Traffic signal lamp monitoring method and device, electronic equipment and storage medium

Abstract

The invention provides a traffic signal lamp monitoring method and device, electronic equipment and a storage medium, and relates to the technical field of traffic signal lamp monitoring, and the technical scheme is characterized in that the method comprises the steps: obtaining the environment light intensity information and traffic flow information of an intersection where a traffic signal lamp is located; according to the ambient light intensity information and the traffic flow information, calculating a sampling frequency for monitoring the state of a traffic signal lamp; and monitoring the traffic signal lamp according to the calculated sampling frequency. The traffic signal lamp monitoring method and device, the electronic equipment and the storage medium provided by the invention have the advantages that the sampling frequency can be dynamically adjusted according to the environment change, so that the monitoring effect is ensured, and meanwhile, the energy consumption is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of traffic light monitoring, and in particular to a traffic light monitoring method, device, electronic device and storage medium. Background Art

[0002] In modern urban traffic management, real-time monitoring of traffic lights is crucial to maintaining traffic order and improving road traffic efficiency. However, in some urban environments, such as busy intersections with unstable power supply and limited sunlight, traditional traffic light monitoring systems face many challenges.

[0003] Take a city intersection as an example. This intersection is not only an important node of urban traffic, but also a key location for emergency passages. Because it is located on the edge of the power supply network, the intersection often suffers from short-term power outages. To solve the power supply problem, the management department adopted a solar power supply solution. However, the surrounding high-rise buildings have limited sunlight exposure time, which seriously affects the power generation efficiency of solar panels. In this case, how to achieve all-weather and high-efficiency traffic light monitoring under limited power supply conditions has become an urgent problem to be solved.

[0004] Existing traffic light monitoring systems usually use a fixed-frequency continuous sampling method. Although this method performs well in ensuring monitoring accuracy and real-time performance, it has the problem of excessive energy consumption. With limited power supply, it is difficult to meet the needs of all-weather monitoring. In addition, traditional systems lack the ability to adapt to environmental changes and cannot dynamically adjust the working mode according to actual conditions, resulting in high power consumption during low-demand periods, causing energy waste.

[0005] In view of the above problems, the present application proposes a new technical solution. Summary of the invention

[0006] The purpose of the present application is to provide a traffic light monitoring method, device, electronic device and storage medium, which have the advantage of being able to dynamically adjust the sampling frequency according to environmental changes, thereby reducing energy consumption while ensuring the monitoring effect.

[0007] This application provides a traffic light monitoring method, and the technical solution is as follows: The method comprises: obtaining ambient light intensity information and traffic flow information of the intersection where the traffic light is located; calculating a sampling frequency for monitoring the state of the traffic light according to the ambient light intensity information and traffic flow information; and monitoring the traffic light according to the calculated sampling frequency.

[0008] Furthermore, the present application also proposes that the step of calculating the sampling frequency for monitoring the state of a traffic light based on the ambient light intensity information and the traffic flow information includes: making the calculation weight of the traffic flow information greater than the calculation weight of the ambient light intensity information; calculating the sampling frequency for monitoring the state of a traffic light based on the traffic flow information, the larger the traffic flow information is, the higher the calculated sampling frequency is; calculating the sampling frequency for monitoring the state of a traffic light based on the ambient light intensity information, the darker the ambient light intensity information is, the higher the calculated sampling frequency is.

[0009] Furthermore, the present application also proposes that the step of calculating the sampling frequency for monitoring the status of traffic lights based on the ambient light intensity information and traffic flow information includes: obtaining the ambient light intensity information and traffic flow information within a preset time period, and predicting the ambient light intensity and traffic flow for the next time period; calculating the sampling frequency of the traffic light status for the next time period based on the predicted ambient light intensity and traffic flow for the next time period, so as to achieve an early response to environmental changes.

[0010] Furthermore, the present application also proposes that the step of calculating the sampling frequency of the traffic light status in the next time period based on the predicted ambient light intensity and traffic flow in the next time period includes: obtaining the difference between the predicted ambient light intensity and traffic flow in the next time period and the ambient light intensity and traffic flow in the current time period; judging whether the difference exceeds a preset threshold; when it exceeds the preset threshold, increasing the buffer sampling frequency on the basis of the calculated sampling frequency for the next time period to avoid monitoring blind spots caused by sudden environmental changes.

[0011] Further, the present application also proposes that the step of calculating the traffic signal status sampling frequency for the next time period according to the predicted environmental light intensity and traffic flow in the next time period includes: obtaining the environmental light intensities L(t), L(t - 1) and traffic flows F(t), F(t - 1) in the current time period t and the previous time period t - 1, as well as the predicted environmental light intensity L(t + 1) and traffic flow F(t + 1) in the next time period t + 1; calculating the environmental change difference value D, where D = |L(t + 1) - L(t)| + |F(t + 1) - F(t)|; calculating the environmental change trend T, where T = sign((L(t + 1) - L(t)) * (L(t) - L(t - 1)) + (F(t + 1) - F(t)) * (F(t) - F(t - 1))); calculating the environmental change rate R, where R = (|L(t + 1) - L(t)| / |L(t) - L(t - 1)| + |F(t + 1) - F(t)| / |F(t) - F(t - 1)|) / 2; determining the gain coefficient f(R) according to the environmental change rate R, when R > R1, f(R) = α; when R2 < R ≤ R1, f(R) = β; when R ≤ R2, f(R) = γ; where R1 and R2 are preset change rate thresholds, and R1 > R2, α, β, γ are preset gain coefficients, and α > β > γ; calculating the traffic signal status sampling frequency S for the next time period, where S = S_base + k1 * max(0, D - D_threshold) * (1 + k2 * T) * f(R); where, S_base is the basic sampling frequency, k1 is the first adjustment coefficient, k2 is the second adjustment coefficient, D_threshold is the difference value threshold; ensuring that the calculated sampling frequency S satisfies the constraint condition S_min ≤ S ≤ S_max, where S_min is the minimum sampling frequency and S_max is the maximum sampling frequency; taking the calculated sampling frequency S as the traffic signal status sampling frequency for the next time period.

[0012] Further, the present application also proposes that the step of obtaining the environmental light intensity information and traffic flow information within a preset time period and predicting the environmental light intensity and traffic flow in the next time period includes: according to the preset time period, obtaining the historical environmental light intensity information and historical traffic flow information corresponding to the next time period to be predicted in the preset time period; obtaining the mutation points based on the historical environmental light intensity information and historical traffic flow information and their corresponding time information; predicting the environmental light intensity and traffic flow in the next time period according to the time information of the detected light intensity and traffic flow mutation points.

[0013] Furthermore, the present application also proposes that the step of predicting the ambient light intensity and traffic flow in the next time period based on the time information of the detected light intensity and traffic flow mutation points includes: counting the frequency of occurrence of light intensity and traffic flow mutation points detected in multiple historical time periods within a preset time period; based on the frequency of occurrence of the mutation points within the time period, calculating the probability of a mutation in light intensity and traffic flow in the next time period; and adjusting the predicted ambient light intensity and traffic flow values ​​for the next time period based on the probability of a mutation in light intensity and traffic flow in the next time period.

[0014] Furthermore, the present application also proposes a traffic light monitoring device, comprising: an acquisition module, used to obtain ambient light intensity information and traffic flow information at the intersection where the traffic light is located; a calculation module, used to calculate the sampling frequency for monitoring the status of the traffic light based on the ambient light intensity information and traffic flow information; and a monitoring module, used to monitor the traffic light according to the calculated sampling frequency.

[0015] Furthermore, the present application also proposes an electronic device, including a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the above method are executed.

[0016] Furthermore, the present application also proposes a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above method are executed.

[0017] From the above, it can be seen that the traffic light monitoring method, device, electronic device and storage medium provided by the present application dynamically adjust the sampling frequency according to the ambient light intensity and traffic flow, thereby reducing energy consumption while ensuring the monitoring effect. It has the advantage of being able to dynamically adjust the sampling frequency according to environmental changes, thereby reducing energy consumption while ensuring the monitoring effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flow chart of a traffic light monitoring method provided in this application.

[0019] Figure 2 A schematic diagram of the structure of a traffic light monitoring method provided in this application.

[0020] In the figure: 210, acquisition module; 220, calculation module; 230, monitoring module. DETAILED DESCRIPTION

[0021] The technical solutions in the present application will be clearly and completely described below in conjunction with the drawings in the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0022] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0023] At busy urban intersections, traffic light monitoring systems face a trade-off between energy consumption and monitoring efficiency. Although the existing fixed-frequency continuous sampling method can ensure monitoring accuracy and real-time performance, it is difficult to meet all-weather monitoring needs when power supply is limited. This method cannot dynamically adjust the working mode according to actual environmental changes, resulting in high power consumption during low-demand periods, causing energy waste. At the same time, the system lacks the ability to respond quickly to emergencies, making it difficult to quickly switch to high-frequency sampling mode to meet emergency monitoring needs while ensuring daily low-power operation.

[0024] In this regard, refer to Figure 1 , the present application proposes a traffic signal light monitoring method, comprising: S110, obtaining ambient light intensity information and traffic flow information at the intersection where the traffic light is located; S120, calculating a sampling frequency for monitoring the state of a traffic light according to the ambient light intensity information and the traffic flow information; S130: Monitor the traffic light according to the calculated sampling frequency.

[0025] The ambient light intensity information refers to the light intensity data at the intersection where the traffic light is located. Specifically, a light sensor can be used to achieve real-time measurement. This information is used to evaluate the impact of current light conditions on traffic light monitoring.

[0026] Traffic flow information refers to the number and density of vehicles passing through the intersection. Specifically, vehicle detectors or cameras can be used to implement statistics. This information is used to determine the current traffic conditions and monitoring needs of the intersection.

[0027] The sampling frequency refers to the time interval at which the monitoring system samples the status of traffic lights. Specifically, an adjustable timer can be used to achieve dynamic control. The adjustment of this frequency directly affects the energy consumption and monitoring accuracy of the system.

[0028] In some specific implementations, the monitoring system may be a camera, which is aimed at a traffic light to monitor the state of the traffic light based on image recognition, and the traffic flow information may be collected simultaneously based on the camera.

[0029] The core innovation of this application is to propose a traffic light monitoring method that dynamically adjusts the sampling frequency based on ambient light intensity and traffic flow. This method obtains environmental data in real time and calculates the optimal sampling frequency, thereby significantly reducing system energy consumption while ensuring the monitoring effect.

[0030] The working principle of this application can be described in detail as follows: First, obtain ambient light intensity information and traffic flow information in real time.

[0031] Next, the acquired data is input into a preset calculation algorithm, which takes into account the impact of light intensity on the difficulty of signal light recognition and the impact of traffic flow on the importance of monitoring. For example, when the light is dim, the sampling frequency needs to be increased to ensure accurate recognition of the signal light status; while when the traffic flow is large, the sampling frequency needs to be increased to capture the changes of the signal light in time.

[0032] The calculated sampling frequency is then applied to the monitoring system. The system dynamically adjusts its sampling time interval based on the calculated results. For example, when the light suddenly dims or the traffic volume increases sharply, the system can quickly increase the sampling frequency to ensure monitoring quality.

[0033] Ambient light intensity and traffic flow are selected as key parameters because these two factors directly affect the importance of signal light monitoring. By dynamically adjusting the sampling frequency, the system can maximize energy savings while ensuring monitoring results, thereby extending the system's working time and improving overall efficiency.

[0034] As a preferred implementation, the present application can be implemented at a busy intersection. The intersection is equipped with traffic lights in four directions, and a high-definition camera is installed in each direction for signal light status recognition, and a light sensor is used to measure the ambient light intensity, and the camera is used to count traffic flow.

[0035] The system's central processing unit is set to acquire environmental data every 5 seconds. For example, at a certain moment, the light intensity measured by the light sensor is 500 lux, and the number of vehicles passing through in 5 minutes is 300. Based on the preset algorithm, the optimal sampling frequency can be calculated to be 2Hz, taking into account historical data and current trends.

[0036] The system then adjusts the sampling frequency to 2Hz, which means that the signal light status is sampled every 0.5 seconds. The camera captures images at this frequency and uses image processing algorithms to identify the current status of the signal light (red, yellow, green). If a change in the status of the signal light is detected, the system will immediately record and report it.

[0037] During the next monitoring process, if the light intensity suddenly drops to 100 lux (for example, the weather turns cloudy), the system will quickly recalculate the sampling frequency. The new calculation result may increase the sampling frequency to 4Hz to cope with the increased recognition difficulty caused by dimming light.

[0038] In the solution of the present application, the sampling frequency of calculating the ambient light intensity information and the traffic flow information is mainly to reduce power consumption and optimize monitoring efficiency.

[0039] Among them, the relationship between ambient light intensity and sampling frequency is reflected in the visual characteristics of the human eye and the visibility of traffic lights. When the ambient light intensity is high, the brightness of the traffic lights is relatively not so prominent due to the strong ambient light. In order to make the human eye clearly identify, the traffic lights usually work at a higher brightness. At this time, even if the sampling frequency is slightly lower, due to the high brightness and good contrast of the traffic lights, it is not easy for the human eye to miss the change of the traffic light status.

[0040] In low ambient light, traffic lights are relatively conspicuous even at low brightness due to the low ambient light. However, if a traffic light fails (such as sudden brightness reduction, flickering, etc.), it may be more difficult for the human eye to detect it at low brightness. Therefore, when the ambient light is dim, appropriately increasing the sampling frequency can capture the abnormal state of the traffic light more timely.

[0041] Regarding traffic flow and monitoring needs, usually during the day, traffic flow is larger, and the real-time and reliability requirements of traffic lights are higher. Although the ambient light intensity is high during the day, combined with traffic flow information, a higher sampling frequency may still be required to ensure timely detection of signal light failures and avoid traffic congestion or accidents.

[0042] Traffic volume is smaller at night, and the monitoring pressure on traffic lights is relatively small, so the sampling frequency can be appropriately reduced.

[0043] Through this dynamic adjustment, the system can reduce the sampling frequency to save energy when the light is sufficient and the traffic volume is light, and increase the sampling frequency to ensure the monitoring quality when the light is insufficient or the traffic volume is heavy. This adaptive method not only improves the energy utilization efficiency, but also enhances the system's ability to adapt to complex environmental changes.

[0044] In some of the above embodiments of the present application, it is proposed to calculate the sampling frequency for monitoring the status of traffic lights based on the ambient light intensity information and the traffic flow information in order to dynamically adjust the sampling frequency. However, in this process, there is still room for optimization in how to reasonably allocate the weights of the ambient light intensity information and the traffic flow information, and how to calculate the sampling frequency based on these two types of information.

[0045] In this regard, the present application further proposes the steps of calculating the sampling frequency for monitoring the status of traffic lights based on ambient light intensity information and traffic flow information, including: making the calculation weight of traffic flow information greater than the calculation weight of ambient light intensity information; calculating the sampling frequency for monitoring the status of traffic lights based on traffic flow information, the greater the traffic flow information, the higher the calculated sampling frequency; calculating the sampling frequency for monitoring the status of traffic lights based on ambient light intensity information, the darker the ambient light intensity information, the higher the calculated sampling frequency.

[0046] The technical solution of this application effectively solves the problem of how to optimize the sampling frequency calculation by reasonably allocating the weights of ambient light intensity information and traffic flow information and establishing the relationship between these two types of information and the sampling frequency. Specifically, the solution gives priority to traffic flow information, which meets the needs of actual traffic management because traffic flow directly reflects the busyness of the intersection.

[0047] In practical applications, the calculation weight of traffic flow information can be set to 0.7, while the calculation weight of ambient light intensity information can be set to 0.3. This weight distribution ensures that when calculating the sampling frequency, the influence of traffic flow information is greater, thereby better reflecting the needs of actual traffic conditions.

[0048] When calculating the sampling frequency based on traffic flow information, a linear or nonlinear mapping relationship can be used. For example, a basic sampling frequency can be set, and then the sampling frequency can be linearly increased according to the increase in traffic flow. Specifically, the formula can be used: sampling frequency = basic frequency + k* traffic flow, where k is a preset coefficient. In this way, when the traffic flow increases, the sampling frequency will increase accordingly, ensuring that during busy traffic periods, the system can monitor the status of traffic lights more frequently, improving the real-time and accuracy of monitoring.

[0049] For ambient light intensity information, the sampling frequency can be calculated using an inverse proportional relationship. For example, the formula can be used: sampling frequency = c / (light intensity + ε), where c is a constant and ε is a small positive number to avoid division by zero. This calculation method ensures that the sampling frequency will be increased accordingly in low light conditions to cope with possible recognition difficulties and improve monitoring quality.

[0050] These two calculation methods can be combined by weighted averaging to form the final sampling frequency calculation formula: final sampling frequency = 0.7*(base frequency + k*traffic flow) + 0.3*(c / (light intensity + ε)).

[0051] In this way, the technical solution of the present application realizes dynamic adjustment of the sampling frequency. During busy traffic periods, the system will increase the sampling frequency to increase monitoring accuracy and real-time performance. When the light conditions are poor, the system will also appropriately increase the sampling frequency to ensure monitoring quality. In other cases, the system can reduce the sampling frequency to save energy. This dynamic adjustment mechanism achieves a balance between monitoring accuracy, real-time response capability, and energy efficiency.

[0052] Specifically, by combining ambient light intensity and traffic flow information, this method can intelligently adjust the sampling frequency: In low light and high traffic conditions: maintain a high sampling frequency to ensure timely detection of faults.

[0053] In strong light and high flow: increase the sampling frequency appropriately.

[0054] In low light and low traffic conditions: the sampling frequency can be appropriately reduced, because the real-time requirements for the traffic lights are not high at this time, and reducing power consumption is given priority. However, the reduction should not be too large, and a certain monitoring frequency must still be guaranteed to prevent hidden faults (such as slight flickering) in the traffic lights from being discovered in time.

[0055] In strong light and low traffic: the sampling frequency can be reduced because the real-time requirements for traffic lights are not high at this time, and the visibility of traffic lights is high under strong light, so faults are easy to be discovered.

[0056] That is, in some preferred embodiments, in various situations of the calculated sampling frequency, weak light, large flow> strong light, large flow> weak light, small flow.

[0057] In some of the above embodiments of the present application, it is proposed to dynamically calculate the traffic light status sampling frequency based on the ambient light intensity information and the traffic flow information to adapt to the monitoring needs under different environmental conditions. However, in this process, there may be a problem of not responding to environmental changes in a timely manner. Since environmental conditions may change suddenly, such as sudden changes in weather or sudden increases in traffic flow, relying solely on the current environmental information to calculate the sampling frequency may cause the monitoring system to be unable to adjust in time, thereby affecting the monitoring effect.

[0058] In this regard, the present application further proposes the steps of calculating the sampling frequency for monitoring the status of traffic lights based on the ambient light intensity information and traffic flow information, including: obtaining the ambient light intensity information and traffic flow information within a preset time period, and predicting the ambient light intensity and traffic flow for the next time period; calculating the sampling frequency of the traffic light status for the next time period based on the predicted ambient light intensity and traffic flow for the next time period, so as to achieve an early response to environmental changes.

[0059] This application introduces a prediction mechanism, which obtains ambient light intensity information and traffic flow information within a preset time period and predicts the environmental conditions of the next time period to calculate the sampling frequency of the next time period. This prediction mechanism can respond to possible environmental changes in advance, making the monitoring system more flexible and efficient.

[0060] Specifically, we first obtain environmental data over a period of time, including light intensity and traffic flow. These historical data provide a basis for predicting future environmental changes. Then, we use these data to predict environmental conditions in the next time period.

[0061] The forecasting mechanism can be implemented in many ways. For example, you can use the moving average method, which calculates the average of the past few time periods to predict the value of the next time period. Another method is exponential smoothing, which gives more weight to the most recent data to better capture the latest trends.

[0062] After obtaining the prediction results, the system calculates the corresponding sampling frequency based on the predicted ambient light intensity and traffic flow in the next time period. This step can be achieved by setting a series of thresholds and corresponding frequencies. For example, when the predicted light intensity is lower than a certain threshold or the traffic flow is higher than a certain threshold, the system will increase the sampling frequency accordingly.

[0063] Some of the above embodiments provide the ability to dynamically adjust the sampling frequency according to the current environmental conditions, while the prediction mechanism of this embodiment further enhances the forward-looking nature of the system. By combining these two aspects, it is possible to prepare for upcoming changes while maintaining a response to the current environment. This synergy enables greater adaptability and stability in the face of complex and changing environments.

[0064] As a preferred embodiment, the technical solution of the present application can be implemented at an intersection. The intersection is equipped with multiple traffic lights and sensors for monitoring ambient light intensity and traffic flow. The system first sets a preset time period, such as 30 minutes. During the 30 minutes, the system records ambient light intensity and traffic flow data every 5 minutes.

[0065] Specifically, light intensity can be measured in lumens (lm), which may range from 100lm (at dusk) to 100,000lm (midday on a sunny day). Traffic flow can be measured in vehicles per hour, which may range from 50 vehicles per hour (late at night) to 2,000 vehicles per hour (peak hour).

[0066] Use this data to predict the environmental conditions for the next 30 minutes. The prediction algorithm can use exponential smoothing to give more weight to recent data. For example, if the average light intensity in the current 30 minutes is 5000lm and the traffic flow is 1000 vehicles / hour, and the prediction algorithm takes into account the upcoming dusk period, it may predict that the average light intensity in the next 30 minutes will be 3000lm and the traffic flow will be 1200 vehicles / hour (taking into account the possible evening rush hour).

[0067] Based on these predicted values, the system calculates the sampling frequency for the next time period. Assume that the system sets the following rules: When the predicted light intensity is <2000lm or the traffic flow is >1500 vehicles / hour, the sampling frequency is set to 2 times per second.

[0068] When 2000lm≤predicted light intensity<5000lm and 1000 vehicles / hour<traffic flow≤1500 vehicles / hour, the sampling frequency is set to once per second.

[0069] In other cases, the sampling frequency is set to once every 2 seconds.

[0070] In this example, based on the predicted light intensity of 3000lm and traffic flow of 1200 vehicles / hour, the system will set the sampling frequency to 1 time per second for the next 30 minutes. In this way, the system can adjust to a higher sampling frequency in advance before the light starts to dim and traffic flow increases, ensuring accurate monitoring when environmental conditions change.

[0071] In this way, energy usage is optimized while ensuring monitoring quality, providing an efficient solution for scenarios with unstable or limited power supply.

[0072] In some of the above embodiments of the present application, it is proposed to calculate the traffic light state sampling frequency of the next time period according to the predicted ambient light intensity and traffic flow of the next time period to achieve early response to environmental changes. However, in this process, if the environment changes suddenly, the accuracy and real-time performance of monitoring may be affected.

[0073] In this regard, the present application further proposes obtaining the difference between the predicted ambient light intensity and traffic flow in the next time period and the ambient light intensity and traffic flow in the current time period; judging whether the difference exceeds a preset threshold; when the preset threshold is exceeded, increasing the buffer sampling frequency based on the calculated sampling frequency for the next time period to avoid monitoring blind spots caused by sudden environmental changes.

[0074] This application identifies possible environmental mutations by obtaining the difference between the predicted ambient light intensity and traffic flow in the next time period and the current time period, and determining whether the difference exceeds a preset threshold. When the difference exceeds the preset threshold, the system will increase the buffer sampling frequency based on the originally calculated sampling frequency for the next time period. This method can increase the sampling frequency when the environment changes significantly, thereby avoiding monitoring blind spots caused by environmental mutations.

[0075] Specifically, the difference value can be calculated in a variety of ways. For example, the difference between the ambient light intensity and the traffic flow can be calculated separately, and then the two differences are added or the maximum value is taken as the final difference value. The preset threshold value can be set based on historical data analysis or expert experience.

[0076] The step of determining whether the difference value exceeds a preset threshold value can be further refined. For example, multiple threshold levels can be set, and different degrees of frequency adjustment can be adopted according to the interval in which the difference value falls. This hierarchical processing can achieve more precise sampling frequency control.

[0077] The buffer sampling frequency that is increased when a preset threshold is exceeded can be proportional to the degree of the difference value. For example, a base increment can be set, and then the buffer sampling frequency can be increased linearly or nonlinearly according to the degree to which the difference value exceeds the threshold. This dynamic adjustment method can better adapt to environmental changes of varying degrees.

[0078] By dynamically adjusting the sampling frequency, the system can respond to environmental changes in a timely manner while maintaining low power consumption, improving the accuracy and real-time performance of monitoring. When the environment is relatively stable, the system maintains a lower sampling frequency to save energy; when it predicts that the environment may change significantly, the system will increase the sampling frequency in advance to ensure that important changes in the status of traffic lights are not missed.

[0079] As a preferred implementation, the technical solution of the present application can be implemented in the following specific scenarios: Suppose at an intersection, the system predicts that the average ambient light intensity in the next 10 minutes will be 500 lux and the average traffic flow will be 200 vehicles / hour, while the actual data for the current period are 450 lux and 180 vehicles / hour respectively.

[0080] The system first calculates the difference value. Assuming a simple linear weighting method, the weights of ambient light intensity and traffic flow are 0.4 and 0.6 respectively. Then the difference value is calculated as follows: Difference = 0.4*|500-450| / 450+0.6*|200-180| / 180≈0.16 Assuming that the preset threshold is 0.15, the system determines that the difference value exceeds the preset threshold.

[0081] Next, the system calculates the buffer sampling frequency. Assuming a linear increase method, the base increment is 10%, and the increment is 1% for every 1% above the threshold, with a maximum of no more than 50%. In this example, the difference value exceeds the threshold by about 6.7%, and assuming that the predicted sampling frequency for the next time period is 1 time per second, the system decides to increase the sampling frequency by 16.7%.

[0082] Therefore, the system adjusts the sampling frequency of the next time period to 1.167 times per second (116.7% of the original frequency). This adjustment can increase the sampling frequency in advance when large environmental changes are expected, effectively avoiding possible monitoring blind spots without excessively increasing the system burden.

[0083] In this way, the technical solution of the present application can effectively solve the problem of monitoring blind spots caused by sudden changes in the environment, improve the accuracy and real-time performance of the traffic light monitoring system, and maintain the low power consumption characteristics of the system. This adaptive mechanism enables the system to maintain efficient and stable operation in various complex environments, providing more reliable technical support for traffic management.

[0084] In some of the above embodiments of the present application, it is proposed to calculate the sampling frequency of the traffic light status in the next time period based on the predicted ambient light intensity and traffic flow in the next time period to achieve early response to environmental changes. However, in this process, calculating the sampling frequency based solely on the predicted value may not fully consider the rate and trend of environmental changes, resulting in the adjustment of the sampling frequency being not precise and flexible enough. In addition, the lack of a rapid response mechanism to sudden environmental changes may result in monitoring blind spots when the environment changes dramatically.

[0085] In response to this, the present application further proposes to obtain the ambient light intensities L(t) and L(t - 1) and traffic flows F(t) and F(t - 1) in the current time period t and the previous time period t - 1, as well as the predicted ambient light intensity L(t + 1) and traffic flow F(t + 1) in the next time period t + 1; Calculate the environmental change difference value D, where D = |L(t + 1) - L(t)| + |F(t + 1) - F(t)|; Calculate the environmental change trend T, where T = sign((L(t + 1) - L(t)) * (L(t) - L(t - 1)) + (F(t + 1) - F(t)) * (F(t) - F(t - 1))); Calculate the environmental change rate R, where R = (|L(t + 1) - L(t)| / |L(t) - L(t - 1)| + |F(t + 1) - F(t)| / |F(t) - F(t - 1)|) / 2; Determine the gain coefficient f(R) according to the environmental change rate R. When R > R1, f(R) = α; when R2 < R ≤ R1, f(R) = β; when R ≤ R2, f(R) = γ; where R1 and R2 are preset change rate thresholds, and R1 > R2, and α, β, and γ are preset gain coefficients, and α > β > γ; Calculate the traffic signal state sampling frequency S in the next time period, where S = S_base + k1 * max(0, D - D_threshold) * (1 + k2 * T) * f(R); Among them, S_base is the base sampling frequency, k1 is the first adjustment coefficient, k2 is the second adjustment coefficient, and D_threshold is the difference value threshold; ensure that the calculated sampling frequency S satisfies the constraint condition S_min ≤ S ≤ S_max, where S_min is the minimum sampling frequency and S_max is the maximum sampling frequency; use the calculated sampling frequency S as the traffic signal state sampling frequency in the next time period.

[0086] The technical solution proposed by the present application introduces multiple key features, and each feature has its specific implementation method and function.

[0087] The calculation of the environmental change difference value D adopts the method of the sum of absolute values. There can be various variants of this method. For example, the Euclidean distance or Manhattan distance can be used to calculate the difference value. The advantage of choosing the sum of absolute values is that the calculation is simple and it can reflect the overall amplitude of environmental changes.

[0088] The sign function is used to calculate the environmental change trend T. Other methods can be considered here, such as using slope or weighted average to calculate the trend. The advantage of the sign function is that it can clearly distinguish between positive, negative and no obvious trend, which is conducive to the subsequent adjustment of the sampling frequency.

[0089] The calculation of the environmental change rate R uses the average value of the relative change rate. Methods such as weighted average or exponential smoothing can be considered to calculate the rate. The advantage of the current method is that it can simultaneously consider the change rate of light intensity and traffic flow, and reduce the impact of extreme values ​​of a single factor on the result through averaging.

[0090] The gain coefficient f(R) is determined in the form of a piecewise function. This method can set more thresholds and corresponding gain coefficients according to actual needs. The advantage of the piecewise function is that the gain coefficient can be flexibly adjusted according to the different degrees of environmental change rate, thereby achieving refined control of the sampling frequency.

[0091] The calculation formula of the sampling frequency S takes into account the base frequency, environmental change differences, trends, and rates. This formula can be further optimized, such as by introducing nonlinear functions or adaptive weights. The advantage of the current formula is that the relationship between the various factors is clear, making it easy to understand and adjust.

[0092] The setting of the constraints S_min and S_max ensures that the sampling frequency is within a reasonable range. It is possible to consider dynamically adjusting these two thresholds based on the system hardware performance and actual application scenarios. The benefit of this constraint mechanism is that it can prevent the sampling frequency from being too high, causing the system to be overloaded, or too low, causing insufficient monitoring accuracy.

[0093] The difference value D of environmental change reflects the magnitude of environmental change, the trend T indicates the direction of change, and the rate R indicates the speed of change. These three factors together determine the direction and magnitude of the adjustment of the sampling frequency. The gain coefficient f(R) provides different degrees of adjustment strength according to the different rates, allowing the system to respond more sensitively to rapidly changing environments. Finally, the calculation formula of the sampling frequency integrates these factors together to achieve comprehensive adjustment of the sampling frequency.

[0094] In actual applications, appropriate parameter values ​​can be set according to specific scenarios. For example, for intersections with large light changes, the weight of light intensity can be set higher; for intersections with frequent traffic flow fluctuations, the thresholds of R1 and R2 can be lowered to make the system more sensitive to rate changes. The specific parameter settings need to be determined through actual testing and optimization.

[0095] The technical solution of the present application realizes the refined dynamic adjustment of the sampling frequency by introducing three dimensions: environmental change difference value, change trend and change rate. This method can more accurately capture the characteristics of environmental changes, thereby realizing precise control of the sampling frequency.

[0096] Specifically, the difference value D of environmental changes reflects the magnitude of environmental changes, allowing the system to respond to significant environmental changes. The change trend T indicates the direction of environmental changes, allowing the system to predict the sustainability of environmental changes and adjust the sampling strategy in advance. The change rate R further refines the characteristics of environmental changes, allowing the system to distinguish between gradual and sudden changes and make corresponding adjustments.

[0097] The introduction of the gain factor f(R) enables the system to flexibly adjust the sampling frequency according to the rate of environmental change. This mechanism can improve the system's response speed when the environment changes rapidly, while avoiding over-adjustment when the environment is relatively stable.

[0098] The sampling frequency calculation formula S=S_base+k1*max(0,D-D_threshold)*(1+k2*T)*f(R) takes all the above factors into consideration. The basic sampling frequency S_base ensures the basic monitoring capability of the system. The adjustment is triggered only when the difference value exceeds the threshold D_threshold, avoiding excessive response to small changes. The trend T affects the adjustment amplitude through the (1+k2*T) term, enabling the system to make forward-looking adjustments based on the changing trend. Finally, the gain factor f(R) further adjusts the sampling frequency according to the rate of change, achieving an accurate response to the speed of environmental changes.

[0099] The constraint condition S_min≤S≤S_max ensures that the sampling frequency always remains within the acceptable range of the system, avoiding system problems caused by too high or too low sampling frequency.

[0100] Compared with simple prediction value calculation, this method that comprehensively considers multiple factors can more comprehensively capture the characteristics of environmental changes, thereby achieving more accurate and flexible sampling frequency adjustment. The system can quickly increase the sampling frequency when the environment changes dramatically to reduce the occurrence of monitoring blind spots; at the same time, it can appropriately reduce the sampling frequency when the environment is relatively stable to save system resources.

[0101] As a preferred implementation, the following specific parameter settings may be considered: Set the base sampling frequency S_base=1Hz, the minimum sampling frequency S_min=0.5Hz, and the maximum sampling frequency S_max=10Hz.

[0102] The environmental light intensity L is measured in lux, and the traffic flow F is measured in the number of vehicles passing through per minute.

[0103] Set the difference value threshold D_threshold = 100. This value indicates that when the environmental light intensity changes by 100 lux or the traffic flow changes by 100 vehicles / minute, the sampling frequency adjustment is triggered.

[0104] Set the first adjustment coefficient k1 = 0.01 and the second adjustment coefficient k2 = 0.5.

[0105] For the gain coefficient f(R), set R1 = 2, R2 = 1, α = 2, β = 1.5, γ = 1. This means that when the environmental change rate R > 2, the maximum gain coefficient 2 is adopted; when 1 < R ≤ 2, the medium gain coefficient 1.5 is adopted; when R ≤ 1, the minimum gain coefficient 1 is adopted.

[0106] Suppose at a certain moment, the system obtains the following data: L(t - 1) = 1000 lux, F(t - 1) = 200 vehicles / minute; L(t) = 1200 lux, F(t) = 250 vehicles / minute; The predicted L(t + 1) = 1500 lux, F(t + 1) = 320 vehicles / minute.

[0107] Based on this data, the system performs the following calculations: Calculate the environmental change difference value D = |1500 - 1200| + |320 - 250| = 370; Calculate the environmental change trend T = sign((1500 - 1200)*(1200 - 1000) + (320 - 250)*(250 - 200)) = 1; Calculate the environmental change rate R = (|1500 - 1200| / |1200 - 200| + |320 - 250| / |250 - 200|) / 2 = 1.75; Since 1 < R ≤ 2, determine the gain coefficient f(R) = β = 1.5; Calculate the sampling frequency S = 1 + 0.01*max(0, 370 - 100)*(1 + 0.5*1)*1.5 = 5.55 Hz; The final sampling frequency S = 5.55 Hz, which satisfies the constraint condition of S_min ≤ S ≤ S_max.

[0108] Through this example, we can see how the system dynamically adjusts the sampling frequency according to the difference value, trend, and rate of environmental changes. In this scenario, because the environmental changes are significant (the difference value far exceeds the threshold), and there is a positive trend, and the rate of change is fast, the system increases the sampling frequency from the basic 1Hz to 5.55Hz to better capture the rapid changes in the environment.

[0109] This dynamic adjustment mechanism can effectively reduce the average power consumption of the system while ensuring monitoring accuracy. When the environment is relatively stable, the system can maintain a low sampling frequency to save energy; when the environment changes rapidly, the system can increase the sampling frequency in time to ensure that important environmental changes are not missed. This adaptive capability makes the system particularly suitable for use in scenarios with limited power supply, such as traffic light monitoring systems powered by solar energy. At the same time, by precisely controlling the sampling frequency, the system can respond quickly in emergency situations and meet the needs of being an important emergency channel in the city.

[0110] In some of the above embodiments of the present application, it is proposed to obtain ambient light intensity information and traffic flow information within a preset time period, and predict the ambient light intensity and traffic flow of the next time period to calculate the traffic light state sampling frequency of the next time period, so as to achieve early response to environmental changes. However, in this process, relying solely on current and predicted environmental information may not accurately reflect the regularity and periodicity of environmental changes, thereby affecting the accuracy of the prediction and the rationality of the sampling frequency.

[0111] In this regard, the present application further proposes to obtain historical ambient light intensity information and historical traffic flow information corresponding to the next time period to be predicted in a preset time period according to a preset time period; obtain mutation points based on the historical ambient light intensity information and historical traffic flow information and their corresponding time information; and predict the ambient light intensity and traffic flow for the next time period based on the time information of the detected light intensity and traffic flow mutation points.

[0112] This technical solution improves the accuracy of environmental change prediction by introducing historical data analysis and mutation point detection. Specifically, the solution first obtains historical data according to a preset time period, which helps to capture the periodic laws of environmental changes. The preset time period can be one day, one week, or one month, etc., depending on the characteristics of the intersection where the traffic light is located. For example, for an intersection with obvious differences in traffic patterns on weekdays and weekends, one week can be selected as the preset time period.

[0113] When obtaining historical data, the system can adopt multiple methods. One method is to directly extract data from the database for the same period in the past. Another method is to use sliding window technology to dynamically obtain data from the last few cycles. This ensures that the historical data used can reflect long-term trends and capture recent changes.

[0114] Next, analyze the mutation points and their time information in the historical data. A variety of algorithms can be used to detect mutation points, such as statistical methods (such as the CUSUM algorithm), which can identify key nodes and abnormal situations of environmental changes. For example, the system may detect a mutation point in traffic flow between 7 and 9 a.m. every morning, which may correspond to the beginning of the morning rush hour.

[0115] The mutation point detection is not limited to a single variable, but can also consider the joint distribution of ambient light intensity and traffic flow. This can capture the mutual influence between the two factors, such as the combined effect of dimming light and reduced traffic flow caused by rainy days.

[0116] Based on the detected mutation point time information, the system can more accurately predict the environmental conditions in the next time period. The prediction method can use time series analysis technology to predict through patterns in historical data and combined with mutation point information.

[0117] This approach enables better responses to periodic changes and emergencies. For example, if historical data shows that traffic flow often changes between 4 and 6 p.m. every Friday, the system can adjust the sampling frequency in advance to cope with possible traffic congestion. Similarly, if a sudden change pattern of light intensity under specific weather conditions is detected, the system can adjust the prediction strategy accordingly.

[0118] In addition, the method can be combined with real-time data to further improve the accuracy of the forecast. For example, the system can compare the differences between the real-time observed data and the historical patterns, and if significant deviations are found, the forecast results can be adjusted in a timely manner.

[0119] By adopting this prediction method based on historical data and mutation point analysis, the technical solution can more accurately predict the ambient light intensity and traffic flow in the next time period. This improvement allows the system to adjust the sampling frequency more accurately, thereby optimizing energy use while ensuring monitoring accuracy. Especially in the face of periodic changes and emergencies, this method can provide more reliable prediction results, thereby supporting the system to make more intelligent decisions, improving the adaptability and efficiency of the entire traffic light monitoring system.

[0120] In order to better understand the technical solution of the present application, a specific embodiment is provided below: Assume that a traffic light monitoring system is deployed at a busy city intersection. The system is set to a preset time period of one week because the traffic pattern at this intersection is significantly different on weekdays and weekends. The system collects ambient light intensity and traffic flow data every 15 minutes.

[0121] First, the system obtains historical data. For example, if it is currently 3 p.m. on Wednesday, the system extracts data from 3 p.m. to 4 p.m. on Wednesdays for the past four weeks. This data includes ambient light intensity (in lux) and traffic volume (in vehicles / hour) every 15 minutes.

[0122] Next, detect mutation points. Suppose the system detects that in the past four weeks of data, there are frequent mutation points in traffic volume between 3:30 and 3:45 pm on Wednesdays, with an average increase of about 30%. At the same time, the system also notices that on cloudy days, the light intensity during this time period will suddenly drop by about 20%.

[0123] Based on this information, predict the environmental conditions for the next hour, taking into account the periodic patterns in the historical data, the detected mutation points, and the current weather forecast.

[0124] Assume that the current time is 3:00 PM on Wednesday and the weather forecast shows that it may be cloudy. The system predicts that between 3:30 and 3:45, the traffic volume may increase from 1000 vehicles / hour to 1300 vehicles / hour, while the light intensity may decrease from 5000 lux to 4000 lux.

[0125] Based on this prediction, it was decided to gradually increase the sampling frequency from once per minute to once every 15 seconds starting at 3:25 to cope with the upcoming traffic rush and light changes. In this way, the system can capture the state changes of traffic lights in time when environmental changes occur, while keeping a lower sampling frequency at other time periods to save energy.

[0126] In this way, the technical solution of the present application can significantly reduce the average power consumption of the system while ensuring monitoring accuracy, while still being able to respond to emergencies and environmental changes in a timely manner. This not only prolongs the system's battery life, but also improves its adaptability and reliability in complex environments.

[0127] In some of the above embodiments of the present application, it is proposed to predict the ambient light intensity and traffic flow of the next time period based on the time information of the detected light intensity and traffic flow mutation point to calculate the traffic light status sampling frequency of the next time period. However, in this process, relying solely on the mutation point information of a single time period may result in the prediction results being inaccurate and unstable. The changes in ambient light intensity and traffic flow may have certain periodicity and regularity, and the data of a single time period cannot fully reflect these long-term change trends. In addition, emergencies or abnormal situations may cause abnormal fluctuations in the data of a certain time period. If the prediction is based only on the data of this time period, it may cause unnecessary fluctuations in the sampling frequency, affecting the stability and energy efficiency of the system.

[0128] In this regard, the present application further proposes to statistically calculate the frequency of occurrence of light intensity and traffic flow mutation points detected in multiple historical time periods within a preset time period; based on the frequency of occurrence of mutation points within the time period, calculate the probability of sudden changes in light intensity and traffic flow in the next time period; and adjust the predicted ambient light intensity and traffic flow values ​​for the next time period based on the probability of sudden changes in light intensity and traffic flow in the next time period.

[0129] This application introduces a method of statistical analysis and probability prediction, which predicts future environmental changes by analyzing data from multiple historical time periods. Specifically, firstly, the data from multiple historical time periods within a preset time period are counted, and the frequency of occurrence of sudden changes in light intensity and traffic flow is counted. This step can capture the long-term trends and periodic patterns of environmental changes. Different time periods can be selected for statistics, such as 24 hours, 7 days or 30 days, to adapt to the laws of environmental changes at different scales.

[0130] Furthermore, based on the frequency of occurrence of mutation points obtained by statistics, the probability of mutation occurring in the next time period is calculated, for example, using a simple frequency statistics method or a more complex time series analysis.

[0131] Finally, according to the calculated mutation probability, the predicted ambient light intensity and traffic flow values ​​for the next time period are adjusted. This step can make corresponding corrections to the prediction results according to the magnitude of the mutation probability. The adjustment can be linear or nonlinear. For example, when the mutation probability is high, the predicted value can be adjusted toward the historical mutation value, while when the mutation probability is low, the original predicted value can be kept unchanged or only slightly adjusted.

[0132] This method effectively solves the limitation of single time period data prediction by introducing statistical analysis and probabilistic prediction. It can better capture the long-term trend and periodic pattern of environmental changes, improve the accuracy and stability of prediction. At the same time, it can better deal with emergencies and abnormal situations, and avoid prediction deviation caused by abnormal data in a single time period.

[0133] By combining historical data analysis with real-time data collection, environmental changes can be predicted more accurately, allowing the sampling frequency to be adjusted more reasonably. This not only improves monitoring accuracy, but also optimizes energy efficiency. For example, if a sudden change is predicted in the next time period, the system can increase the sampling frequency in advance to ensure that important traffic light status changes are not missed. Conversely, in periods when the environment is predicted to be more stable, the system can appropriately reduce the sampling frequency to save energy.

[0134] In the specific implementation process, the sliding time window method can be used to process historical data. For example, a 30-day sliding window is set, and the statistical data is updated once a day. This ensures that the latest historical data is always used, while capturing the change patterns of a longer time span. In addition, a weight factor can be introduced so that the more recent data has a greater impact on the prediction results, thereby increasing the model's sensitivity to recent changes.

[0135] To further improve the accuracy of the prediction, the ambient light intensity and traffic flow can be modeled separately. Since these two factors may have different changing patterns and periodicities, modeling them separately can capture their respective characteristics more accurately. For example, the ambient light intensity may be mainly affected by the natural sunlight cycle, while the traffic flow may be more affected by factors such as weekdays and holidays.

[0136] In practical applications, multiple thresholds can be set to refine the impact of mutation probability. For example, when the mutation probability is less than 20%, keep the original prediction value unchanged; when the mutation probability is between 20% and 50%, make a small adjustment to the prediction value; when the mutation probability is between 50% and 80%, make a moderate adjustment; when the mutation probability is higher than 80%, make a large adjustment. This hierarchical adjustment strategy can make the system more flexible to respond to different degrees of environmental change possibilities.

[0137] As a preferred implementation, a feedback mechanism can be set up in the system. By comparing the difference between the actual observed environmental changes and the predicted results, the system can dynamically adjust the parameters of the prediction model, such as the mutation point determination threshold, probability calculation method, etc. This adaptive mechanism can enable the system to maintain an efficient operating state under different environmental conditions.

[0138] In a specific embodiment, assuming that at a city intersection, the preset time period is 7 days. The system collects ambient light intensity and traffic flow data every 10 minutes and stores these data in a local database. At 12:00 midnight every day, the system will conduct a comprehensive analysis of the data of the past 7 days and count the frequency of occurrence of mutation points.

[0139] The criteria for determining a mutation point can be set as: the ambient light intensity changes by more than 50 lux, or the traffic flow changes by more than 100 vehicles per hour. The system records the specific time when each mutation point occurs. For example, in the past 7 days, the system may observe the following patterns: Sudden change in ambient light intensity: There are 6 mutations between 5:30 and 6:00 in the morning (corresponding to sunrise); There are 5 mutations between 18:00 and 18:30 in the evening (corresponding to sunset); Traffic flow changes: There are 5 sudden changes between 7:30 and 8:00 on weekday mornings (corresponding to the morning rush hour); There are 5 sudden changes between 17:30 and 18:00 on weekday afternoons (corresponding to the evening peak); There are 2 sudden changes between 11:30 and 12:00 noon on weekends (corresponding to the weekend shopping peak); Based on these statistics, the system can calculate the probability of a sudden change in the next time period. For example, if the next time period is 7:45 am on a weekday, the system will calculate that the probability of a sudden change in traffic flow in this time period is 71.4% (5 times / 7 days).

[0140] Based on this probability, the predicted traffic flow value is adjusted accordingly. Assuming that the originally predicted traffic flow is 300 vehicles per hour, considering the 71.4% probability of mutation, the system may adjust the predicted value to 450 vehicles per hour. This adjusted predicted value will be used to calculate the traffic light status sampling frequency for the next time period.

[0141] This method allows the system to more accurately predict environmental changes and allocate energy resources more rationally. When a sudden change is predicted, the system will increase the sampling frequency to ensure that important traffic signal changes are not missed. When the environment is predicted to be relatively stable, the system can reduce the sampling frequency to save energy.

[0142] Second, refer to Figure 2 , the present application further proposes a traffic signal light monitoring device, comprising: An acquisition module 210 is used to acquire ambient light intensity information and traffic flow information at the intersection where the traffic light is located; A calculation module 220, for calculating a sampling frequency for monitoring the state of a traffic light according to the ambient light intensity information and the traffic flow information; The monitoring module 230 is used to monitor the traffic light according to the calculated sampling frequency.

[0143] By dynamically adjusting the sampling frequency according to the ambient light intensity and traffic flow, energy consumption is reduced while ensuring the monitoring effect. It has the advantage of being able to dynamically adjust the sampling frequency according to environmental changes, thereby reducing energy consumption while ensuring the monitoring effect.

[0144] In addition, in some preferred embodiments, a traffic signal light monitoring device proposed in the present application can perform any one of the steps in the above method.

[0145] In a third aspect, the present application also provides an electronic device, including a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the above method are executed.

[0146] Through the above technical solution, the processor and the memory are interconnected and communicate with each other through a communication bus and / or other forms of connection mechanisms (not marked), and the memory stores computer-readable instructions executable by the processor. When the electronic device is running, the processor executes the computer-readable instructions to execute the method in any optional implementation of the above embodiment to achieve the following functions: obtain ambient light intensity information and traffic flow information at the intersection where the traffic light is located; calculate the sampling frequency for monitoring the status of the traffic light based on the ambient light intensity information and the traffic flow information; and monitor the traffic light based on the calculated sampling frequency.

[0147] In a fourth aspect, the present application also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method are executed.

[0148] Through the above technical solution, when the computer program is executed by the processor, the method in any optional implementation method of the above embodiments is executed to achieve the following functions: obtaining the ambient light intensity information and traffic flow information of the intersection where the traffic light is located; calculating the sampling frequency for monitoring the status of the traffic light based on the ambient light intensity information and traffic flow information; and monitoring the traffic light based on the calculated sampling frequency.

[0149] Among them, the computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0150] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0151] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0152] Furthermore, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0153] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A traffic light monitoring method, characterized in that: include: Obtain the ambient light intensity information and traffic flow information at the intersection where the traffic light is located; Calculating a sampling frequency for monitoring the state of a traffic light according to the ambient light intensity information and the traffic flow information; The traffic light is monitored according to the calculated sampling frequency.

2. A traffic signal light monitoring method according to claim 1, characterized in that: The step of calculating the sampling frequency for monitoring the state of the traffic light according to the ambient light intensity information and the traffic flow information comprises: Making the calculation weight of the traffic flow information greater than the calculation weight of the ambient light intensity information; Calculate the sampling frequency for monitoring the state of the traffic light according to the traffic flow information, the greater the traffic flow information, the higher the calculated sampling frequency; The sampling frequency for monitoring the state of the traffic light is calculated according to the ambient light intensity information. The darker the ambient light intensity information is, the higher the calculated sampling frequency is.

3. A traffic signal light monitoring method according to claim 1, characterized in that: The step of calculating the sampling frequency for monitoring the state of the traffic light according to the ambient light intensity information and the traffic flow information comprises: Obtaining ambient light intensity information and traffic flow information within a preset time period, and predicting ambient light intensity and traffic flow in the next time period; According to the predicted ambient light intensity and traffic flow in the next time period, the traffic light status sampling frequency in the next time period is calculated to achieve early response to environmental changes.

4. A traffic signal light monitoring method according to claim 3, characterized in that: The step of calculating the traffic light state sampling frequency for the next time period according to the predicted ambient light intensity and traffic flow for the next time period comprises: Obtain the difference between the predicted ambient light intensity and traffic flow in the next time period and the ambient light intensity and traffic flow in the current time period; Determining whether the difference value exceeds a preset threshold; When the preset threshold is exceeded, the buffer sampling frequency is increased based on the calculated sampling frequency for the next time period to avoid monitoring blind spots caused by sudden changes in the environment.

5. A traffic signal light monitoring method according to claim 3, characterized in that: The step of calculating the traffic light state sampling frequency for the next time period according to the predicted ambient light intensity and traffic flow for the next time period comprises: Obtain the ambient light intensity L(t), L(t-1) and traffic flow F(t), F(t-1) of the current time period t and the previous time period t-1, as well as the predicted ambient light intensity L(t+1) and traffic flow F(t+1) of the next time period t+1; Calculate the environmental change difference value D, where D = |L(t+1)-L(t)|+|F(t+1)-F(t)|; Calculate the environmental change trend T, where T=sign((L(t+1)-L(t))*(L(t)-L(t-1))+(F(t+1)-F(t))*(F(t)-F(t-1))); Calculate the environmental change rate R, where R=(|L(t+1)-L(t)| / |L(t)-L(t-1)|+|F(t+1)-F(t)| / |F(t)-F(t-1)|) / 2; Determine the gain coefficient f(R) according to the environmental change rate R. When R > R1, f(R) = α; when R2 < R ≤ R1, f(R) = β; when R ≤ R2, f(R) = γ; where R1 and R2 are preset change rate thresholds, and R1 > R2, and α, β, and γ are preset gain coefficients, and α > β > γ; Calculate the sampling frequency S of the traffic signal light state in the next time period, where S = S_base + k1 * max(0, D - D_threshold) * (1 + k2 * T) * f(R); where S_base is the basic sampling frequency, k1 is the first adjustment coefficient, k2 is the second adjustment coefficient, and D_threshold is the difference value threshold; Ensure that the calculated sampling frequency S satisfies the constraint condition S_min ≤ S ≤ S_max, where S_min is the minimum sampling frequency and S_max is the maximum sampling frequency; Use the calculated sampling frequency S as the sampling frequency of the traffic signal light state in the next time period.

6. A traffic signal light monitoring method according to claim 3, characterized in that: The steps of obtaining the environmental light intensity information and traffic flow information within a preset time period and predicting the environmental light intensity and traffic flow in the next time period include: According to the preset time period, obtain the historical environmental light intensity information and historical traffic flow information corresponding to the next time period to be predicted within the preset time period; Obtain the mutation points based on the historical environmental light intensity information and historical traffic flow information and their corresponding time information; Predict the environmental light intensity and traffic flow in the next time period according to the time information of the detected light intensity and traffic flow mutation points.

7. A traffic signal light monitoring method according to claim 6, characterized in that: The steps of predicting the environmental light intensity and traffic flow in the next time period according to the time information of the detected light intensity and traffic flow mutation points include: Statistically calculate the occurrence frequency of the detected light intensity and traffic flow mutation points within the time period in multiple historical time periods within the preset time period; Based on the occurrence frequency of the mutation points within the time period, calculate the probability of the occurrence of light intensity and traffic flow mutations in the next time period; According to the probability of the occurrence of light intensity and traffic flow mutations in the next time period, adjust the predicted environmental light intensity and traffic flow values in the next time period.

8. A traffic light monitoring device, characterized in that: Include: An acquisition module for acquiring the environmental light intensity information and traffic flow information of the intersection where the traffic signal light is located; A calculation module for calculating the sampling frequency for monitoring the traffic signal light state according to the environmental light intensity information and traffic flow information; A monitoring module for monitoring the traffic signal light according to the calculated sampling frequency.

9. An electronic device, characterized in that: Includes a processor and a memory, and the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in any one of the methods as claimed in claims 1 - 7 are run.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps in any one of the methods as claimed in claims 1 - 7 are run.

Images