LED display screen multi-scene adaptive dimming method and system

Abstract

The invention discloses a multi-scene self-adaptive dimming method and system for an LED display screen, and relates to the technical field of self-adaptive control. The method comprises the following steps: judging whether a target display screen is in a fixed scene or a non-fixed scene based on position information corresponding to the target display screen; under the condition of a fixed scene, constructing an environment sensing network comprising a plurality of sensing nodes, and obtaining corresponding environment parameters according to each sensing node; according to the environment sensing network, generating an update brightness adjustment strategy among a plurality of environment parameters associated with the sensing node; the technical key points are as follows: the whole scheme is from scene recognition, environment modeling, strategy making, fuzzy reasoning, weight correction to self-adaptive dimming, all links are progressive layer by layer and support each other, display screen brightness adjustment is jointly promoted to move from a traditional empirical mode to a data driving mode, and static control is changed to dynamic self-adaptive control; and the intelligent display management target is really realized.

Description

Technical Field

[0001] The present invention relates to the field of adaptive control technology, and in particular to a multi-scene adaptive dimming method and system for an LED display screen. Background Art

[0002] An adaptive control system is a system that automatically adjusts itself to achieve optimal performance according to some predetermined criteria. When adaptive dimming is performed on LED screens, the LED screens are usually adjusted according to real-time environmental changes and cannot be perceived and adjusted in advance. At the same time, the currently adapted adaptive control systems are aimed at fixed-installed LED screens.

[0003] Traditional brightness adjustment technologies generally rely on fixed sensor feedback mechanisms or perform brightness control based on a single empirical rule, making them difficult to adapt to complex and changing actual usage environments. For example, in outdoor advertising screens or in-vehicle displays, traditional systems often adjust brightness based solely on light intensity, ignoring the influence of key factors such as humidity and the orientation of the display. This leads to problems such as brightness being too high and glaring or too low and invisible in special scenarios such as rainy days, in tunnels, or at night. In addition, traditional solutions lack the ability to predict moving scenes. For example, when a vehicle enters a tunnel, the screen brightness cannot be adjusted in advance, causing visual mutations and safety hazards. At the same time, most systems use a static weight distribution method and are unable to dynamically optimize brightness decisions based on environmental changes. The response is delayed and the adjustment is inaccurate. These problems seriously limit the intelligence level and scope of application of brightness control systems. Summary of the Invention

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: A multi-scene adaptive dimming method for an LED display screen, the method comprising: Based on the location information corresponding to the target display screen, determine whether it is in a fixed scene or a non-fixed scene; When in a fixed scene, an environmental perception network consisting of several perception nodes is constructed, and the corresponding environmental parameters are obtained according to each perception node; According to the environment perception network, an updated brightness adjustment strategy is generated between multiple environment parameters associated with the perception node; Determine the actual weight between each sensing node and the multiple environmental parameters according to the environmental parameters, and add each actual weight to the corresponding updated brightness adjustment strategy to obtain the corresponding weighted brightness adjustment strategy; The preset brightness control model is trained according to multiple weighted brightness adjustment strategies to obtain a standard brightness control model; When in a non-fixed scene, GPS data and map navigation data are combined to predict the type of area to be entered. When the area type matches any type in the area adjustment library, the preliminary dimming preparation mechanism is triggered; After the initial dimming preparation mechanism is completed and the target display enters the target area, the regional environmental data corresponding to the target area is continuously monitored and a dynamic fine-tuning strategy is implemented: input parameters are fuzzified, fuzzy rules are constructed, and fuzzy reasoning is performed to obtain the recommended brightness value; After executing the dynamic fine-tuning strategy, the change rate of the regional environmental data is detected within the preset acquisition period, and the weight of the corresponding regional environmental data in the fuzzy rule is dynamically adjusted according to the detection results. After the weight adjustment, the final brightness value corresponding to the acquisition period is calculated and the adaptive dimming action is completed.

[0005] Furthermore, in the process of obtaining the corresponding environmental parameters, the data received by each sensing node is preprocessed to screen the environmental parameters; wherein, the preprocessing process is: decoding and cleaning, identifying the data packet format, eliminating invalid or abnormal values, and then executing the verification mechanism; the screened environmental parameters are classified and stored by type; wherein, the environmental parameters include at least: light intensity and the first environmental humidity.

[0006] Furthermore, a fusion algorithm is used to analyze the correlation between different environmental parameters, and the algorithms used in the fusion algorithm include at least: statistical correlation analysis, machine learning model and cluster analysis; the analysis process is: using the Pearson correlation coefficient to analyze the linear correlation between two continuous variables, which is used to judge the strength of the correlation between each environmental parameter and the final brightness value; constructing a random forest prediction model, and at the same time outputting the influence weight of each input variable on the output; dividing the scenes into several types according to different combinations of environmental parameters, including at least: daytime outdoor mode, rainy mode in tunnels and urban road mode; based on the analysis results, formulating several sets of brightness adjustment strategies; wherein the variables are environmental parameters.

[0007] Furthermore, the process of obtaining the corresponding weighted brightness adjustment strategy is as follows: collecting brightness adjustment records of historical time periods as training data; using a machine learning algorithm to train a brightness control model; wherein the machine learning algorithm is either a neural network or a decision tree; and evaluating the performance of the brightness control model through a cross-validation method.

[0008] Furthermore, the process of predicting the type of area to be entered is as follows: based on the current position information and moving direction of the target display screen, combined with map navigation data, and using a path planning algorithm to analyze the moving path of the target display screen, the type of area to be entered is predicted.

[0009] Furthermore, when the preliminary dimming preparation mechanism is triggered, the time required for the real-time target display screen to reach the target area is calculated based on the moving speed of the target display screen and the distance from the target area type; when the required time is equal to the preset time, the preliminary dimming action is performed: the brightness adjustment is completed according to the brightness adjustment gradient within the preset time, so that the final brightness value is adjusted to the set brightness value pre-marked by the target area type at the same time as the target display screen reaches the target area.

[0010] Furthermore, the regional environmental data includes at least: a first light intensity, a second light intensity and a second ambient humidity; wherein the first light intensity is the light intensity within the target area, the second light intensity is the light intensity facing the target display screen, and the second ambient humidity is the ambient humidity within the target area.

[0011] Furthermore, the process of fuzzifying input parameters is as follows: each input regional environmental data is divided into three fuzzy sets: low, medium, and high; each fuzzy set corresponds to a membership value; the process of constructing fuzzy rules is as follows: constructing an IF-THEN fuzzy rule engine for brightness adjustment decision-making; wherein, the input variables include at least: any fuzzy set of the three fuzzy sets in each regional environmental data, and the output variables are brightness recommendation values, which are respectively low brightness 40%, medium brightness 60%, and high brightness 80%; the process of completing fuzzy reasoning is as follows: after completing the fuzzification of the input variables, matching the current membership with the IF-THEN fuzzy rule engine, calculating the activation degree of each fuzzy rule, and combining the rule weight to obtain the brightness recommendation value.

[0012] Furthermore, the process of obtaining the final brightness value is as follows: when calculating the change rate of each regional environmental data detected within the preset acquisition cycle, for each regional environmental data, calculate its difference from the value at the previous moment in each acquisition cycle, and divide it by the time interval to obtain the change rate; set a corresponding change threshold for each regional environmental data. When the change rate of a certain regional environmental data exceeds its change threshold, the weight adjustment engine is triggered, and the original weight corresponding to the regional environmental data is multiplied by the set ratio value. The result is the adjusted weight; the set ratio value range is between 1 and 1.5; after completing the weight adjustment, the adjusted weight is used and recalculated using the weighted average method to obtain the final brightness value.

[0013] A multi-scene adaptive dimming system for an LED display screen, the system comprising: Scene recognition unit: Based on the location information of the target display screen, it determines whether it is in a fixed scene or a non-fixed scene; Environment construction unit: When in a fixed scene, it builds an environment perception network consisting of several perception nodes and obtains the corresponding environment parameters according to each perception node; Brightness modeling unit: generates an updated brightness adjustment strategy based on multiple environmental parameters associated with the perception node based on the environment perception network; Weighted adjustment unit: determines the actual weight between each sensing node and multiple environmental parameters according to the environmental parameters, and adds each actual weight to the corresponding updated brightness adjustment strategy to obtain the corresponding weighted brightness adjustment strategy; Training and verification unit: trains the preset brightness control model according to multiple weighted brightness adjustment strategies to obtain a standard brightness control model; Regional prediction unit: When in a non-fixed scene, it combines GPS data and map navigation data to predict the type of area it is about to enter. When the area type matches any type in the regional adjustment library, the preliminary dimming preparation mechanism is triggered; Fuzzy reasoning unit: After the initial dimming preparation mechanism is completed and the target display enters the target area, it continues to monitor the regional environmental data corresponding to the target area and implements a dynamic fine-tuning strategy: fuzzifying the input parameters, building fuzzy rules and completing fuzzy reasoning to obtain the recommended brightness value; Dynamic correction unit: After executing the dynamic fine-tuning strategy, it detects the rate of change of regional environmental data within the preset acquisition cycle, dynamically adjusts the weight of the corresponding regional environmental data in the fuzzy rule based on the detection results, and calculates the final brightness value corresponding to the acquisition cycle after the weight adjustment, and completes the adaptive dimming action.

[0014] The present invention provides a multi-scene adaptive dimming method and system for an LED display screen, which has the following beneficial effects: 1) This solution uses GPS positioning to determine scene types and build a differentiated environmental perception network. This not only improves the flexibility of overall deployment, but also enables precise adaptation to different usage scenarios. This addresses the problem that traditional fixed brightness adjustment cannot cope with complex environmental changes. The overall solution can serve both static advertising screens and dynamic scenarios such as in-vehicle displays, significantly enhancing the overall solution's applicability and responsiveness. 2) This solution uses a fusion algorithm based on statistical analysis and machine learning to extract correlations between environmental parameters and formulate multiple brightness adjustment strategies accordingly. This not only improves the scientific nature of brightness adjustment, but also provides the overall solution with the ability to switch between multiple modes, thus addressing the problem of traditional brightness adjustment strategies being single and lacking in scene differentiation. This demonstrates the technical advantages of a data-driven and model-driven dual-wheel drive, greatly enhancing the intelligence and adaptability of brightness adjustment and making it suitable for a variety of application scenarios. 3) This solution introduces weighted fuzzy reasoning, which not only improves the accuracy of brightness recommendation output but also enhances its ability to handle the synergistic effects of multiple variables. This addresses the problem of traditional brightness adjustment ignoring the mutual influence of parameters, which can lead to misjudgments. Subsequent solutions will further incorporate the environmental change rate to dynamically adjust the weights, enabling the overall solution to quickly respond to sudden environmental changes, significantly improving the stability and accuracy of brightness adjustment. 4) This solution dynamically adjusts rule weights based on the environmental change rate, enhancing both adaptability to sudden environmental changes and the real-time and robustness of brightness adjustment. This addresses the issues of delayed response and untimely adjustment in traditional brightness adjustment, shifting brightness adjustment from passive response to active prediction. This significantly improves user experience and system intelligence, and is a key component in building a closed-loop brightness control system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flowchart of the overall steps of a multi-scene adaptive dimming method for an LED display screen in the present invention. DETAILED DESCRIPTION

[0016] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0017] Example 1: See also Figure 1 This embodiment provides a multi-scene adaptive dimming method for an LED display screen, and the specific steps are as follows: S1. Determine whether the target display screen is in a fixed scene or a non-fixed scene based on the location information corresponding to the target display screen; The location information is obtained through the GPS module configured in the target display screen, which is used to obtain the positioning data or latitude and longitude data of the target display screen in real time. By analyzing whether the positioning data has moved, the determination of whether the scene is fixed or non-fixed is achieved. That is, if the positioning data does not move at all, it is a fixed scene. If there is continuous or intermittent movement, it indicates that it is a non-fixed scene.

[0018] S2. When the scene is fixed, an environmental perception network consisting of several perception nodes is constructed, and the corresponding environmental parameters are obtained from each perception node. Based on the constructed environmental perception network, the environmental parameters recorded by each perception node are extracted. These environmental parameters will serve as an important basis for the subsequent formulation of the brightness adjustment strategy. The environmental perception network consists of several different types of sensors, including light sensors and humidity sensors. Each sensor is deployed at a different location within the target display screen. Each sensor collects corresponding data at a set time interval and transmits the data to the central control system. All sensors are connected through wireless communication protocols to form a complete environmental perception network. Environmental parameters include light intensity and first ambient humidity.

[0019] Specific implementation steps

[0020] S2.1. Preprocess the data received by each sensing node and screen the environmental parameters. The preprocessing process includes decoding and cleaning the data, extracting meaningful environmental parameters, identifying the data packet format, eliminating invalid or abnormal values, and then ensuring data integrity through a verification mechanism. S2.2. Classify and store the extracted environmental parameters by type for subsequent processing. For example, taking an outdoor fixed advertising screen as an example, the light intensity value measured by the photosensor is extracted as the environmental parameter, and the data provided by the humidity sensor is also included.

[0021] S3. Generate an updated brightness adjustment strategy based on multiple environmental parameters associated with the sensing node based on the environmental perception network. Specifically, this step formulates or updates the existing brightness adjustment strategy based on the previously collected environmental parameters and corresponding feature vectors. This step requires considering how to reasonably adjust the brightness of the target display, i.e., the LED display, based on environmental changes to ensure both good visual effects and energy conservation.

[0022] Specific implementation steps

[0023] S3.1. Use a fusion algorithm to analyze the correlations between different environmental parameters. The analysis process is as follows: S3.1.1. Calculate the correlation matrix: Use the Pearson correlation coefficient to analyze the linear correlation between two continuous variables to determine the strength of the correlation between each environmental parameter and the final brightness value. S3.1.2. Build a model and extract feature importance: Build a random forest prediction model and output the weight of each input variable (i.e., each environmental parameter) on the output (i.e., the final brightness value). This is used to quantify which environmental parameter contributes most to brightness adjustment. S3.1.3. Use cluster analysis to divide scenes: Based on different combinations of environmental parameters, scenes are divided into several types, including daytime outdoor mode, rainy tunnel mode, and urban road mode, providing a basis for the subsequent development of multiple brightness adjustment strategies. S3.2. Based on the analysis results, develop several brightness adjustment strategies. For example, Strategy A: Set the brightness corresponding to daytime outdoor mode to 70%; Strategy B: Set the brightness corresponding to urban road mode to 55%. Therefore, the fusion algorithm used in S3.1 is a fusion of statistical correlation analysis, machine learning model (i.e., random forest), and cluster analysis.

[0024] By using GPS positioning to determine the scene type and build a differentiated environmental perception network, the flexibility of the overall deployment is improved on the one hand, and on the other hand, it achieves precise adaptation to different usage scenarios, thus solving the problem that traditional fixed brightness adjustment cannot cope with complex environmental changes, and embodies the concept of environment-driven intelligent decision-making. Specifically, this solution integrates a GPS module to analyze in real time whether the target display is in a fixed or non-fixed scene, and then triggers the corresponding brightness control mechanism. In fixed scenes, a multi-node, multi-parameter environmental perception network is built to collect key data including light and humidity. In mobile scenes, the type of area to be entered is predicted in combination with map navigation information, and dimming preparations are started in advance. This scene recognition mechanism avoids the drawbacks of the one-size-fits-all traditional brightness control system, ensuring that the overall solution can serve both static advertising screens and dynamic scenes such as vehicle-mounted displays, significantly enhancing the applicability and responsiveness of the overall solution.

[0025] S4. Determine the actual weight between each sensing node and the multiple environmental parameters based on the environmental parameters, and add each actual weight to the corresponding updated brightness adjustment strategy to obtain a corresponding weighted brightness adjustment strategy. Specifically, the importance of each environmental parameter is calculated and assigned different weight values to enable more precise brightness adjustment. This helps avoid situations where minor factors excessively affect the brightness setting.

[0026] Specific implementation steps

[0027] S4.1. Based on the fusion algorithm used in S3.1, the weights of the influence of each input variable (i.e., each environmental parameter) on the output (i.e., the final brightness value) are derived, and the weight calculation process is completed. S4.2. The calculated weights are applied to the brightness adjustment strategy to form a weighted brightness adjustment strategy. For example, if it is found that the impact of humidity on brightness is much smaller than that of light intensity, a lower weight coefficient is assigned to temperature; conversely, a higher weight coefficient is assigned to light intensity.

[0028] S5. Training a preset brightness control model according to multiple weighted brightness adjustment strategies to obtain a standard brightness control model. The brightness control model is trained using historical data to enable it to predict future brightness requirements, thereby improving the accuracy and response speed of brightness adjustment.

[0029] Specific implementation steps

[0030] S5.1. Collect brightness adjustment records for historical time periods as training data; the historical time period can be one week or one month, which can be set according to actual needs. The brightness adjustment records for the historical time period include: light intensity, humidity, and the brightness value finally set by the system or user at different time periods; it should be noted that these data are used to train the model to identify the patterns of brightness settings under different environmental conditions, so that the model has the ability to learn from experience; S5.2. Model training: Use a machine learning algorithm to train the brightness control model; the machine learning algorithm is either a neural network or a decision tree; S5.3. Evaluate the performance of the brightness control model through cross-validation; an example is: use the brightness adjustment records of the same time period every day in the past month to train the model to ensure that the model can accurately predict future brightness needs; so that the final output standard brightness control model can automatically and intelligently adjust the brightness according to the current environment, improving accuracy and response speed.

[0031] S6. When in a non-fixed scene, GPS data and map navigation data are combined to predict the type of area to be entered. When the area type matches any type in the area adjustment library, the preliminary dimming preparation mechanism is triggered; For a moving target display, GPS data and geographic location information provided by the navigation system (i.e., map navigation data) are combined to predict the type of area the target display will pass through. Brightness settings are adjusted in advance to ensure the display brightness is close to ideal upon entering the new environment, thereby reducing discomfort caused by sudden brightness changes. The area adjustment library includes at least tunnels, sections under overpasses, and forested roads. Other types exist, which will not be detailed here. The only requirement is to ensure that the area types in the library are somewhat different from those in the normal environment.

[0032] Specific implementation steps

[0033] S6.1. Combine the original GPS module with the external navigation system to obtain the location information of the target display screen in real time; S6.2. Based on the current location information and moving direction, combined with the map navigation data, and using the path planning algorithm to analyze the moving path of the target display screen, predict the type of area to be entered or arrived; the path planning algorithm uses the Dijkstra algorithm; S6.3. When the preliminary dimming preparation mechanism is triggered, the time required for the target display screen to reach the target area is calculated based on the moving speed of the target display screen and the distance to the target area type. When the required time is equal to the preset time, the process is executed. Perform preliminary dimming: Complete brightness adjustment according to a uniform brightness adjustment gradient within a preset duration, so that the final brightness value is adjusted to the preset brightness value of the target area type at the same time as the target display screen reaches the target area. The preset brightness value of the target area type is basically 50% brightness, that is, night mode. The preset duration is generally set to 10s, that is, the brightness adjustment starts at a uniform speed at 10s. On the one hand, it can prevent people viewing the target display screen from suddenly entering a different environment and cause discomfort. On the other hand, the target display screen can be adjusted slowly without affecting its normal use. The target area given above is the area corresponding to any type in the area adjustment library.

[0034] The above solution extracts the correlation between environmental parameters based on a fusion algorithm of statistical analysis and machine learning, and formulates multiple sets of brightness adjustment strategies accordingly. On the one hand, it improves the scientific nature of brightness adjustment, and on the other hand, it provides the ability to switch between multiple modes for the overall solution, thus solving the problem of the traditional brightness adjustment strategy being single and lacking scene differentiation. It embodies the technical advantages of data-driven + model-driven dual-wheel drive. By performing Pearson correlation analysis on historical and real-time data, random forest feature importance evaluation, and clustering scene division, it can accurately identify the core factors affecting brightness and divide the environment into modes such as outdoor daytime and rainy tunnels, and set different brightness strategies accordingly. The establishment of this strategy library enables brightness adjustment to no longer rely on a single empirical rule, but instead has the ability to learn from data, greatly improving the intelligence and adaptability of brightness adjustment, and is suitable for a variety of application scenarios.

[0035] S7. After the preliminary dimming preparation mechanism is completed and the target display screen enters the target area, the regional environmental data corresponding to the target area is continued to be monitored, and a dynamic fine-tuning strategy is executed; wherein, the regional environmental data is different from the environmental parameters mentioned in S2, but there are also commonalities. The regional environmental data here include: a first light intensity, a second light intensity, and a second ambient humidity. The first light intensity here is the light intensity in the target area, while the light intensity mentioned in S2 is the light intensity of the environment outside the target area where the target display screen is located. The second light intensity is: the light intensity facing the target display screen, which is used to indicate whether there is strong light facing the screen; the second ambient humidity is the ambient humidity in the target area, which affects the reflectivity of the target display screen. The difference between the second ambient humidity and the first ambient humidity here is similar to the difference between the first light intensity and the light intensity, so no further details will be given here. After monitoring the regional environmental data corresponding to the target area, data preprocessing is also required, that is, denoising and normalization processing is performed to improve the accuracy of subsequent calculations. The specific implementation steps for executing the dynamic fine-tuning strategy are: S7.1. Fuzzification of Input Parameters: Each input regional environmental data, i.e., the first light intensity, the second light intensity, and the second ambient humidity, is divided into three fuzzy sets: low, medium, and high. Each fuzzy set corresponds to a membership value, indicating the degree to which the current measurement value belongs to that set. Among them, the fuzzification step of the first light intensity in the low interval is: subtract the preset minimum light intensity value from the currently measured first light intensity to obtain the first difference, subtract the preset minimum light intensity value from the preset low interval upper limit light intensity value to obtain the second difference, divide the first difference by the second difference to obtain a ratio result, and the ratio result is the membership value of the current first light intensity belonging to the low fuzzy set; the fuzzification step of the first light intensity in the middle interval is: subtract the preset middle interval center point light intensity value from the currently measured first light intensity, take its absolute value, obtain the third difference, divide the preset middle interval width number by the third difference, and obtain the third difference. The value is divided by 2 to obtain a reference value, the third difference value is divided by the reference value to obtain a quotient value, and the quotient value is subtracted from 1. If the result is less than 0, it is directly taken as 0. The result is the membership value of the current first illumination intensity belonging to the medium fuzzy set; the fuzzification step of the first illumination intensity in the high interval is: subtract the preset maximum illumination intensity value from the currently measured first illumination intensity to obtain a first difference value, subtract the preset high interval lower limit illumination intensity value from the preset maximum illumination intensity value to obtain a second difference value, and divide the first difference by the second difference to obtain a ratio result. The ratio result is the membership value of the current first illumination intensity belonging to the high fuzzy set; The fuzzification step of the second light intensity in the low interval is as follows: subtract the preset minimum light intensity value from the currently measured second light intensity to obtain the first difference value, subtract the preset minimum light intensity value from the preset low interval upper limit light intensity value to obtain the second difference value, divide the first difference value by the second difference value to obtain a ratio result, and the ratio result is the membership value of the current second light intensity to the low fuzzy set; the fuzzification step of the second light intensity in the middle interval is as follows: subtract the preset middle interval center point light intensity value from the currently measured second light intensity, take its absolute value, obtain the third difference value, and divide the preset middle interval width value by the third difference value. Divide by 2 to obtain a reference value, divide the third difference value by the reference value to obtain a quotient, subtract this quotient value from 1, and if the result is less than 0, directly take 0, and the result is the membership value of the current second light intensity belonging to the medium fuzzy set; the fuzzification step of the second light intensity in the high interval is: subtract the currently measured second light intensity from the preset maximum light intensity value to obtain a first difference, subtract the preset high interval lower limit light intensity value from the preset maximum light intensity value to obtain a second difference, divide the first difference by the second difference to obtain a ratio result, and the ratio result is the membership value of the current second light intensity belonging to the high fuzzy set; The fuzzification step of the second ambient humidity in the low interval is as follows: subtract the preset minimum humidity value from the currently measured second ambient humidity to obtain the first difference, subtract the preset minimum humidity value from the preset low interval upper limit humidity value to obtain the second difference, divide the first difference by the second difference to obtain a ratio result, and the ratio result is the membership value of the current second ambient humidity to the low fuzzy set; the fuzzification step of the second ambient humidity in the middle interval is as follows: subtract the preset middle interval center point humidity value from the currently measured second ambient humidity, take its absolute value, obtain the third difference, divide the preset middle interval width value by the third difference. Use 2 to obtain a reference value, divide the third difference by the reference value to obtain a quotient, subtract this quotient from 1, and if the result is less than 0, directly take 0, and the result is the membership value of the current second ambient humidity to the medium fuzzy set; the fuzzification steps of the second ambient humidity in the high interval are: subtract the currently measured second ambient humidity from the preset maximum humidity value to obtain a first difference, subtract the preset high interval lower limit humidity value from the preset maximum humidity value to obtain a second difference, divide the first difference by the second difference to obtain a ratio result, and the ratio result is the membership value of the current second ambient humidity to the high fuzzy set; S7.2. Build a rule engine: Based on historical experience and data, build an if-then fuzzy rule engine for brightness adjustment decisions. The input variables include any of the three fuzzy sets in the environmental data for each area, and the output variables are recommended brightness values, namely, 40% for low brightness, 60% for medium brightness, and 80% for high brightness. Among them, the list of IF-THEN fuzzy rule engines is as follows: If the first light intensity is low, and the second light intensity is low, and the second ambient humidity is low, the brightness should be set to high; if the first light intensity is low, and the second light intensity is low, and the second ambient humidity is medium, the brightness should be set to medium; if the first light intensity is low, and the second light intensity is low, and the second ambient humidity is high, the brightness should be set to medium; if the first light intensity is low, and the second light intensity is medium, and the second ambient humidity is low, the brightness should be set to high; if the first light intensity is low, and the second light intensity is medium, and the second ambient humidity is medium, the brightness The brightness should be set to medium; if the first light intensity is low, the second light intensity is medium, and the second ambient humidity is high, then the brightness should be set to medium; if the first light intensity is low, the second light intensity is high, and the second ambient humidity is low, then the brightness should be set to medium; if the first light intensity is low, the second light intensity is high, and the second ambient humidity is medium, then the brightness should be set to medium; if the first light intensity is low, the second light intensity is high, and the second ambient humidity is high, then the brightness should be set to low; continue to expand to all possible combinations according to similar logic, a total of 27 fuzzy rules; S7.3, fuzzy reasoning: After completing the fuzzification of the input variables, match the current membership with the IF-THEN fuzzy rule engine, calculate the activation degree of each fuzzy rule, and combine the rule weights to obtain the brightness recommendation value; the reasoning steps are as follows: S7.3.1, obtain the membership value of the input variable; for example: for example, the membership value of the first light intensity belonging to low is 0.5, the membership value of the second light intensity belonging to medium is 0.6, and the membership value of the second ambient humidity belonging to high is 0.7; S7.3.2, match each fuzzy rule; for example: for the 6th fuzzy rule in the above S7.2: if the first light intensity is low, the second light intensity is medium, and the second ambient humidity is high, then the activation degree of the fuzzy rule is calculated as: take the minimum value of the three memberships, for example: min(0.5, 0.6, 0.7) = 0.5; S7.3.3, multiply by the weight corresponding to the fuzzy rule; for example: if the weight of the sixth fuzzy rule is 0.6, then the weighted brightness value = 0.5×60% = 30; S7.3.4, repeat the above S7.3.1 to S7.3.3 for all fuzzy rules in the IF-THEN fuzzy rule engine to obtain the weighted brightness value of each fuzzy rule, and accumulate them to obtain the total weighted brightness value. At the same time, accumulate the product of the activation degree and weight of all fuzzy rules to obtain the total weight value; S7.3.5, calculate the recommended brightness value: the recommended brightness value is the total weighted brightness value divided by the total weight value; the unit of the recommended brightness value is percentage, which represents the brightness value that the target display should be set to; the above reasoning process ensures that in a complex and changing environment, multiple factors and their influence can be comprehensively considered, the brightness can be dynamically adjusted, and more accurate and intelligent control can be achieved.

[0036] This solution introduces weighted fuzzy reasoning, which not only improves the accuracy of brightness recommendation output but also enhances the ability to handle the synergistic effects of multiple variables. This solves the problem of traditional brightness adjustment ignoring the mutual influence between parameters and causing misjudgment, and embodies the technological innovation of fuzzy logic + weight optimization fusion control. In S7.3, an IF-THEN fuzzy rule engine is used to combine the membership of the input parameters with the rule weights to derive the brightness recommendation value through the weighted average method. The application of this solution not only considers the status of the environmental data of each area (such as light intensity), but also reflects its actual impact on brightness in the current environment through weight allocation. Subsequently, the environmental change rate is further introduced in the subsequent solution to dynamically correct the weights, enabling the overall solution to respond quickly to sudden environmental changes, significantly improving the stability and accuracy of brightness adjustment.

[0037] S8. After executing the dynamic fine-tuning strategy, the brightness control of the target display screen is completed when the brightness recommendation value is obtained. The change rate of the regional environmental data is detected within the preset acquisition period. The weight of the corresponding regional environmental data in the fuzzy rule is dynamically adjusted according to the detection result. After the weight adjustment, the weighted average method is used to recalculate and obtain the final brightness value corresponding to the dynamic state in the acquisition period. At this time, the target display screen performs adaptive dimming action based on the final brightness value.

[0038] Specific implementation steps

[0039] S8.1. Calculate the rate of change of each regional environmental data detected within a preset collection cycle. For each regional environmental data, calculate the difference between its value and the previous value within each collection cycle, and divide it by the time interval to obtain the rate of change. An example is given: the current first light intensity is I_now, the last time is I_last, and the time interval is Δt (for example, 1 second). The rate of change of the first light intensity is: the numerator is the current value minus the last value, obtaining the first difference; the denominator is the time interval, obtaining the second difference; the result is the rate of change of the first light intensity. The calculation of the second light intensity and the second ambient humidity in the regional environmental data is similar, so they are not described in detail here. S8.2. Set a corresponding change threshold for each regional environmental data (indicating whether a significant change has occurred). The change threshold can be set according to actual needs. In this embodiment, an example is given: the change threshold corresponding to the second light intensity is ≥100 lux / s, and the change threshold corresponding to the second light intensity is 50 lux / s. The change threshold corresponding to the second ambient humidity is ≥5% / s. When the change rate of the environmental data of a certain area exceeds its change threshold, it is considered that the environment in which the environmental data of the area is located is undergoing drastic changes, and the corresponding rule weight should be increased; S8.3. When the environmental data of any area exceeds its corresponding change threshold, the weight adjustment engine is triggered, and the original weight corresponding to the regional environmental data is multiplied by the set ratio value, and the result is the adjusted weight; wherein the set ratio value range is between 1 and 1.5, and the value in this embodiment is 1.2; it should be noted that the maximum weight limit can be set, that is, no more than 1.5 times the original weight, to prevent excessive amplification of the influence of the environmental data of a certain area; S8.4. After completing the weight adjustment, the adjusted weight is used and recalculated using the weighted average method to obtain the final brightness value; the application of the weighted average method is specifically: weighted brightness = Σ(rule brightness recommendation × adjusted weight), total weight = Σ(adjusted weight), final brightness value = weighted brightness / total weight; The practical examples for S8.1 to S8.4 above are as follows: Scenario description: A target display screen installed at the rear of a vehicle enters a tunnel from a brightly lit road, where the environment changes rapidly. The initial regional environmental data (before entering the tunnel) is: first light intensity: 800 lux; second light intensity: 700 lux; second ambient humidity: 40%. Based on the rule matching results obtained in S7.2 above, Rule A states: If the first light intensity is low, the second light intensity is high, and the second ambient humidity is medium, then the brightness should be set to medium, i.e., 40%, with a weight of 0.6. Rule B states: If the first light intensity is low, the second light intensity is high, and the second ambient humidity is medium, then the brightness should be set to medium, i.e., 60%, with a weight of 0.4. Reasoning result: Recommended brightness value = (0.6 × 40 + 0.4 × 60) / (0.6 + 0.4) = 48%. At this point, after entering the tunnel, there is a sudden change, with the second ambient humidity rising to 90%. The corresponding rate of change is: (90 - 40) / 1 = 50% / s; the weight is adjusted as follows: because the change rate corresponding to the first light intensity exceeds 100 lux / s, its corresponding weight is increased from 0.6 to 0.6×1.2=0.72. Therefore, the final brightness value = (0.72×40+0.4×60) / (0.72+0.4)=47.14% (keep two decimal places). In fact, due to the sharp drop in the first light intensity, the brightness should decrease faster. However, because this change trend is recognized, the weight of the relevant rules is promptly increased, making the brightness adjustment more sensitive.

[0040] By dynamically adjusting rule weights based on the rate of environmental change, the system enhances adaptability to sudden environmental changes while improving the real-time and robustness of brightness adjustment, thereby resolving the issues of delayed response and untimely adjustment associated with traditional brightness adjustment. Specifically, by monitoring the rate of change of the first and second light intensities, as well as the second ambient humidity, the system automatically enhances the influence of mutation parameters in the inference process, thereby accelerating brightness adjustment response. For example, when entering a tunnel, the system can quickly identify a sudden drop in light intensity and increase the weight of the relevant rules, allowing the brightness to be adjusted down more quickly to match the new environment. This mechanism shifts brightness adjustment from passive response to active prediction, significantly improving user experience and system intelligence, and is a key component in building a closed-loop brightness control system. To sum up, the overall solution covers scene recognition, environment modeling, strategy formulation, fuzzy reasoning, weight correction, and finally adaptive dimming. Each link is progressive and mutually supportive, jointly promoting the display brightness adjustment from traditional experience-based to data-driven, and from static control to dynamic adaptive control, truly realizing the goal of intelligent display management.

[0041] Example 2: Based on Example 1, this embodiment also provides a multi-scene adaptive dimming system for an LED display screen, which includes: a scene recognition unit: based on the position information corresponding to the target display screen, determining whether it is in a fixed scene or a non-fixed scene; an environment construction unit: when it is in a fixed scene, constructing an environment perception network including several perception nodes, and obtaining corresponding environment parameters according to each perception node; a brightness modeling unit: generating an updated brightness adjustment strategy between multiple environment parameters associated with the perception node according to the environment perception network; a weighted adjustment unit: determining the actual weight between each perception node and multiple environment parameters according to the environment parameters, and adding each actual weight to the corresponding updated brightness adjustment strategy to obtain a corresponding weighted brightness adjustment strategy; a training and verification unit: training a preset brightness control model according to multiple weighted brightness adjustment strategies, and obtaining a standard brightness adjustment strategy. Quasi-brightness control model; regional prediction unit: when in a non-fixed scene, it combines GPS data and map navigation data to predict the type of area to be entered. When the area type matches any type in the regional adjustment library, the preliminary dimming preparation mechanism is triggered; fuzzy reasoning unit: after the preliminary dimming preparation mechanism is completed and the target display enters the target area, it continues to monitor the regional environmental data corresponding to the target area and executes a dynamic fine-tuning strategy: fuzzify the input parameters, construct fuzzy rules and complete fuzzy reasoning to obtain the recommended brightness value; dynamic correction unit: after executing the dynamic fine-tuning strategy, the change rate of the regional environmental data is detected within the preset acquisition period, and the weight of the corresponding regional environmental data in the fuzzy rule is dynamically adjusted according to the detection result. After the weight adjustment, the final brightness value corresponding to the acquisition period is calculated and the adaptive dimming action is completed.

[0042] The effect is described as follows: when a certain parameter suddenly changes, this system can react faster, thereby speeding up the response speed. By dynamically adjusting the weights, the lag or misjudgment that may occur under fixed weights is avoided, and the adjustment accuracy is further improved. It is especially suitable for complex and changeable mobile scenarios, such as entering and exiting tunnels, sudden weather changes, etc., to achieve enhanced environmental adaptability. The entire solution is completely automatically completed based on sensor data, has adaptive capabilities, and does not require manual intervention. In summary, by introducing a dynamic weight correction mechanism based on the change rate of regional environmental data, the intelligence and adaptability of the brightness adjustment system can be significantly improved without relying on user feedback. Combined with the original regional environmental data, the mechanism achieves faster response speed, higher adjustment accuracy and stronger environmental robustness. It is suitable for scenarios such as vehicle-mounted displays, outdoor advertising screens, and intelligent transportation equipment that require stable operation in complex environments.

[0043] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0044] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0045] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A multi-scene adaptive dimming method for LED display screen, characterized in that: The method includes: Based on the location information corresponding to the target display screen, determine whether it is in a fixed scene or a non-fixed scene; When in a fixed scene, an environmental perception network consisting of several perception nodes is constructed, and the corresponding environmental parameters are obtained according to each perception node; According to the environment perception network, an updated brightness adjustment strategy is generated between multiple environment parameters associated with the perception node; Determine the actual weight between each sensing node and the multiple environmental parameters according to the environmental parameters, and add each actual weight to the corresponding updated brightness adjustment strategy to obtain the corresponding weighted brightness adjustment strategy; The preset brightness control model is trained according to multiple weighted brightness adjustment strategies to obtain a standard brightness control model; When in a non-fixed scene, GPS data and map navigation data are combined to predict the type of area to be entered. When the area type matches any type in the area adjustment library, the preliminary dimming preparation mechanism is triggered; After the initial dimming preparation mechanism is completed and the target display enters the target area, the regional environmental data corresponding to the target area is continuously monitored and a dynamic fine-tuning strategy is implemented: input parameters are fuzzified, fuzzy rules are constructed, and fuzzy reasoning is performed to obtain the recommended brightness value; After executing the dynamic fine-tuning strategy, the change rate of the regional environmental data is detected within the preset acquisition period, and the weight of the corresponding regional environmental data in the fuzzy rule is dynamically adjusted according to the detection results. After the weight adjustment, the final brightness value corresponding to the acquisition period is calculated and the adaptive dimming action is completed.

2. The LED display multi-scene adaptive dimming method according to claim 1, characterized in that: In the process of obtaining the corresponding environmental parameters, the data received by each sensing node is preprocessed to screen the environmental parameters; the preprocessing process is as follows: decoding and cleaning, identifying the data packet format, eliminating invalid or abnormal values, and then executing the verification mechanism; the screened environmental parameters are classified and stored by type; the environmental parameters include at least: light intensity and the first environmental humidity.

3. The LED display multi-scene adaptive dimming method according to claim 1, characterized in that: A fusion algorithm is used to analyze the correlation between different environmental parameters, and the algorithms used in the fusion algorithm include at least: statistical correlation analysis, machine learning model and cluster analysis; the analysis process is: using the Pearson correlation coefficient to analyze the linear correlation between two continuous variables, which is used to judge the strength of the correlation between each environmental parameter and the final brightness value; constructing a random forest prediction model, and simultaneously outputting the influence weight of each input variable on the output; dividing the scenes into several types according to different combinations of environmental parameters, including at least: daytime outdoor mode, rainy mode in tunnels and urban road mode; based on the analysis results, formulating several sets of brightness adjustment strategies; wherein the variables are environmental parameters.

4. The LED display multi-scene adaptive dimming method according to claim 1, characterized in that: The process of obtaining the corresponding weighted brightness adjustment strategy is as follows: collecting brightness adjustment records of historical time periods as training data; using a machine learning algorithm to train a brightness control model; wherein the machine learning algorithm is either a neural network or a decision tree; and evaluating the performance of the brightness control model through a cross-validation method.

5. The LED display multi-scene adaptive dimming method according to claim 1, characterized in that: The process of predicting the type of area to be entered is as follows: based on the current location information and movement direction of the target display screen, combined with map navigation data, and using a path planning algorithm to analyze the movement path of the target display screen, the type of area to be entered is predicted.

6. The LED display multi-scene adaptive dimming method according to claim 5, characterized in that: When the preliminary dimming preparation mechanism is triggered, the time required for the real-time target display screen to reach the target area is calculated based on the moving speed of the target display screen and the distance from the target area type. When the required time is equal to the preset time, the preliminary dimming action is performed: the brightness adjustment is completed according to the brightness adjustment gradient within the preset time, so that the final brightness value is adjusted to the set brightness value pre-marked by the target area type at the same time as the target display screen reaches the target area.

7. The LED display multi-scene adaptive dimming method according to claim 1, characterized in that: The regional environmental data includes at least: a first light intensity, a second light intensity and a second ambient humidity; wherein the first light intensity is the light intensity within the target area, the second light intensity is the light intensity facing the target display screen, and the second ambient humidity is the ambient humidity within the target area.

8. The LED display multi-scene adaptive dimming method according to claim 7, characterized in that: The process of fuzzification of input parameters is as follows: each input regional environmental data is divided into three fuzzy sets: low, medium and high; each fuzzy set corresponds to a membership value; The process of constructing fuzzy rules is as follows: building an IF-THEN fuzzy rule engine for brightness adjustment decision-making; wherein, the input variables include at least: any fuzzy set of three fuzzy sets in each regional environmental data; the output variable is the recommended brightness value, which is 40% for low brightness, 60% for medium brightness, and 80% for high brightness; the process of completing fuzzy reasoning is as follows: after completing the fuzzification of the input variables, matching the current membership with the IF-THEN fuzzy rule engine, calculating the activation degree of each fuzzy rule, and combining the rule weights to obtain the recommended brightness value.

9. The LED display multi-scene adaptive dimming method according to claim 1, characterized in that: The process of obtaining the final brightness value is as follows: when calculating the change rate of each regional environmental data detected within the preset acquisition cycle, for each regional environmental data, calculate its difference from the value at the previous moment in each acquisition cycle, and divide it by the time interval to obtain the change rate; set a corresponding change threshold for each regional environmental data. When the change rate of a certain regional environmental data exceeds its change threshold, the weight adjustment engine is triggered, and the original weight corresponding to the regional environmental data is multiplied by the set ratio value. The result is the adjusted weight; the set ratio value range is between 1 and 1.5; after completing the weight adjustment, use the adjusted weight and recalculate using the weighted average method to obtain the final brightness value.

10. A multi-scene adaptive dimming system for LED display screen, characterized by: The system includes: Scene recognition unit: Based on the location information corresponding to the target display screen, it determines whether it is in a fixed scene or a non-fixed scene; Environment construction unit: When in a fixed scene, it builds an environment perception network consisting of several perception nodes and obtains the corresponding environment parameters according to each perception node; Brightness modeling unit: generates an updated brightness adjustment strategy based on multiple environmental parameters associated with the perception node based on the environment perception network; Weighted adjustment unit: determines the actual weight between each sensing node and multiple environmental parameters according to the environmental parameters, and adds each actual weight to the corresponding updated brightness adjustment strategy to obtain the corresponding weighted brightness adjustment strategy; Training and verification unit: trains the preset brightness control model according to multiple weighted brightness adjustment strategies to obtain a standard brightness control model; Regional prediction unit: When in a non-fixed scene, it combines GPS data and map navigation data to predict the type of area it is about to enter. When the area type matches any type in the regional adjustment library, the preliminary dimming preparation mechanism is triggered; Fuzzy reasoning unit: After the initial dimming preparation mechanism is completed and the target display enters the target area, it continues to monitor the regional environmental data corresponding to the target area and implements a dynamic fine-tuning strategy: fuzzifying the input parameters, building fuzzy rules and completing fuzzy reasoning to obtain the recommended brightness value; Dynamic correction unit: After executing the dynamic fine-tuning strategy, it detects the rate of change of regional environmental data within the preset acquisition cycle, dynamically adjusts the weight of the corresponding regional environmental data in the fuzzy rule based on the detection results, and calculates the final brightness value corresponding to the acquisition cycle after the weight adjustment, and completes the adaptive dimming action.

Images