Intelligent adaptive adjustment method and device for traffic signal lamp

Abstract

The invention provides a traffic signal lamp intelligent adaptive adjustment method and device, and relates to the field of traffic signal lamp intelligent control. The method comprises the following steps: obtaining individual attribute data and real-time waiting time length data corresponding to waiting tourists in a waiting area; obtaining traffic flow data corresponding to the running vehicles in the waiting area; on the basis of the individual attribute data and the real-time waiting time length data, traffic urgency features corresponding to the waiting tourists are constructed; on the basis of the traffic flow information, traffic priority features corresponding to the running vehicles are constructed; based on the above characteristics and the real-time waiting time length data, a traffic pressure value of the waiting tourist is output through a signal lamp intelligent adjustment model; and adjusting traffic lights in the waiting area based on the traffic pressure value. According to the invention, the problem that in special urban spaces with high tourist density, such as tourist attractions or cultural blocks, the traditional traffic signal lamp control mode can reduce the guiding capability, so that the occurrence probability of traffic accidents is increased is solved.

Description

Technical Field

[0001] The present application relates to the field of intelligent control of traffic lights, and in particular to a method and device for intelligent adaptive adjustment of traffic lights. Background Art

[0002] Intelligent adaptive adjustment technology for traffic lights is evolving towards high integration, multi-dimensional linkage, and global optimization. Its development trend is not only reflected in the intelligent replacement of traditional control strategies, but also in the all-round leap in system architecture, decision-making mechanisms, data dimensions, and behavioral modeling.

[0003] Currently, most traffic lights use a "maximum and minimum green light interval control + fixed priority rule" approach to adjust signal timing. This involves setting minimum and maximum green light duration boundaries at intersections based on empirical parameters. If the queue length or number of vehicles in a certain direction exceeds a threshold, the green light duration in that direction is extended. Otherwise, the signal phase is switched according to the rotation cycle, and in the event of a conflict between pedestrians and vehicles, motor vehicles are prioritized. For example, by default, motor vehicles going straight have priority over pedestrian green lights. This approach is adaptable to a certain extent in conventional urban environments with a stable traffic structure and a single traffic entity. However, in special urban spaces with a high tourist density, such as tourist attractions or cultural districts, this traditional control method suffers from a reduced guidance capability, leading to an increased probability of traffic accidents.

[0004] Therefore, there is an urgent need for an intelligent adaptive adjustment method and device for traffic lights. Summary of the Invention

[0005] The present application provides a method and device for intelligent adaptive adjustment of traffic lights, which solves the problem that in special urban spaces with high tourist density, such as tourist attractions or cultural blocks, traditional traffic light control methods have a reduced guidance ability, thereby increasing the probability of traffic accidents.

[0006] In a first aspect of the present application, a method for intelligent adaptive adjustment of traffic lights is provided, the method comprising: obtaining individual attribute data and real-time waiting time data corresponding to tourists waiting in a waiting area; obtaining traffic flow data corresponding to vehicles traveling in the waiting area; constructing a traffic urgency feature corresponding to waiting tourists based on the individual attribute data and real-time waiting time data, the traffic urgency feature including tourist identity features and tourist behavior pattern features; constructing a traffic priority feature corresponding to traveling vehicles based on traffic flow information, the traffic priority feature including lane traffic density features and vehicle type composition features in the current time period; inputting the traffic urgency feature, traffic priority feature and real-time waiting time data into an intelligent traffic light adjustment model, and outputting a traffic pressure value for waiting tourists through the intelligent traffic light adjustment model; and adjusting the traffic lights in the waiting area based on the traffic pressure value.

[0007] Optionally, individual attribute data corresponding to tourists waiting in the waiting area are obtained, specifically including: collecting initial individual attribute data corresponding to waiting tourists through a multimodal sensing device, the multimodal sensing device including a visible light camera, an infrared thermal imaging device, and a voice recognition microphone array; inputting the initial individual attribute data into a tourist attribute classification model, and outputting identity feature label data corresponding to different waiting tourists based on the tourist attribute classification model; constructing an identity feature data set corresponding to the classified identity feature label data, and normalizing the identity feature data set; and using the normalized identity feature data set as the individual attribute data.

[0008] Optionally, based on individual attribute data, a passage urgency feature corresponding to waiting tourists is constructed: based on identity feature label data, a passage urgency feature judgment index is constructed, and the passage urgency feature judgment index includes travel status index, language index and behavior status index; based on the passage urgency feature judgment index and combined with real-time waiting time data, a passage urgency feature corresponding to each waiting tourist is constructed.

[0009] Optionally, based on the traffic flow information, a traffic priority feature corresponding to the moving vehicles is constructed, specifically including: obtaining vehicle quantity data, driving speed data and spacing data of the lane corresponding to the waiting area in the current time period; constructing a lane traffic density feature based on the vehicle quantity data, driving speed data and spacing data; obtaining vehicle type data of the lane corresponding to the waiting area in the current time period; and constructing a vehicle type composition feature based on the vehicle type data.

[0010] Optionally, before inputting the traffic urgency features, traffic priority features and real-time waiting time data into the traffic light intelligent space intelligent adjustment model and outputting the traffic pressure value of waiting tourists through the traffic light intelligent space intelligent adjustment model, it is necessary to construct a traffic light intelligent space intelligent adjustment model, specifically including: uniformly encoding the traffic urgency features, traffic priority features and real-time waiting time data; extracting the deep correlation features between the traffic urgency features, traffic priority features and real-time waiting time data after uniform encoding through a multi-layer neural network structure; performing multi-feature dimensional fusion of the deep correlation features by introducing a weighted attention mechanism; and constructing a traffic light intelligent space intelligent adjustment model based on the deep correlation features after multi-feature dimensional fusion.

[0011] Optionally, the traffic pressure value of waiting tourists is outputted through the intelligent air conditioning model of the traffic light, specifically including: calculating the traffic pressure value through the following formula: in, is the flow pressure value, represents the Sigmoid function, Intelligently adjust the weight of the output layer of the traffic light model. is the bias term, represents the natural exponential function.

[0012] Optionally, the traffic lights in the waiting area are adjusted based on the traffic pressure value, specifically including: obtaining a target traffic pressure level interval corresponding to the traffic pressure value; based on the traffic pressure level, obtaining an intelligent adjustment scheme corresponding to the pressure level interval in a preset traffic light adjustment database, the preset traffic light adjustment database being used to store the correspondence between the pressure level interval and the intelligent adjustment scheme; based on the intelligent adjustment scheme, adjusting the traffic lights in the waiting area.

[0013] In a second aspect of the present application, a traffic light intelligent adaptive adjustment device is provided, the device comprising an acquisition module and a processing module, wherein: The acquisition module is used to obtain individual attribute data and real-time waiting time data corresponding to tourists waiting in the waiting area; and obtain traffic flow data corresponding to vehicles traveling in the waiting area.

[0014] The processing module is used to construct the traffic urgency characteristics corresponding to waiting tourists based on individual attribute data and real-time waiting time data. The traffic urgency characteristics include tourist identity characteristics and tourist behavior pattern characteristics; based on traffic flow information, the traffic priority characteristics corresponding to moving vehicles are constructed. The traffic priority characteristics include lane traffic density characteristics and vehicle type composition characteristics in the current time period; the traffic urgency characteristics, traffic priority characteristics and real-time waiting time data are input into the traffic light intelligent space intelligent adjustment model, and the traffic pressure value of waiting tourists is output through the traffic light intelligent space intelligent adjustment model; the traffic lights in the waiting area are adjusted based on the traffic pressure value.

[0015] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device performs any of the methods described above.

[0016] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to perform any of the above methods.

[0017] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. Obtain individual attribute data and real-time waiting time data for tourists waiting in the waiting area; obtain traffic flow data for vehicles traveling in the waiting area; construct a traffic urgency feature for the waiting tourists based on the individual attribute data and real-time waiting time data. This traffic urgency feature includes characteristics of the tourist's identity and behavior patterns; construct a traffic priority feature for the traveling vehicles based on the traffic flow information. This traffic priority feature includes characteristics of lane traffic density and vehicle type composition within the current time period; input the traffic urgency, traffic priority, and real-time waiting time data into the intelligent traffic light adjustment model, which then outputs a traffic pressure value for the waiting tourists; and adjust the traffic lights in the waiting area based on this pressure value, improving the signal light control system's responsiveness to changes in tourist traffic demand and matching the traffic window. This effectively addresses the problem of traditional traffic light control methods that suffer from reduced guidance capabilities and increased probability of traffic accidents in special urban spaces with high tourist density, such as tourist attractions or cultural districts.

[0018] 2. Based on identity feature label data, construct a traffic urgency feature judgment indicator. The traffic urgency feature judgment indicator includes a travel status indicator, a language indicator, and a behavioral status indicator. Based on the traffic urgency feature judgment indicator and combined with real-time waiting time data, construct the traffic urgency feature corresponding to each waiting tourist, so that the traffic urgency feature is used as the core input variable in the intelligent signal control model, effectively representing the actual demand intensity of tourists for the right of way in the current traffic environment, and providing a behavioral driving basis for the subsequent calculation of traffic pressure values, thereby improving the signal system's ability to fine-tune the traffic response of non-local groups in complex scenarios.

[0019] 3. Obtain the target traffic pressure level interval corresponding to the traffic pressure value; based on the traffic pressure level, obtain the intelligent adjustment scheme corresponding to the pressure level interval from the preset traffic light adjustment database, which is used to store the correspondence between the pressure level interval and the intelligent adjustment scheme; based on the intelligent adjustment scheme, adjust the traffic lights in the waiting area, so that the signal control strategy can achieve refined and differentiated response according to the traffic demand intensity of the tourist group. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of a method for intelligent adaptive adjustment of traffic lights provided in an embodiment of the present application; Figure 2 This is a module diagram of an intelligent adaptive adjustment device for traffic lights provided in an embodiment of the present application; Figure 3 This is a structural diagram of an electronic device provided in an embodiment of the present application.

[0021] Explanation of the reference numerals: 21, acquisition module; 22, processing module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0023] The terms used in the following examples of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification of the present application, the singular expressions "a", "an", "said", "above", "the", and "this" are intended to include plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more of the listed items.

[0024] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of the technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of this application, unless otherwise specified, "plurality" means two or more.

[0025] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0026] Please refer to Figure 1 , which shows a flow chart of a method for intelligent adaptive adjustment of traffic lights provided in an embodiment of the present application, the flow chart mainly includes the following steps: S101 to S106.

[0027] Step S101, obtaining individual attribute data and real-time waiting time data corresponding to tourists waiting in the waiting area.

[0028] Specifically, in unique urban spaces such as tourist attractions or cultural districts, characterized by high visitor density, uncertain visitor mobility patterns, and significant differences in cognitive abilities, traffic signal timing struggles to fully adapt to the behavioral characteristics of non-local tourists. This is particularly true while waiting for traffic signals. Language barriers, poor signal recognition, and distracted behavior often lead tourists to misjudge signal states, cross traffic boundaries, or remain in traffic areas for extended periods, leading to traffic conflicts and safety hazards. To address these technical issues, we first collect individual attribute data and real-time wait time data for tourists in the waiting area. Tourists typically include individual travelers, self-guided tour groups, group tours, elderly tourists, families, and foreign-speaking tourists, each with significant differences in behavior patterns, travel intentions, and signal comprehension. Individual attribute data is primarily used to distinguish these different types of tourists, while real-time wait time data records the total wait time from the moment they enter the waiting area at a traffic signal to the current moment.

[0029] In a possible implementation, step S101 further includes: collecting initial individual attribute data corresponding to waiting tourists through a multimodal sensing device, the multimodal sensing device including a visible light camera, an infrared thermal imaging device, and a voice recognition microphone array; inputting the initial individual attribute data into a tourist attribute classification model, and outputting identity feature label data corresponding to different waiting tourists based on the tourist attribute classification model; constructing an identity feature data set corresponding to the classified identity feature label data, and normalizing the identity feature data set; and using the normalized identity feature data set as individual attribute data.

[0030] Specifically, visible light cameras are first deployed at the boundaries and core observation points of the waiting area to collect the appearance images and behavioral dynamics of tourists, and identify basic identity attributes such as the tourists' gender, estimated age, whether they are traveling with a tour group, and whether they are accompanied by children or elderly people. Infrared thermal imaging equipment is used to obtain the tourists' degree of stillness, movement mode (standing still, walking, wandering), population density and individual heat source aggregation trend to assist in determining whether they are currently in a behavioral state such as queuing, taking pictures, waiting for passage or stopping for sightseeing. At the same time, a voice recognition microphone array is used to detect whether tourists are using voice guide equipment or interacting with companions, and further extract language information, communication intentions and indicators related to signal recognition ability.

[0031] The initial perception data collected by the above-mentioned different modal devices is used as input and imported into the pre-trained tourist attribute classification model. The model can perform feature fusion and label prediction on multimodal input based on the convolutional neural network and attention mechanism structure, and output identity feature label data including "whether the tourist is a non-local", "whether the tourist is a foreign language tourist", "whether the tourist is a family group", "whether the tourist has difficulty in understanding traffic rules", "whether the tourist is carrying a tour guide device", "whether the tourist is taking photos or in a non-traffic behavior", etc.

[0032] The above-mentioned label data are constructed into a structured identity feature dataset, and a data normalization processing mechanism is introduced to encode and standardize different label attributes according to a unified numerical range to improve the consistency and expression efficiency of subsequent model inputs, while providing a comparable input basis for multi-factor weight operations in the signal control model.

[0033] Finally, the normalized identity feature dataset is used as individual attribute data and input into the travel urgency calculation model together with the real-time waiting time data, thereby achieving detailed modeling of tourists' travel needs and dynamic support for traffic light adjustment strategies.

[0034] Step S102: Obtain traffic flow data corresponding to vehicles traveling in the waiting area.

[0035] Specifically, to accurately characterize and dynamically perceive traffic flow characteristics at typical pedestrian-vehicle intersections in tourist attractions or cultural districts, multi-source traffic detection equipment must be deployed in each lane of the target intersection to obtain high-temporal-resolution vehicle traffic data. Traffic detection equipment includes geomagnetic vehicle detectors, video recognition cameras, millimeter-wave radar sensors, and license plate recognition units, enabling non-contact, real-time acquisition of vehicle status without disrupting traffic. For example, geomagnetic detectors combined with millimeter-wave radars can be used to calculate the number of vehicles passing through each detection point per unit time, including the total number of vehicles passing per minute in a single lane and the simultaneous passing rate of vehicles in adjacent lanes. Radar equipment and inter-frame image difference methods can be used to measure the speed of each vehicle, reflecting the instantaneous congestion state and queuing trends of the road section. Video recognition and rear-end detection algorithms can be used to calculate the distance between consecutive vehicles, including metrics such as average headway, maximum and minimum headway, and headway fluctuation frequency, for modeling traffic flow continuity and density.

[0036] At the same time, vehicle type data is extracted based on the license plate recognition system and vehicle contour recognition model, including the category distribution of small passenger cars, buses, tourist buses, taxis, freight vehicles and emergency vehicles, and the vehicle type composition characteristics are constructed by counting the traffic proportions of each category of vehicles in the current period to reflect the structural characteristics of vehicle traffic in this time period.

[0037] In addition, the above-mentioned traffic flow data can be timestamped and synchronized by combining the historical timing records of the traffic controller and the dynamic signal execution data, so that the various types of traffic flow data obtained have time continuity and stage markings, providing high-reliability and high-precision basic data support for the subsequent construction of traffic priority features and adjustment of signal control strategies.

[0038] Step S103 : constructing a passage urgency feature corresponding to the waiting tourists based on the individual attribute data and the real-time waiting time data. The passage urgency feature includes the tourist identity feature and the tourist behavior pattern feature.

[0039] Specifically, the urgency of travel feature is a comprehensive indicator that characterizes the intensity of a waiting tourist's subjective demand for the current right of way and the urgency of their behavioral response in a specific traffic environment. It dynamically reflects their drive to respond to signal adjustments and is one of the core inputs for building intelligent traffic signal control models. Tourist identity features, such as whether they are non-locals, whether they are traveling in a group, whether they use guides, whether they have language barriers, and whether they are elderly or traveling with children, reflect their cognitive abilities and understanding of traffic rules. Behavior features, such as their current behavior (e.g., stopping, taking photos, walking, talking), their gaze toward the signal, whether they actively pay attention to the signal, their positional stability, and behavioral shifts during the waiting period, are used to identify whether tourists are ready to pass. Combined with real-time waiting time data, these identity and behavioral features are weighted and fused to construct a set of urgency features. This feature reflects the urgency of travel for individuals or groups within the current cycle, providing a behavioral driver for signal duration adjustment and priority determination.

[0040] In a possible implementation, step S103 also includes: constructing a passage urgency feature judgment index based on identity feature label data, the passage urgency feature judgment index including a travel status index, a language index, and a behavior status index; based on the passage urgency feature judgment index and combined with real-time waiting time data, constructing a passage urgency feature corresponding to each waiting tourist.

[0041] Specifically, based on the normalized identity feature tag data obtained in step S102, several urgency indicators are extracted and constructed to quantitatively assess the subjective intensity and behavioral activity of tourists' travel needs. The travel status indicator reflects the consistency and initiative of tourists' travel intentions by determining whether they are traveling in a group, a parent-child pair, an accompanying elderly group, or an independent traveler. The language indicator identifies the language used by tourists based on voice recognition or guide device information and determines whether it is a local language, reflecting their ability to understand traffic instructions and their reliance on system guidance. The behavioral status indicator, based on video behavior recognition and thermal imaging dynamic data, identifies whether tourists are currently stationary, walking, taking photos, looking around, or frequently turning, and is used to assess whether they are about to pass or are anxious. Next, the above indicators, combined with tourists' real-time waiting time data and their cumulative dwell time, are input into a travel urgency quantification function. This function incorporates a behavioral decay weighting mechanism. When waiting time continues to increase while tourists remain highly active or exhibit behavioral uncertainty, their travel urgency output value is increased. Meanwhile, for short waiting times or those exhibiting low travel intent behaviors such as taking photos or chatting, their travel urgency is assigned a low weight. Ultimately, a travel urgency feature is generated for each tourist. This feature, represented as a numerical vector, encompasses the tourist's subjective driving force for travel during the current cycle and their response priority to traffic system adjustments. This feature is used to subsequently calculate travel pressure values ​​and output signal control strategies.

[0042] Step S104: Based on the traffic flow information, a traffic priority feature corresponding to the moving vehicles is constructed. The traffic priority feature includes a lane traffic density feature and a vehicle type composition feature in the current time period.

[0043] Specifically, geomagnetic detectors, millimeter-wave radars, and video recognition equipment deployed at target intersections capture the number of vehicles passing through each lane per unit time, along with their average speed and distance to each other. These are then combined with road traffic status to form lane traffic density features, which are used to characterize the current traffic saturation and road load. Furthermore, license plate recognition systems and contour recognition algorithms extract information about each type of vehicle, and calculate the proportion of passenger cars, tourist buses, public buses, and freight trucks within the current cycle to form a vehicle type composition feature that reflects the importance, rigidity, and control flexibility of traffic within that cycle. These features collectively constitute the traffic priority characteristics of moving vehicles, providing a basis for subsequent right-of-way allocation and signal phase adjustment.

[0044] In one possible implementation, step S104 further includes: obtaining vehicle quantity data, driving speed data, and spacing data for the lane corresponding to the waiting area within the current time period; constructing lane traffic density characteristics based on the vehicle quantity data, driving speed data, and spacing data; obtaining vehicle type data for the lane corresponding to the waiting area within the current time period; and constructing vehicle type composition characteristics based on the vehicle type data.

[0045] Specifically, geomagnetic detectors and millimeter-wave radar equipment deployed in the corresponding lanes of the waiting area can collect real-time information on vehicle counts passing through detection points, the instantaneous speed of each vehicle, and the distance between vehicles. Vehicle count data reflects the total number of vehicles passing through the lane per unit time, speed data indicates whether traffic flow is smooth or congested, and distance data is used to assess the compactness and continuity of traffic flow. After normalizing the data in these three dimensions, a lane traffic density feature is constructed based on a weighted combination algorithm to quantify the traffic load level of the lane during the current time period.

[0046] Using a high-definition video recognition system and license plate recognition terminals, the system captures the contour features, dimensions, and license plate type of passing vehicles, extracting the vehicle type identifiers for all vehicles passing during the current time period. Combining these with vehicle classification rules, the system then calculates the composition of small passenger cars, tourist buses, public buses, taxis, freight vehicles, construction vehicles, and emergency vehicles. Furthermore, based on the traffic weights of each vehicle type in different scenarios (e.g., buses take priority over private cars, tourist buses take precedence during scenic hours), a vehicle type composition feature is constructed and structured into a multidimensional feature vector for input into the signal control model.

[0047] Finally, the lane traffic density characteristics and vehicle type composition characteristics are used as the traffic priority characteristics of the current traffic flow at the intersection into the input model, which is used to compare and integrate with the tourist traffic urgency characteristics to support the subsequent traffic pressure value evaluation and dynamic adjustment of traffic signal timing.

[0048] Step S105: input the traffic urgency feature, traffic priority feature and real-time waiting time data into the traffic light intelligent space intelligent adjustment model, and output the traffic pressure value of waiting tourists through the traffic light intelligent space intelligent adjustment model.

[0049] Specifically, the tourist urgency features (including their identity and behavioral patterns), priority features (including lane density and vehicle type composition), and real-time wait time data constructed in the previous steps are fed as joint input variables into the trained traffic light intelligent control and adjustment model. This model utilizes a deep neural network structure based on a multi-source attention mechanism, dynamically sensing the intensity of pedestrian and vehicle traffic demand and comprehensively assessing the traffic resistance faced by tourists during the current cycle. Through internal feature fusion and weight calculation, the model outputs a numerical traffic pressure value, representing the comprehensive demand intensity of the tourist group for right of way within the current time window. This serves as a key basis for subsequent traffic light phase adjustment and release strategy generation.

[0050] In a possible implementation, step S105 also includes: uniformly encoding the urgency of passage features, the priority of passage features, and the real-time waiting time data; extracting the deep correlation features between the uniformly encoded urgency of passage features, the priority of passage features, and the real-time waiting time data through a multi-layer neural network structure; performing multi-feature dimensional fusion of the deep correlation features by introducing a weighted attention mechanism; and constructing a traffic light intelligent space intelligent adjustment model based on the deep correlation features after the multi-feature dimensional fusion.

[0051] Specifically, the traffic urgency feature, traffic priority feature and real-time waiting time data are uniformly coded: let the traffic urgency feature be a vector group in, A standardized characteristic dimension of tourists, such as whether they are foreign-speaking tourists, whether they travel in groups, whether they use tour guides, etc. ,and is a positive integer, is the number of vectors in the vector group; Assume that the traffic priority feature is a vector group: in, Indicates the density and type composition of vehicles on one side, such as lane saturation, bus ratio, average spacing, etc. ,and is a positive integer, is the number of vectors in the vector group; Let the waiting time be a separate variable: The real-time waiting time of a tourist in the waiting area. Represents the real number field, and constructs the full feature input vector as: in, Through the multi-layer neural network structure, the deep correlation features between the unified coded traffic urgency features, traffic priority features and real-time waiting time data are extracted: the feature input vector Input multi-layer perception network (MLP), each layer of neural units uses ReLU activation function: in and are the weight matrix and bias vector of the first and second layers respectively, The output vector of the first hidden layer in the traffic light intelligent control adjustment model is After that, the intermediate feature expression result is obtained by linear transformation and activation function processing of the first layer of neural network. Represents the result of nonlinear feature expression. By introducing the attention mechanism, the deep correlation features are integrated into multiple feature dimensions: In order to enhance the dominant role of tourist behavior in the model, the adaptive attention mechanism is introduced to allocate the weight of tourist-side features: in, Indicates the The relative importance weight of each traffic urgency feature in the overall traffic urgency feature, is the feature transformation matrix, is the context query vector, represents the transpose of the following query vector, represents the natural exponential function, Indicates the The original standardized input features of tourists come from the travel urgency feature vector group, It indicates the expression of the urgency of passage under the guidance of behavioral attention. 、 and After combination, it is input into the backbone network again to perform multi-feature dimensional fusion and obtain the fused deep correlation features. Based on the deep correlation features after multi-feature dimension fusion, a traffic light intelligent space intelligent adjustment model is constructed: the final output layer of the traffic light intelligent space intelligent adjustment model adopts Sigmoid activation function to output the tourist traffic pressure value , the value range is , the calculation formula is: in, is the output layer weight, is the bias term, Represents the Sigmoid function.

[0052] Step S106: Adjust the traffic lights in the waiting area based on the traffic pressure value.

[0053] Specifically, the flow pressure value It represents the comprehensive demand intensity of tourist groups for right of way in the current cycle. The signal control system dynamically evaluates the priority of tourist passage based on this value and performs corresponding signal adjustment operations accordingly, thereby realizing a tourist-oriented signal control strategy.

[0054] In a possible implementation, step S106 also includes: obtaining a target traffic pressure level interval corresponding to the traffic pressure value; based on the traffic pressure level, obtaining an intelligent adjustment scheme corresponding to the pressure level interval in a preset traffic light adjustment database, the preset traffic light adjustment database being used to store the correspondence between the pressure level interval and the intelligent adjustment scheme; based on the intelligent adjustment scheme, adjusting the traffic lights in the waiting area.

[0055] Specifically, obtain the currently calculated traffic pressure value , and determine the target traffic pressure level range it is in, for example It is divided into three levels: "low traffic demand area", "medium traffic demand area" and "high traffic demand area", and each area corresponds to a clear signal control strategy.

[0056] The system then searches a preset traffic light adjustment database for the intelligent adjustment solution corresponding to that level range. The database stores traffic light control strategies corresponding to different traffic pressure levels in a key-value structure. These strategies include traffic light phase adjustment rules, phase switching conditions, pedestrian green light extension strategies, and lane green light compression mechanisms, allowing for adaptation to diverse traffic environments and visitor behavior patterns.

[0057] For example, when When in a high-traffic-demand area, the intelligent adjustment plans preset in the database include but are not limited to: releasing the pedestrian green light in advance, extending the duration of the pedestrian green light, inserting an intermediate phase release, or activating local pedestrian priority channels. When in a low-traffic-demand area, the existing timing plan is maintained or the duration of the pedestrian green light is slightly compressed to ensure traffic efficiency.

[0058] Ultimately, the system automatically adjusts the phase configuration and duration of traffic lights within the waiting area based on the retrieved intelligent adjustment solution, achieving real-time responsive control driven by tourist traffic pressure. This implementation approach offers the advantages of high modularity, transparent rules, and ease of engineering deployment. It can map complex traffic behaviors into structured control actions without increasing computational burden, adapting to the practical operational demands of multi-scene, fast-changing, and high-traffic traffic flows in scenic road systems.

[0059] Please refer to Figure 2 , which shows a module schematic diagram of a traffic light intelligent adaptive adjustment device provided by an embodiment of the present application, the device includes an acquisition module 21 and a processing module 22, wherein, The acquisition module 21 is used to acquire individual attribute data and real-time waiting time data corresponding to tourists waiting in the waiting area; and to acquire traffic flow data corresponding to vehicles traveling in the waiting area.

[0060] The processing module 22 is used to construct the traffic urgency characteristics corresponding to waiting tourists based on individual attribute data and real-time waiting time data, and the traffic urgency characteristics include tourist identity characteristics and tourist behavior pattern characteristics; based on traffic flow information, construct the traffic priority characteristics corresponding to moving vehicles, and the traffic priority characteristics include lane traffic density characteristics and vehicle type composition characteristics in the current time period; input the traffic urgency characteristics, traffic priority characteristics and real-time waiting time data into the traffic light intelligent space intelligent adjustment model, and output the traffic pressure value of waiting tourists through the traffic light intelligent space intelligent adjustment model; adjust the traffic lights in the waiting area based on the traffic pressure value.

[0061] In one possible implementation, the acquisition module 21 is used to obtain individual attribute data corresponding to waiting tourists in the waiting area, specifically including: collecting initial individual attribute data corresponding to waiting tourists through a multimodal sensing device, the multimodal sensing device including a visible light camera, an infrared thermal imaging device, and a voice recognition microphone array; inputting the initial individual attribute data into a tourist attribute classification model, and outputting identity feature label data corresponding to different waiting tourists based on the tourist attribute classification model; constructing an identity feature data set corresponding to the classified identity feature label data, and normalizing the identity feature data set; and using the normalized identity feature data set as individual attribute data.

[0062] In one possible implementation, the processing module 22 is used to construct a passage urgency feature corresponding to waiting tourists based on individual attribute data: construct a passage urgency feature judgment index based on identity feature label data, and the passage urgency feature judgment index includes a travel status index, a language index, and a behavior status index; based on the passage urgency feature judgment index and combined with real-time waiting time data, construct a passage urgency feature corresponding to each waiting tourist.

[0063] In one possible implementation, the processing module 22 is used to construct a traffic priority feature corresponding to the moving vehicles based on the traffic flow information, specifically including: obtaining vehicle quantity data, driving speed data, and spacing data of the lane corresponding to the waiting area in the current time period; constructing a lane traffic density feature based on the vehicle quantity data, driving speed data, and spacing data; obtaining vehicle type data of the lane corresponding to the waiting area in the current time period; and constructing a vehicle type composition feature based on the vehicle type data.

[0064] In one possible implementation, the processing module 22 is used to construct a traffic light intelligent space intelligent adjustment model before inputting the traffic urgency characteristics, traffic priority characteristics and real-time waiting time data into the traffic light intelligent space intelligent adjustment model and outputting the traffic pressure value of waiting tourists through the traffic light intelligent space intelligent adjustment model. Specifically, the model includes: uniformly encoding the traffic urgency characteristics, traffic priority characteristics and real-time waiting time data; extracting the deep correlation features between the traffic urgency characteristics, traffic priority characteristics and real-time waiting time data after uniform encoding through a multi-layer neural network structure; performing multi-feature dimensional fusion of the deep correlation features by introducing a weighted attention mechanism; and constructing a traffic light intelligent space intelligent adjustment model based on the deep correlation features after multi-feature dimensional fusion.

[0065] In a possible implementation, the processing module 22 is configured to output the traffic pressure value of waiting passengers through the traffic light intelligent air intelligent adjustment model, specifically including: calculating the traffic pressure value through the following formula: in, is the flow pressure value, represents the Sigmoid function, Intelligently adjust the weight of the output layer of the traffic light model. is the bias term, represents the natural exponential function.

[0066] In one possible embodiment, the processing module 22 is used to adjust the traffic lights in the waiting area based on the traffic pressure value, specifically including: obtaining the target traffic pressure level interval corresponding to the traffic pressure value; based on the traffic pressure level, obtaining the intelligent adjustment scheme corresponding to the pressure level interval in the preset traffic light adjustment database, the preset traffic light adjustment database is used to store the correspondence between the pressure level interval and the intelligent adjustment scheme; based on the intelligent adjustment scheme, adjusting the traffic lights in the waiting area.

[0067] It should be noted that the above embodiments provide devices that implement their functions using only the division of the above functional modules as examples. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiments are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0068] This application also provides an electronic device. Figure 3 , Figure 3 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. The electronic device may include: at least one processor 301, at least one communication bus 302, a user interface 303, at least one network interface 304, and a memory 305.

[0069] The communication bus 302 is used to implement the connection and communication between these components.

[0070] The user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0071] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0072] The processor 301 may include one or more processing cores. Using various interfaces and circuits, the processor 301 connects to various components within the server. It executes instructions, programs, code sets, or instruction sets stored in the memory 305, as well as accesses data stored in the memory 305, to perform various server functions and process data. Optionally, the processor 301 may be implemented using at least one of the following hardware forms: a digital signal processing (DSP), a field-programmable gate array (FPGA), or a programmable logic array (PLA). The processor 301 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing content displayed on the display screen; and the modem handles wireless communications. It is understood that the modem may not be integrated into the processor 301 but implemented as a separate chip.

[0073] Among them, the memory 305 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may also optionally be at least one storage device located away from the aforementioned processor 301. Refer to Figure 3 , as a computer storage medium, the memory 305 may include an operating system, a network communication module, a user interface module, and a traffic light intelligent adaptive adjustment application.

[0074] exist Figure 3In the electronic device shown, the user interface 303 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 301 can be used to call the traffic light intelligent adaptive adjustment application stored in the memory 305. When executed by one or more processors 301, the electronic device executes one or more of the methods described in the above embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should know that this application is not limited to the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for this application.

[0075] The present application also provides a computer-readable storage medium storing instructions, which, when executed by one or more processors, enable an electronic device to execute one or more of the methods described in the above embodiments.

[0076] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0077] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely schematic, such as the division of units, which is only a logical function division. In actual implementation, there may be other division methods, such as 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 service interface, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0078] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they 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 according to actual needs.

[0079] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0080] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of this application, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of this application. The aforementioned memory includes various media that can store program code, such as USB flash drives, mobile hard drives, magnetic disks, or optical disks.

[0081] The above descriptions are merely exemplary embodiments disclosed in this application and are not intended to limit the scope of this application. That is, any equivalent changes and modifications made based on the teachings disclosed in this application are still within the scope of this application.

[0082] This application is intended to cover any modifications, uses or adaptations disclosed in this application, which follow the general principles disclosed in this application and include common knowledge or customary technical means in the technical field not disclosed in this application.

Claims

1. A method for intelligent adaptive adjustment of traffic lights, characterized in that: The method comprises: Obtain individual attribute data and real-time waiting time data corresponding to tourists waiting in the waiting area; Obtaining traffic flow data corresponding to vehicles traveling in the waiting area; Based on the individual attribute data and the real-time waiting time data, constructing a passage urgency feature corresponding to the waiting tourist, the passage urgency feature including tourist identity features and tourist behavior pattern features; Based on the traffic flow information, constructing a traffic priority feature corresponding to the moving vehicle, the traffic priority feature including a lane traffic density feature and a vehicle type composition feature within a current time period; Input the passage urgency feature, the passage priority feature and the real-time waiting time data into the traffic light intelligent space intelligent adjustment model, and output the passage pressure value of the waiting tourists through the traffic light intelligent space intelligent adjustment model; A traffic light in the waiting area is adjusted based on the traffic pressure value.

2. The method according to claim 1, characterized in that The obtaining of individual attribute data corresponding to the tourists waiting in the waiting area specifically includes: Collecting initial individual attribute data corresponding to the waiting tourists through a multimodal sensing device, wherein the multimodal sensing device includes a visible light camera, an infrared thermal imaging device, and a voice recognition microphone array; Inputting the initial individual attribute data into a tourist attribute classification model, and outputting identity feature label data corresponding to different waiting tourists based on the tourist attribute classification model; Constructing an identity feature data set corresponding to the classified identity feature label data, and performing normalization processing on the identity feature data set; The normalized identity feature data set is used as the individual attribute data.

3. The method according to claim 2, characterized in that Based on the individual attribute data, the passage urgency feature corresponding to the waiting tourists is constructed: Based on the identity feature tag data, constructing a travel urgency feature judgment index, the travel urgency feature judgment index including a travel status index, a language index, and a behavior status index; Based on the passage urgency feature judgment index and combined with the real-time waiting time data, the passage urgency feature corresponding to each waiting tourist is constructed.

4. The method according to claim 1, wherein The constructing of a traffic priority feature corresponding to the moving vehicle based on the traffic flow information specifically includes: Obtaining vehicle quantity data, driving speed data, and spacing data for the lane corresponding to the waiting area within a current time period; Constructing the lane traffic density feature according to the vehicle quantity data, the driving speed data, and the spacing data; Obtaining vehicle type data for the lane corresponding to the waiting area within a current time period; The vehicle type constituent features are constructed based on the vehicle type data.

5. The method according to claim 1, characterized in that Before inputting the passage urgency feature, the passage priority feature, and the real-time waiting time data into the traffic light intelligent space intelligent adjustment model and outputting the passage pressure value of the waiting tourists through the traffic light intelligent space intelligent adjustment model, it is necessary to construct the traffic light intelligent space intelligent adjustment model, specifically including: uniformly encoding the passage urgency feature, the passage priority feature, and the real-time waiting time data; Extracting deep correlation features between the passage urgency feature, the passage priority feature, and the real-time waiting time data after the unified coding through a multi-layer neural network structure; By introducing a weighted attention mechanism, the deep correlation features are fused into multiple feature dimensions; Based on the deep correlation features after the multi-feature dimension fusion, the traffic light intelligent space intelligent adjustment model is constructed.

6. The method according to claim 5, characterized in that Outputting the passage pressure value of the waiting tourists through the traffic light intelligent air intelligent adjustment model specifically includes: The flow pressure value is calculated by the following formula: in, is the passage pressure value, represents the Sigmoid function, To perform the deep correlation feature after the fusion of multiple feature dimensions, The weight of the output layer of the intelligent air-conditioning model of the traffic light is is the bias term, represents the natural exponential function.

7. The method according to claim 1, characterized in that The adjusting the traffic lights in the waiting area based on the traffic pressure value specifically includes: Obtaining a target traffic pressure level interval corresponding to the traffic pressure value; Based on the traffic pressure level, obtaining an intelligent adjustment solution corresponding to the pressure level interval from a preset traffic light adjustment database, wherein the preset traffic light adjustment database is used to store a correspondence between the pressure level interval and the intelligent adjustment solution; Based on the intelligent adjustment scheme, the traffic lights in the waiting area are adjusted.

8. An intelligent adaptive adjustment device for traffic lights, characterized in that: The device includes an acquisition module and a processing module, wherein: The acquisition module is used to acquire individual attribute data and real-time waiting time data corresponding to tourists waiting in the waiting area; and to acquire traffic flow data corresponding to vehicles traveling in the waiting area; The processing module is used to construct the traffic urgency characteristics corresponding to the waiting tourists based on the individual attribute data and the real-time waiting time data, and the traffic urgency characteristics include the tourist identity characteristics and the tourist behavior pattern characteristics; based on the traffic flow information, construct the traffic priority characteristics corresponding to the moving vehicles, and the traffic priority characteristics include the lane traffic density characteristics and vehicle type composition characteristics in the current time period; input the traffic urgency characteristics, the traffic priority characteristics and the real-time waiting time data into the traffic light intelligent space intelligent adjustment model, and output the traffic pressure value of the waiting tourists through the traffic light intelligent space intelligent adjustment model; adjust the traffic lights in the waiting area based on the traffic pressure value.

9. An electronic device, characterized in that: The electronic device comprises a processor, a communication bus, a user interface, a network interface and a memory, wherein the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is performed.

Images