on driving condition characteristics; Calculate the average vehicle speed ratio and the dispersion ratio respectively; The sum of the average vehicle speed ratio adjusted according to the first contribution ratio and the dispersion ratio adjusted according to the second contribution ratio is determined as the vehicle condition stability value. The driving state characteristics include the dispersion of average vehicle speed and vehicle spacing.
7. The tunnel entrance section supplementary lighting system based on visual recognition according to claim 6, characterized in that, The vehicle condition analysis module is used to classify supplementary lighting needs into categories, including: If the vehicle condition stability value is greater than or equal to the preset vehicle condition stability value, it is classified as a strong supplementary lighting requirement category. If the vehicle condition stability value is less than the preset vehicle condition stability value, it is classified as a weak supplementary lighting requirement category.
8. The tunnel entrance section supplementary lighting system based on visual recognition according to claim 1, characterized in that, The supplementary lighting adjustment module is used to identify densely trafficked areas, including: The predetermined road area is divided into several path segments, and the corresponding road traffic condition images are retrieved. Mark the vehicles whose driving distance within each of the aforementioned path segments is less than the preset driving distance; Count the number of marked vehicles and calculate the ratio of the number of marked vehicles to the total number of vehicles; If the quantity ratio is greater than or equal to a preset quantity ratio, the corresponding path segment is identified as the densely trafficked segment.
9. The tunnel entrance section supplementary lighting system based on visual recognition according to claim 1, characterized in that, The supplementary lighting adjustment module is used to determine the degree of supplementary lighting optimization required, including: The supplementary lighting optimization requirement is determined based on the deviation of the ratio of the light intensity in front of the vehicle to the light intensity behind the vehicle in the densely trafficked section, as well as the distance between the densely trafficked section and the entrance section. Calculate the deviation ratio and distance ratio separately; The sum of the deviation ratio adjusted according to the first influence ratio and the distance ratio adjusted according to the second influence ratio is determined as the supplementary lighting optimization requirement.
10. The tunnel entrance section supplementary lighting system based on visual recognition according to claim 9, characterized in that, The supplementary lighting adjustment module is used to adjust the initial supplementary lighting intensity, including: The adjustment amount of the initial supplemental lighting intensity is positively correlated with the degree of supplemental lighting optimization requirement.
Citation Information
Patent Citations
Highway tunnel entrance section illumination regulation and control system for relieving black hole effect
CN116963343A
