Tunnel light attenuation shed brightness threshold optimization method and system based on subjective and objective fusion

Abstract

This invention discloses a method and system for optimizing the brightness threshold of tunnel light-reducing canopies based on a fusion of objective and subjective data. It relates to the fields of traffic safety engineering and tunnel lighting design technology. The method involves acquiring the natural light brightness value outside the tunnel and constructing a light environment with initial road surface brightness values ​​in a simulated light-reducing canopy section. Objective and subjective data of the initial road surface brightness value light environment are acquired. A comprehensive evaluation index is calculated based on the objective and subjective data. Based on the initial road surface brightness value and the comprehensive evaluation index, a nonlinear regression prediction model is used to guide the optimization of the road surface brightness value to determine the optimal brightness threshold. A comprehensive evaluation index incorporating objective visibility rate and subjective driving confidence is constructed. Combined with a scientific and efficient algorithm, it can accurately obtain the minimum brightness threshold that meets both safety and comfort requirements, providing a quantitative basis for the scientific design and energy-saving operation of light-reducing canopies.

Description

Technical Field

[0001] This invention relates to the fields of traffic safety engineering and tunnel lighting design technology, and in particular to a method and system for optimizing the brightness threshold of tunnel light reduction canopies based on a fusion of subjective and objective factors. Background Technology

[0002] The statements in this section are merely background information relating to this disclosure and do not necessarily constitute prior art.

[0003] In the construction of highways and other transportation infrastructure, tunnels serve as crucial transportation arteries, making the lighting design of their entrance sections paramount. Due to the significant brightness difference between the inside and outside of a tunnel during the day, drivers encounter a strong "black hole effect" upon entering, causing temporary blind spots or adaptation delays in their visual adaptation system, severely impacting driving safety. Therefore, in engineering practice, light-reducing canopies are typically installed before the tunnel entrance. These physical structures create a smooth brightness transition zone between the high-brightness environment outside the tunnel and the low-brightness environment inside.

[0004] However, existing methods for determining the brightness or design parameters of light-reducing canopy surfaces mainly suffer from the following technical bottlenecks: First, the evaluation indicators are too simplistic and lack an effective integration of subjective and objective dimensions. Existing technical solutions tend to focus on objective considerations, typically fitting a brightness reduction coefficient by measuring the driver's "visibility distance" to detect objects. While "visibility distance" reflects the driver's physiological ability to "see" obstacles, it cannot comprehensively quantify the driver's psychological state under sudden changes in lighting conditions. In actual high-speed driving, "seeing" is not entirely equivalent to "safety." Even if the ambient brightness meets the minimum visibility requirements, if the driver still feels visually oppressed, panicked, or lacks confidence in the road ahead (i.e., insufficient driving confidence), it may still lead to rigid operation, excessive braking, or hesitant judgment. Conversely, relying solely on subjective judgment lacks quantitative standards due to significant individual differences.

[0005] Secondly, the lack of precise threshold optimization algorithms makes it difficult to balance safety and energy efficiency. Existing design methods often focus on the structural design of the light-reducing canopy (such as length and material transmittance calculations). The determination of its brightness requirements is usually based on theoretical dark adaptation curve calculations, or in experiments, a simple, crude adjustment method of "linearly increasing brightness if not met" is used. This unidirectional, linear adjustment method is inefficient and makes it difficult to accurately capture the critical brightness threshold point that "meets safety and comfort requirements while minimizing energy consumption." In actual engineering projects, to ensure acceptance, the design often has a large redundancy, resulting in an excessively large scale of light-reducing canopy construction or excessively high internal supplemental lighting energy consumption, leading to waste of construction costs and operating energy; while if the design is inadequate, safety hazards cannot be eliminated. Summary of the Invention

[0006] To overcome the shortcomings of existing technologies, such as the single evaluation index for the brightness design of light-reducing canopies, the lack of a subjective-objective integrated evaluation model, and the lack of a precise threshold optimization algorithm, this invention provides a method and system for optimizing the brightness threshold of tunnel light-reducing canopies based on subjective-objective integration. It mathematically integrates objective visibility rate with subjective driving confidence and uses intelligent algorithms to accurately find the optimal brightness threshold for the light environment of the light-reducing canopy, aiming to comprehensively formulate brightness standards for light-reducing canopies.

[0007] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: In a first aspect, the present invention provides a method for optimizing the brightness threshold of a tunnel light reduction canopy based on a fusion of subjective and objective factors, including: Obtain the natural light intensity value outside the tunnel, and construct the light environment with the initial road surface brightness value in the simulated light-reducing canopy section; Acquire objective and subjective data on the initial road surface brightness under the light environment; A comprehensive evaluation index is calculated based on objective and subjective data. Based on the initial pavement brightness value and the comprehensive evaluation index, a nonlinear regression prediction model is used to guide the optimization of the pavement brightness value and determine the optimal brightness threshold.

[0008] A further technical solution involves organizing multiple experimental drivers to drive vehicles through a simulated light-reducing canopy section at a designed speed, recording the drivers' binary visual recognition results of the target objects, and obtaining micro-operation data of the drivers' driving behavior to obtain objective data.

[0009] A further technical solution involves having drivers fill out a driving confidence quantification form after passing through a simulated light-reducing canopy section, and then conducting a post-experimental comparison and scoring to obtain subjective data.

[0010] A further technical solution is that the comprehensive evaluation index is expressed as:

[0011]

[0012]

[0013] in, , These are the weighting coefficients. For safety performance index, For comfort performance index, For years of driving experience, This is the dark adaptation time.

[0014] A further technical solution involves constructing a nonlinear regression model based on a sample set, fitting road surface brightness and a comprehensive evaluation index, and considering the logarithmic characteristics of human visual perception and the S-shaped threshold characteristics established by psychological confidence to build a light environment-evaluation index prediction model.

[0015] A further technical solution is that the light environment-evaluation index prediction model is expressed as follows:

[0016] in, For road surface brightness, The inflection point brightness, The maximum increase in the comprehensive evaluation index. This serves as the baseline value for the comprehensive evaluation index. This is the sensitivity coefficient for the brightness-confidence relationship.

[0017] A further technical solution involves using guided optimization to solve for the theoretical brightness value that makes the predicted evaluation value equal to the target threshold by inverse operation based on the currently updated prediction model. The comprehensive evaluation index under this theoretical brightness value is then recalculated, and its deviation from the comprehensive evaluation target threshold is calculated. When the deviation is less than the convergence accuracy, the current theoretical brightness value is output as the optimal brightness threshold.

[0018] Secondly, this invention provides a tunnel light reduction canopy brightness threshold optimization system based on subjective and objective fusion, comprising: The environmental parameter acquisition module is configured to: acquire the natural light brightness value outside the tunnel and construct the light environment of the initial road surface brightness value in the simulated light reduction canopy section; The objective and subjective data acquisition module is configured to acquire objective and subjective data under the initial road surface brightness value light environment. The comprehensive evaluation module is configured to calculate a comprehensive evaluation index based on objective and subjective data. The threshold optimization module is configured to: based on the initial road surface brightness value and the comprehensive evaluation index, use a nonlinear regression prediction model to perform guided optimization of the road surface brightness value and determine the optimal brightness threshold.

[0019] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the tunnel light reduction shed brightness threshold optimization method based on subjective-objective fusion as described in the first aspect.

[0020] Fourthly, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the tunnel light reduction shed brightness threshold optimization method based on subjective and objective fusion as described in the first aspect.

[0021] The above one or more technical solutions have the following beneficial effects: This invention constructs a comprehensive evaluation index that includes objective visibility rate and subjective driving confidence. Combined with a scientific and efficient algorithm, it can accurately obtain the minimum brightness threshold that meets both safety and comfort requirements, providing a quantitative basis for the scientific design and energy-saving operation of light-reducing sheds.

[0022] This invention proposes a Q-index model that mathematically integrates objective "visibility" with subjective "sense of security," resulting in a more comprehensive evaluation. Compared to existing technologies that solely rely on visual distance, this invention avoids design biases such as "objective visibility but subjective fear" (leading to rigid operation) or "blind confidence" (leading to ignoring risks), ensuring that the established brightness standards better meet the actual physiological and psychological needs of drivers under the "black hole effect."

[0023] This invention abandons traditional linear increases in brightness or theoretical calculation methods, and innovatively introduces a guided optimization algorithm based on a nonlinear regression prediction model. This algorithm can converge quickly and accurately capture the critical brightness point (i.e., the lowest energy consumption point) that meets the safety and comfort conditions. It effectively solves the problems of wasteful construction costs and excessive operating energy consumption caused by excessive safety margins in traditional designs, and achieves the best balance between safety and energy saving. Attached Figure Description

[0024] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0025] Figure 1 This is a flowchart of the tunnel light reduction canopy brightness threshold optimization method based on subjective and objective fusion according to an embodiment of the present invention; Figure 2 This is a flowchart of the guided optimization algorithm based on a nonlinear regression prediction model in an embodiment of the present invention. Detailed Implementation

[0026] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0027] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0028] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0029] Example 1 like Figure 1 As shown in the figure, this embodiment discloses a method for optimizing the brightness threshold of a tunnel light reduction canopy based on subjective and objective fusion. The method includes the following steps: S1: Obtain the natural light intensity value outside the tunnel and construct the light environment of the initial road surface brightness value in the simulated light reduction canopy section; In this embodiment, the natural light intensity value outside the tunnel under specific weather conditions is obtained. In the simulated light-reducing canopy section, the lighting device is controlled to simulate the initial road surface brightness value. The lighting environment for (initial test brightness). Natural light brightness value. This represents the actual natural light level outside the tunnel; it is a measured value of the external brightness and serves as the input parameter for the experiment, primarily used to initialize the experimental environment. Initial road surface brightness value. Its purpose is to simulate the road surface brightness at the entrance of the light-reducing canopy, and it needs to form a reasonable transition relationship with the natural light brightness value outside the tunnel. According to the tunnel lighting design principle, the natural light brightness value is taken. 10%-20% as the initial road surface brightness value .

[0030] Select a tunnel or closed road section about 100m long as a simulated light-reducing canopy section, and install a high-power dimmable LED array on the top of the simulated light-reducing canopy section.

[0031] A high-precision luminance meter was used to measure the natural light intensity outside the tunnel on a given day. For example, the luminance outside the tunnel was measured at noon on a clear day. The system uses PWM dimming control of the LED array based on preset initial test values ​​to bring the road surface brightness within the simulated light-reducing canopy section to the initial value. .

[0032] S2: Obtain objective and subjective data on the initial road surface brightness under the light environment; In this embodiment, multiple experimental drivers drove vehicles at a designed speed through a simulated light-reducing canopy section, recording the binary visual recognition results for the target object. Acquire driver's micro-operation data and obtain a quantitative score of driver confidence. To obtain subjective and objective experimental data.

[0033] Several experimental drivers (30 in this embodiment) were selected (covering different ages, genders, and driving experience) and required to drive the vehicle through a simulated road section at a design speed (e.g., 80 km / h). Inside the simulated light-reducing canopy road section, a cubic target object was placed at a certain distance from the entrance (15m in this embodiment), with its surface covered by a gray diffuse reflection coating with a reflectivity of 20%.

[0034] In some implementations, the target object is a standardized 20cm×20cm×20cm cube with N5 gray matte paint sprayed on its surface.

[0035] The driver wears an eye tracker or provides feedback via an onboard button to confirm whether an obstacle was detected before approaching. If the obstacle was detected beyond the safe stopping visual distance, it is recorded. Otherwise remember 0, obtaining the binary visual recognition result; simultaneously, driver's micro-operation data is collected via the vehicle's CAN bus, including the standard deviation of the steering wheel angle and longitudinal acceleration fluctuation at the moment of entering the light-reducing canopy; the binary visual recognition result and the driver's micro-operation data constitute objective data. The safe parking sight distance is shown in Table 1 below, and can be set and adjusted according to specific circumstances; this embodiment does not impose specific limitations.

[0036] Table 1 Examples of Safe Parking Sight Distances

[0037] The safety performance index is calculated based on objective data and is expressed as follows:

[0038]

[0039] in, For safety performance index, Total number of drivers For the first The binary visual recognition results of the driver. The penalty coefficient for operational volatility. This means that even if the identification is successful, if the operation fluctuates greatly, the safety performance will be deducted proportionally. To The normalization result, For the first A comprehensive measurement of the fluctuations in the driver's micro-operations. The normalized standard deviation of the steering wheel angle. The normalized longitudinal acceleration fluctuation, , These are the weights for the standard deviation of the steering wheel angle and the longitudinal acceleration fluctuation, respectively.

[0040] After completing the test, drivers filled out a "Driving Confidence Quantitative Evaluation Form" on an in-vehicle tablet. The evaluation used a 5-point scale (1 - very panicked, 3 - somewhat confident, 5 - very confident), resulting in a confidence score. After the experiment, drivers were randomly shown first-person visual video clips of their driving under different test brightness levels on a high-definition screen and asked to rate these clips again, obtaining a post-experimental comparative score. This effectively eliminates memory bias and rating scale drift between different experimental rounds, obtaining more sensitive and stable subjective comparative data for brightness differences, which can be used for subsequent model validation.

[0041] The comfort performance index is calculated based on subjective data and is expressed as follows:

[0042] in, For comfort performance index, For the first Driver confidence rating For the first The subjective rating calibration coefficient for each driver is calculated from their post-event comparative rating.

[0043] The post-hoc comparative scores are normalized to obtain normalized post-hoc comparative scores, and the consistency index of the two driver rating sequences is calculated. , If the driver's confidence score and post-mortem comparison score are highly consistent, then A value close to 1 indicates that the driver's initial confidence score is stable and reliable, and the calibration coefficient is close to 1, meaning the initial confidence score has been fully adopted; if the two are inconsistent or negatively correlated, that is... A low or negative score indicates that the driver's initial confidence score may be arbitrary or scale-shifted, and the calibration coefficient will be less than 1, thereby reducing the weight of the unstable score in the overall comfort performance index.

[0044] S3: A comprehensive evaluation index is calculated based on objective and subjective data; In this embodiment, before the experiment begins, the driver undergoes basic visual function tests (such as contrast sensitivity and dark adaptation time) and his / her years of driving experience are recorded.

[0045] Summarize the experimental data from one round and calculate the comprehensive evaluation index. The calculation formula is:

[0046]

[0047]

[0048] in, , These are the weighting coefficients. For years of driving experience, This is for dark adaptation time. For drivers with weaker visual function or novice drivers, the weight of objective visual performance can be automatically increased. The system can identify the differences in needs among different driver groups, providing a data foundation for developing graded lighting standards, making the optimization results more universal and precise.

[0049] Existing light-reducing canopy designs often only aim to meet standard brightness or a single visibility rate, neglecting the driver's psychological experience and driving load during actual driving. This invention, through the aforementioned technical features, simultaneously collects the driver's objective visibility results (binary recognition rate) and subjective driving confidence (quantitative score), constructing a comprehensive evaluation index Q that includes both safety and comfort performance indices. This index weighted and integrates the "visible" safety baseline with the "stable driving" comfort experience, providing a more comprehensive and scientific reflection of the quality of the light environment in the light-reducing canopy. Furthermore, this invention calibrates the driver's initial subjective score by introducing post-event comparative scoring, effectively eliminating individual score drift and memory bias, significantly improving the stability and reliability of subjective data.

[0050] When calculating the comprehensive evaluation index, individual driver characteristics, such as years of driving experience and dark adaptation time, were incorporated as weighting factors for objective visual performance. For novice drivers with weaker visual function or less experience, the system automatically increased the weight of objective visual performance (i.e., "visibility"). This allows the optimal brightness threshold obtained through optimization to cover the needs of a wider range of drivers, making the optimization results more universal and refined. This provides a scientific basis and quantitative means for developing differentiated light-reducing canopy lighting standards for different driver groups in the future.

[0051] S4: Based on the initial road surface brightness value and the comprehensive evaluation index, a nonlinear regression prediction model is used to guide the optimization of the road surface brightness value and determine the optimal brightness threshold.

[0052] In this embodiment, a comprehensive evaluation target threshold is set. and convergence accuracy The optimal brightness threshold under the current operating conditions is determined by a model-guided closed-loop iterative algorithm. Specifically, it includes the following steps: (1) Initialize sample acquisition: within the preset brightness search range Within this range, boundary values ​​and median points are selected as the initial test brightness. Perform steps S2 to S3 respectively to obtain the corresponding comprehensive evaluation index and construct the initial sample set. ,in, For the first The road surface brightness value (unit: cd / m²) at each initial test point corresponds to the test brightness set in the simulated light-reducing canopy section; For the first The comprehensive evaluation index corresponding to each initial test point.

[0053] The minimum brightness in the interval is given in Table 2 below, according to the "Detailed Specifications for Lighting Design of Highway Tunnels" (JTG / T D70 / 2-01-2014).

[0054] Table 2 Basic Brightness Table Inside the Tunnel

[0055] Pick 20% is set as the upper limit.

[0056] The initial sample set is set to 3 sample points by default, which is determined based on the minimum sample requirement for fitting a nonlinear regression model. The Sigmoid model used in this invention contains 4 parameters to be fitted ( , , , Three sample points are sufficient for the initial fitting of the model, while avoiding an excessive number of initial experiments. Only three core points are needed to start the optimization algorithm in the initial stage.

[0057] The specific steps for constructing the initial sample set, combined with the guided optimization algorithm, are as follows: Determine the brightness search range: based on standards and visual adaptation theory. (e.g., when the design speed is 80km / h) cd / m² cd / m²).

[0058] Select initial test brightness: Select 3 key brightness points within the range: : (Lower limit of the range, corresponding to the minimum safe brightness); : (Upper limit of the range, corresponding to sufficiently comfortable brightness); : (Midpoint of the range, corresponding to intermediate transition brightness).

[0059] Data collection: For each Organize driving experiments to collect objective and subjective data, and calculate the corresponding comprehensive evaluation index. .

[0060] Constructing the sample set: 3 groups Data combined into an initial sample set It is used to fit the initial light environment-evaluation index prediction model.

[0061] For example, when =30、 When the value is 800, the initial test brightness is 30, 800, and 415 cd / m². The corresponding Q1, Q2, and Q3 are obtained respectively, which can construct the initial sample set and provide basic data for subsequent nonlinear regression fitting and guided optimization.

[0062] (2) Constructing a light environment-evaluation index prediction model: based on the sample set Fitting road surface brightness With comprehensive evaluation index Nonlinear regression model Considering the logarithmic nature of human visual perception (Weber-Fechner law) and the S-shaped threshold characteristic of psychological confidence establishment (describing the nonlinear growth of the driver's comprehensive evaluation index with the change of road surface brightness under reduced light, named for its curve resembling the Sigmoid function, reflecting the driver's psychological transition from "visual panic" to "sufficient driving confidence"), the following improved Sigmoid prediction model is established:

[0063] in, For model parameters, The inflection point brightness, The maximum growth rate of the comprehensive evaluation index (the vertical stretching coefficient of the S-curve determines the maximum range of change of the comprehensive evaluation index from the baseline value to the saturation value). The baseline value of the comprehensive evaluation index (the vertical offset of the S-curve, corresponding to the minimum comprehensive evaluation index when the brightness approaches 0). The sensitivity coefficient for the brightness-confidence relationship (the slope coefficient of the S-curve, which determines the steepness of the curve; the larger the k, the steeper the curve, indicating that the change in brightness has a more significant impact on the comprehensive evaluation index).

[0064] Furthermore, the sample set is used to fit the parameters of the improved Sigmoid prediction model, and the specific process is as follows: Parameter fitting stage: Initial sample set Substitute the parameters into the model and solve for the optimal parameters using the least squares method or the Gauss-Newton method. .

[0065] Predictive application phase: Once the model parameters are determined, input any road surface brightness value. The corresponding comprehensive evaluation index can be predicted through the model. .

[0066] Furthermore, the sample set is fitted using a piecewise linear approximation-linear regression method: the S-shaped brightness-confidence relationship is divided into three linear intervals: a low brightness growth segment, a medium brightness steep increase segment, and a high brightness saturation segment. The slope and intercept of each segment are fitted using ordinary least squares to quickly estimate the optimal parameters of the prediction model. Details are as follows: By using the three-segment linear approximation of the Sigmoid curve, the nonlinear model is transformed into a linear regression problem: 1) Segmented logic Low brightness segment ( It is approximately linearly increasing, expressed as: .

[0067] Medium brightness range ( It approximates a steep linear growth, expressed as: ).

[0068] High brightness segment ( It is approximately linearly saturated, expressed as: ).

[0069] 2) Solution steps Use the low-brightness points of the initial sample set (e.g.) , ) and medium brightness point ( , Fit the slope of the low-brightness segment and intercept ; Use medium brightness points and high brightness points ( , The slope of the fitted high-brightness segment and obtained ; Inflection point brightness Directly take the luminance value of the mid-brightness point (e.g., 200 cd / m²), sensitivity coefficient Depend on Estimate (e.g.) but ).

[0070] The error of linear approximation is usually within 5%, which meets the needs of rapid estimation in engineering.

[0071] (3) Guided prediction of the next test point: based on the currently updated model The inverse operation is used to solve the problem so that the predicted evaluation value equals the target threshold. Theoretical brightness value at time ,Right now .

[0072] (4) Experimental verification and feedback: Adjust the road surface brightness of the simulated light-reducing canopy section to the predicted value. Repeat steps S2 to S3 to obtain the measured comprehensive evaluation index at this brightness level. .

[0073] (5) Convergence judgment and model update: Calculate the measured deviation .

[0074] like If the system converges, the current brightness will be output. The optimal brightness threshold under this operating condition ; like Then the newly acquired test data will be applied to Expand to sample set Return to step (2) to retrain and update the prediction model, and proceed to the next iteration.

[0075] This method uses historical data to continuously refine the prediction model, enabling it to quickly approach the optimal operating point of a nonlinear system with the fewest experimental iterations, thus balancing optimization efficiency and result accuracy.

[0076] Furthermore, a comprehensive evaluation of the target threshold. The determination is based on statistical analysis of both subjective and objective data, specifically as follows: Objective visual recognition rate benchmark: The target recognition rate R is required to be ≥90% (that is, more than 90% of drivers can identify obstacles beyond the safe stopping sight distance), corresponding to a normalized R value of 0.9.

[0077] Subjective driving confidence benchmark: The normalized driving confidence index C is required to be ≥0.7 (i.e., the average driver confidence score is ≥4 points, with a maximum score of 5 points).

[0078] Weighted fusion calculation: In this embodiment, it is set to , Calculate based on this formula . , It can be adjusted according to actual needs, without specific limitations.

[0079] Furthermore, convergence accuracy The quantification error is determined based on the evaluation index, specifically as follows: Subjective rating error: the smallest distinguishable difference in driver confidence ratings The score is 0.5 out of 5, corresponding to a normalized error of 0.1.

[0080] Objective visual recognition rate error: Due to the limited sample size (typically 30 drivers), the statistical error of the target recognition rate R based on binary visual recognition results. It is approximately ±5%, corresponding to a normalized error of 0.05.

[0081] Comprehensive error calculation: After considering weighting coefficients, the comprehensive evaluation index is calculated. The minimum resolvable error is .

[0082] Through the above technical solution, this invention introduces a guided optimization algorithm based on a nonlinear regression model, achieving accurate and efficient optimization of the brightness threshold. Addressing the nonlinearity of human visual perception and the S-shaped threshold characteristics of psychological confidence, it abandons traditional linear optimization or simple dichotomy methods and innovatively establishes an improved Sigmoid light environment-evaluation index prediction model. This model accurately describes the nonlinear growth law of the comprehensive evaluation index with brightness (the transition from "visual panic" to "sufficient driving confidence"). Based on this, a guided closed-loop iterative algorithm is employed, using model prediction to guide the selection of the next test point and continuously refining the model using measured data. Compared to traditional ergonomic experimental methods, this invention can quickly and accurately approximate the minimum brightness threshold (optimal operating point) that meets safety and comfort requirements with the fewest experimental iterations, avoiding energy waste caused by over-design and safety hazards caused by under-design.

[0083] In some implementations, the algorithm can be replaced with a PID algorithm, or physiological indicators such as heart rate can be introduced as a basis for subjective evaluation.

[0084] In summary, this invention aims to address the problems of single evaluation indicators and lack of precise optimization algorithms in the existing design of light-reducing canopies. It includes: acquiring the natural light intensity outside the canopy and constructing an initial brightness environment on a simulated road section; conducting driving experiments to collect objective visibility rates of target objects and drivers' subjective confidence scores; constructing a comprehensive evaluation index Q that integrates subjective and objective factors, weighting and fusing "visibility rate" and "sense of safety"; and utilizing a guided optimization algorithm based on a nonlinear regression prediction model to accurately approximate the minimum brightness threshold that meets safety and comfort requirements. This invention abandons complex color temperature variables and focuses on the precise optimization of brightness parameters, providing scientific and quantitative data support for the energy-saving design of light-reducing canopies, effectively avoiding energy waste caused by excessive design margins or safety hazards caused by insufficient design.

[0085] Example 2 This embodiment discloses a tunnel light reduction canopy brightness threshold optimization system based on subjective and objective fusion, including: The environmental parameter acquisition module is configured to: acquire the natural light brightness value outside the tunnel and construct the light environment of the initial road surface brightness value in the simulated light reduction canopy section; The objective and subjective data acquisition module is configured to acquire objective and subjective data under the initial road surface brightness value light environment. The comprehensive evaluation module is configured to calculate a comprehensive evaluation index based on objective and subjective data. The threshold optimization module is configured to: based on the initial road surface brightness value and the comprehensive evaluation index, use a nonlinear regression prediction model to perform guided optimization of the road surface brightness value and determine the optimal brightness threshold.

[0086] In this embodiment, the adjustment of the road surface brightness of the simulated light reduction canopy section is achieved through a simulated light environment control system. The simulated light environment control system includes a dimmable lighting array (high-power LED module) set in the simulated light reduction canopy section and a main controller (receiving instructions and controlling the output luminous flux of the LEDs through DALI or 0-10V protocol). The main controller is configured to adjust the luminous flux of the lighting array according to the brightness instructions in order to form the target brightness in the simulated road section.

[0087] A data interaction terminal was installed inside the experimental vehicle to record the visual recognition results of the experimental drivers and input their driving confidence scores.

[0088] The processing unit receives data from the environmental parameter acquisition module and the subjective and objective evaluation module, executes the methods and steps of the comprehensive evaluation module and the threshold optimization module, and outputs the optimal brightness threshold.

[0089] Example 3 The purpose of this embodiment is to provide a computing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method of Embodiment 1.

[0090] Example 4 The purpose of this embodiment is to provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the method of Embodiment 1.

[0091] The steps and methods involved in the apparatuses of Embodiments 3 and 4 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0092] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0093] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0094] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for optimizing the brightness threshold of a tunnel light reduction canopy based on a fusion of subjective and objective factors, characterized in that: include: Obtain the natural light intensity value outside the tunnel, and construct the light environment with the initial road surface brightness value in the simulated light-reducing canopy section; Acquire objective and subjective data on the initial road surface brightness under the light environment; A comprehensive evaluation index is calculated based on objective and subjective data. Based on the initial pavement brightness value and the comprehensive evaluation index, a nonlinear regression prediction model is used to guide the optimization of the pavement brightness value and determine the optimal brightness threshold.

2. The method for optimizing the brightness threshold of a tunnel light reduction canopy based on subjective and objective fusion as described in claim 1, characterized in that, Multiple experimental drivers were organized to drive vehicles through a simulated light-reducing canopy section at a designed speed. The drivers' binary visual perception of the target object was recorded, and the drivers' micro-operation data were obtained to acquire objective data.

3. The method for optimizing the brightness threshold of a tunnel light reduction canopy based on subjective and objective fusion as described in claim 2, characterized in that, After driving through a simulated light-reducing canopy section, drivers filled out a driving confidence evaluation form and then compared their scores after the experiment to obtain subjective data.

4. The method for optimizing the brightness threshold of a tunnel light reduction canopy based on subjective and objective fusion as described in claim 1, characterized in that, The comprehensive evaluation index is expressed as follows: in, , These are the weighting coefficients. For safety performance index, For comfort performance index, For years of driving experience, This is the dark adaptation time.

5. The method for optimizing the brightness threshold of a tunnel light reduction canopy based on subjective and objective fusion as described in claim 1, characterized in that, Based on the sample set, a nonlinear regression model was constructed by fitting the road surface brightness and the comprehensive evaluation index. Considering the logarithmic characteristics of human visual perception and the S-shaped threshold characteristics of psychological confidence, a light environment-evaluation index prediction model was established.

6. The method for optimizing the brightness threshold of a tunnel light reduction canopy based on subjective and objective fusion as described in claim 5, characterized in that, The light environment-evaluation index prediction model is expressed as follows: in, For road surface brightness, The inflection point brightness, The maximum increase in the comprehensive evaluation index. This serves as the baseline value for the comprehensive evaluation index. This is the sensitivity coefficient for the brightness-confidence relationship.

7. The method for optimizing the brightness threshold of a tunnel light reduction canopy based on subjective and objective fusion as described in claim 1, characterized in that, During guided optimization, based on the currently updated prediction model, the theoretical brightness value that makes the predicted evaluation value equal to the target threshold is solved through inverse operation. The comprehensive evaluation index under this theoretical brightness value is recalculated, and the deviation between it and the comprehensive evaluation target threshold is calculated. When the deviation is less than the convergence accuracy, the current theoretical brightness value is output as the optimal brightness threshold.

8. A tunnel light reduction canopy brightness threshold optimization system based on subjective and objective fusion, characterized in that, include: The environmental parameter acquisition module is configured to: acquire the natural light brightness value outside the tunnel and construct the light environment of the initial road surface brightness value in the simulated light reduction canopy section; The objective and subjective data acquisition module is configured to acquire objective and subjective data under the initial road surface brightness value light environment. The comprehensive evaluation module is configured to calculate a comprehensive evaluation index based on objective and subjective data. The threshold optimization module is configured to: based on the initial road surface brightness value and the comprehensive evaluation index, use a nonlinear regression prediction model to perform guided optimization of the road surface brightness value and determine the optimal brightness threshold.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the method for optimizing the brightness threshold of a tunnel light reduction shed based on subjective and objective fusion as described in any one of claims 1-7.

10. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the method for optimizing the brightness threshold of a tunnel light reduction shed based on subjective and objective fusion as described in any one of claims 1-7.

Images