A Highway Traffic Congestion Identification Method Based on One-Dimensional Fuzzy Clouds

Abstract

This invention discloses a method for identifying highway traffic congestion based on one-dimensional fuzzy clouds, belonging to the field of traffic engineering. The method includes: constructing a congestion index database for a specific section of the highway; selecting the top three key indicators as congestion discrimination indicators using the entropy weight method; fuzzifying the congestion discrimination indicators and classifying them into multiple levels; calculating a quantitative value of congestion severity using a weighted average of traffic density and human perception scores, and classifying multiple congestion levels; constructing congestion discrimination rules for single-indicator inputs based on the classification thresholds of the congestion discrimination indicators and congestion levels, combined with expert experience, and constructing a one-dimensional fuzzy cloud model for each single indicator using an inverse cloud algorithm; based on the one-dimensional fuzzy cloud model, constructing a one-dimensional fuzzy cloud multi-rule generator for each discrimination indicator, outputting the congestion severity value of each indicator, calculating the comprehensive congestion severity value and matching it with the congestion level to obtain the final judgment result. This invention improves the comprehensiveness and scientific rigor of congestion discrimination.

Description

Technical Field

[0001] This invention belongs to the field of traffic engineering technology, and in particular relates to a method for identifying highway traffic congestion based on one-dimensional fuzzy clouds. Background Technology

[0002] As the backbone of the modern transportation system, highways directly impact regional economic development and public travel experience through their traffic efficiency. Currently, highway traffic congestion identification primarily relies on threshold-based methods and single-indicator models. Threshold-based methods classify congestion levels (e.g., smooth flow, congested) by setting fixed threshold values ​​for key traffic parameters (such as speed and density), offering the advantages of simplicity and intuitiveness. Some methods attempt to incorporate subjective human evaluation to assist in scoring congestion levels, reflecting drivers' actual perceptions. Furthermore, some models base congestion judgments on historical statistical patterns (such as travel time ratio) or single core indicators (such as traffic density). These methods constitute the mainstream technologies for current highway congestion identification.

[0003] However, existing technologies face significant shortcomings: First, traditional threshold methods are poorly adaptable to complex and ever-changing traffic scenarios. Fixed thresholds are insufficient to effectively address differences in traffic flow characteristics across different road sections, time periods (such as peak and off-peak hours, holidays), and weather conditions, leading to decreased recognition accuracy. Second, existing methods cannot scientifically quantify the fuzzy boundaries in the gradual transition of congestion, particularly the transition from "moderate congestion" to "severe congestion," which is inherently random and fuzzy. Simple threshold segmentation cannot accurately characterize this continuous change. Furthermore, methods relying on a single indicator or simply combining subjective scoring have limitations: a single indicator cannot comprehensively reflect the multidimensional characteristics of congestion (spatial occupancy, time delay, etc.), while manual evaluation introduces subjective perception but lacks a systematic integration mechanism with objective data and is difficult to apply on a large scale in real time. These problems collectively result in insufficient accuracy and practicality in highway traffic congestion identification, necessitating a more scientific and adaptable identification method. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a highway traffic congestion identification method based on one-dimensional fuzzy clouds, thereby resolving the issues present in the prior art.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a method for identifying highway traffic congestion based on one-dimensional fuzzy clouds, comprising:

[0006] By integrating historical traffic data with real-time traffic data, a congestion index database for a specific section of a highway can be constructed.

[0007] Based on the aforementioned congestion indicator library, the top three key indicators are selected as congestion discrimination indicators using the entropy weight method, and the weight of each indicator is determined.

[0008] The congestion identification indicators are fuzzified and divided into multiple levels;

[0009] The degree of congestion is quantified by weighting traffic density and human perception scores, and then divided into multiple congestion levels.

[0010] Based on the threshold values ​​for the congestion discrimination index and the threshold values ​​for the congestion level, a congestion discrimination rule with a single index input is constructed by combining expert experience, and a one-dimensional fuzzy cloud model with a single index is constructed by using the reverse cloud algorithm.

[0011] Based on the one-dimensional fuzzy cloud model, a one-dimensional fuzzy cloud multi-rule generator for each discrimination index is constructed.

[0012] The congestion level of each indicator is output based on the one-dimensional fuzzy cloud multi-rule generator. The congestion level of each indicator is combined with the weight to calculate the comprehensive congestion level. The comprehensive congestion level is matched with the congestion level to obtain the final judgment result.

[0013] Preferably, the process of constructing a congestion index database for a certain section of a highway includes:

[0014] Extract traffic volume indicators, congestion duration indicators, maximum queue length indicators, travel time ratio indicators, average travel time indicators, and average travel speed indicators from historical data;

[0015] Extract traffic flow indicators, speed indicators, density indicators, number of blocked vehicles indicators, and space occupancy indicators from real-time data;

[0016] The extracted indicators are processed for missing values ​​and outliers are removed. Based on the processed indicators, a congestion indicator database for a certain section of the highway is constructed.

[0017] Preferably, the process of selecting the top three key indicators as congestion identification indicators using the entropy weight method includes:

[0018] The indicators in the congestion indicator library are standardized to obtain standardized formulas.

[0019] Using the standardized formula, the entropy value and entropy weight of each index are calculated using the entropy weight method.

[0020] The top three indicators in terms of entropy weight were selected as congestion indicators.

[0021] Preferably, the process of fuzzifying the congestion identification index and dividing it into multiple levels includes:

[0022] Based on ETC gantry detection data, a cumulative frequency curve of congestion discrimination indicators is plotted.

[0023] Based on the frequency cumulative curve, each indicator is divided into five levels according to a preset threshold.

[0024] Preferably, the process of calculating a quantitative value of congestion level by weighting traffic density and human perception scores, and classifying it into multiple congestion levels, includes:

[0025] Traffic density is mapped to the first score, and human perception score is mapped to the second score;

[0026] The first and second scores are weighted and summed to obtain a quantitative value of the congestion level.

[0027] The congestion level is divided into five congestion levels based on preset threshold boundaries.

[0028] Preferably, the process of constructing a congestion discrimination rule based on a single indicator input and constructing a one-dimensional fuzzy cloud model of the single indicator using the inverse cloud algorithm includes:

[0029] Congestion discrimination rules are constructed based on the threshold values ​​for congestion discrimination indicators and congestion level classification.

[0030] The expected value, entropy, and hyperentropy are extracted from historical data using the inverse cloud algorithm and used as feature parameters for a one-dimensional fuzzy cloud model.

[0031] Based on the congestion discrimination rule and the feature parameters, a one-dimensional fuzzy cloud model with a single indicator is constructed.

[0032] Preferably, the process of constructing a one-dimensional fuzzy cloud multi-rule generator for each discrimination index includes:

[0033] For each congestion assessment indicator, construct multiple single-rule generators;

[0034] Calculate the activation intensity for each single rule based on the new input metric value;

[0035] The activation strength is iterated multiple times to generate random values;

[0036] The mean of all random values ​​is used as the congestion level value for each congestion discrimination indicator.

[0037] Preferably, the process of obtaining the final judgment result includes:

[0038] The overall congestion level value is matched with the threshold range of congestion level, and the level that falls into the corresponding range is the final judgment result.

[0039] In a second aspect, the present invention also discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0040] Thirdly, the present invention also discloses a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0041] Compared with the prior art, the present invention has the following advantages and technical effects:

[0042] This invention provides a method for identifying highway traffic congestion based on one-dimensional fuzzy clouds, comprising: First, fusing historical traffic data and real-time traffic data to construct a congestion index database for a certain section of the highway; Second, based on the congestion index database, selecting the top three key indicators as congestion discrimination indicators using the entropy weight method, and determining the weight of each indicator; Third, fuzzifying the congestion discrimination indicators and dividing them into multiple levels; Next, calculating the quantitative value of congestion degree using traffic density and human perception scoring weighted average, and dividing it into multiple congestion levels; Further, based on the classification thresholds of the congestion discrimination indicators and the congestion level classification thresholds, constructing congestion discrimination rules for single indicator inputs in conjunction with expert experience, and constructing a one-dimensional fuzzy cloud model for a single indicator using the inverse cloud algorithm; Based on the one-dimensional fuzzy cloud model, constructing a one-dimensional fuzzy cloud multi-rule generator for each discrimination indicator; Finally, outputting the congestion degree value of each indicator according to the one-dimensional fuzzy cloud multi-rule generator; Combining the congestion degree values ​​of each indicator with their weights to calculate a comprehensive congestion degree value; Matching the comprehensive congestion degree value with the congestion level to obtain the final judgment result.

[0043] This invention integrates historical and real-time data, and uses the entropy weight method to determine the importance of indicators. It constructs a traffic congestion discrimination index system for basic road sections of the expressway network, which improves the adaptability and recognition accuracy of the fuzzy cloud model for expressway traffic congestion identification to actual traffic conditions.

[0044] This invention uses a "traffic density + human perception" method to quantify the degree of congestion, combining objective data statistics with subjective perception scoring to improve the realism and quantification accuracy of congestion identification;

[0045] This invention transforms discrete indicators into continuous cloud droplet distributions through a reverse cloud algorithm. It extracts the distribution characteristics of indicator data from the cloud clusters formed by massive cloud droplets in the data sample and transforms them into qualitative semantic descriptions. This approach takes into account both data distribution patterns and expert experience, thereby improving the applicability and accuracy of congestion prediction methods.

[0046] This invention employs multiple congestion indicators to construct congestion discrimination models based on one-dimensional fuzzy clouds, and improves the comprehensiveness and scientific rigor of congestion discrimination by merging the output results of each model through weighted fusion. Attached Figure Description

[0047] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0048] Figure 1 This is a flowchart of a method according to an embodiment of the present invention;

[0049] Figure 2 This is a schematic diagram of a one-dimensional forward cloud generator according to an embodiment of the present invention;

[0050] Figure 3 This is a schematic diagram of a one-dimensional reverse cloud generator according to an embodiment of the present invention;

[0051] Figure 4 This is a schematic diagram of the distribution of membership degree clouds over the universe of discourse in an embodiment of the present invention;

[0052] Figure 5 This is a schematic diagram of a one-dimensional single rule generator according to an embodiment of the present invention;

[0053] Figure 6 This is a schematic diagram of a one-dimensional multi-rule generator according to an embodiment of the present invention. Detailed Implementation

[0054] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0055] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0056] Example 1

[0057] like Figure 1 As shown, this embodiment provides a highway traffic congestion identification method based on one-dimensional fuzzy clouds, including:

[0058] Step 1: Integrate historical and real-time data to construct a congestion index database for a specific section of the highway.

[0059] The specific process of step 1 is as follows:

[0060] This embodiment proposes an indicator selection framework based on the collaboration of historical and real-time data, aiming to construct a multi-dimensional congestion discrimination index applicable to highway scenarios. Based on historical traffic congestion big data of the highway network, it integrates a historical congestion case library and real-time traffic big data to extract congestion indicators such as traffic volume, congestion duration, maximum queue length, travel time ratio, average travel time, average travel speed, flow rate, speed, density, number of blocked vehicles, and space occupancy rate during holidays and peak hours.

[0061] This embodiment lists the indicators representing traffic congestion status, as shown in Table 1. The traffic volume, congestion duration, maximum queue length, travel time ratio, average travel time, and average travel speed of congestion cases are extracted from the historical database. Real-time data is used to extract indicators such as flow rate, speed, density, number of blocked vehicles, and space occupancy rate, and missing values ​​are processed and outliers are removed.

[0062] Table 1

[0063]

[0064]

[0065] Step 2: Based on the congestion index database of a certain section of the highway obtained in Step 1, the top three key indicators are selected from the index database using the entropy weight method as congestion discrimination indicators, and the corresponding weights of each indicator are determined.

[0066] The specific process of step 2 is as follows:

[0067] This embodiment uses the entropy weight method to calculate the information content of the indicators, screens and sorts the congestion indicators of the highway network, and obtains the objective weight of each congestion indicator by calculating information entropy and assigning weights, thus screening out core indicators with low redundancy and high sensitivity.

[0068] Suppose there are m samples of a certain road segment, and each sample of a certain road segment has n indicators. Then the value of the j-th indicator of the i-th road segment is x. ij After standardizing the indicators, we get a ij .

[0069] For positive indicators (the larger the value, the more congested the traffic), the standardized formula is:

[0070]

[0071] For negative indicators (the smaller the value, the more congested the traffic), the standardized formula is:

[0072]

[0073] Using the entropy weight method, the formula for calculating the weight of the index value is as follows:

[0074]

[0075] In the formula: P ij Let be the weight of the index value of the i-th road segment sample under the j-th index; m is the number of samples of a certain road segment to be evaluated; n is the number of evaluation indicators; a ij Let be the evaluation value of the i-th item under the j-th indicator.

[0076] The formula for calculating the entropy value of the j-th index is:

[0077]

[0078] In the formula: e j The range is 0-1, where j is the entropy value of the j-th index; -1 / lnm is the information entropy coefficient.

[0079] The formula for calculating the entropy weight of the indicator is:

[0080]

[0081] In the formula: k j Let be the entropy weight of the j-th index.

[0082] The top three congestion indicators ranked by entropy weight are selected as the congestion discrimination indicators (X1, X2, X3), and their corresponding entropy weights are k1, k2, and k3, respectively.

[0083] Step 3: The congestion identification indicators selected in Step 2 are fuzzified and divided into multiple levels.

[0084] The specific process of step 3 is as follows:

[0085] This study investigates the fuzzy logic structure for highway traffic congestion identification, fuzzifying the selected input congestion discrimination indicators. Based on ETC gantry detection data, indicators are extracted, statistically analyzed, and their cumulative frequency curves are plotted. Using 15%, 30%, 50%, and 85% as threshold boundaries, the congestion discrimination indicators are divided into five levels: "very low," "low," "medium," "high," and "very high."

[0086] Step 4: Use the "traffic density + human perception" method to quantify the degree of congestion and classify the congestion levels.

[0087] The specific process of step 4 is as follows:

[0088] Since the assessment of congestion level is highly correlated with the psychological feelings of traffic participants, the method of "traffic density + human perception" is used to quantify the degree of congestion.

[0089] Based on the historical congestion density of a road segment, supplemented by manual scoring of congestion video detection data at that moment, a weighted quantitative value for congestion severity is determined. For density, a linear mapping from 0 to 1 is used based on the statistical characteristics of historical data; for manual perception scoring, congestion severity is scored within a range of 0 to 1, with 1 point considered completely congested. The density weight is set to 0.5, and the manual perception score weight is also set to 0.5. The resulting quantitative value for congestion severity is then statistically analyzed, and its cumulative frequency curve is plotted. Thresholds of 15%, 30%, 50%, and 85% are used to divide the congestion into five levels: {"Smooth Traffic," "Mostly Smooth Traffic," "Light Congestion," "Moderate Congestion," and "Severe Congestion"}.

[0090] Step 5: Based on the threshold values ​​for congestion discrimination indicators and congestion levels determined in Steps 3 and 4, and combined with expert experience, construct congestion discrimination rules for single congestion indicator input, and construct a one-dimensional fuzzy cloud model for a single indicator.

[0091] The specific process of step 5 is as follows:

[0092] One-dimensional fuzzy cloud models are based on cloud parameters (expectations). entropy hyperentropy It describes the randomness and fuzziness of a single indicator, and its advantage lies in taking into account both the data distribution pattern and expert experience.

[0093] The essence of constructing a one-dimensional fuzzy cloud model is to extract the distribution characteristics of indicator data from the cloud clusters formed by massive cloud droplets of data samples and transform them into qualitative semantic descriptions.

[0094] We adopted the congestion index classification thresholds, quantification results of congestion levels, and congestion level classification thresholds from previous studies. Based on the single-index thresholds and combined with expert experience, we constructed a congestion discrimination rule base for single congestion index inputs, such as: IF "X1 is very low" THEN "Congestion level is smooth", etc.

[0095] The model input is a single congestion index, X1, and the output is the membership degree of this index to each level of congestion severity. One-dimensional cloud parameters (expected values) are extracted from historical congestion index data using a reverse cloud algorithm. entropy hyperentropy The one-dimensional cloud congestion discrimination model for the discrimination index X1 has been completed. Similarly, for each level of congestion indices X2 and X3, one-dimensional cloud parameters are extracted from historical data based on the principle of reverse cloud. For each congestion level (t) 畅通 t 基本畅通 t 轻度拥堵 t 中度拥堵 t 严重拥堵Based on the principle of reverse cloud, one-dimensional cloud parameters are extracted from historical data.

[0096] Forward and reverse belong to the cloud generator, such as Figure 2 , Figure 3 As shown.

[0097] The formula for a one-dimensional fuzzy cloud is as follows:

[0098] Sample mean for:

[0099]

[0100] In the formula x i Let i be the i-th sample.

[0101] The absolute central distance B of the first-order sample is:

[0102]

[0103] In the formula, N is the number of samples.

[0104] Sample variance S 2 for:

[0105]

[0106] One-dimensional cloud parameter expectation value for:

[0107]

[0108] One-dimensional cloud parameter entropy for:

[0109]

[0110] One-dimensional cloud parameter hyperentropy for:

[0111]

[0112] Once the rules for judgment and the cloud digital features of each rule are obtained, the one-dimensional cloud congestion discrimination model is completed.

[0113] Using a positive cloud generator, membership clouds for each level of each indicator are generated, resulting in a schematic diagram of the membership degree cloud, as shown below. Figure 4 As shown.

[0114] Step 6: Construct a one-dimensional fuzzy cloud multi-rule generator based on the one-dimensional fuzzy cloud model obtained in Step 5.

[0115] The specific process of step 6 is as follows:

[0116] A one-dimensional cloud single-rule generator can be constructed using a one-dimensional X-conditional cloud generator and a one-dimensional Y-conditional cloud generator. For example, a cloud generator diagram with the rule IF "X1 is very low" THEN "Congestion level is smooth" is shown below. Figure 5 As shown.

[0117] Figure 5 In the middle, drop(x1,y) is a two-dimensional point that satisfies the one-dimensional normal cloud distribution law, i.e., a cloud droplet.

[0118] Combining multiple such one-dimensional fuzzy cloud single-rule generators yields a one-dimensional fuzzy cloud multi-rule generator, taking index X1 as an example. Figure 6 As shown.

[0119] Figure 6 The final output result T1 is the output result of the one-dimensional fuzzy cloud multi-rule generator congestion discrimination model, which is the congestion degree value 1, i.e., the congestion discrimination index X1.

[0120] Congestion identification is performed on newly input data based on a one-dimensional fuzzy cloud model. For each single rule, For the expectation, To generate a one-dimensional random value (En) that conforms to a one-dimensional normal distribution, representing the variance. i Based on the newly input congestion index X1, (En) is used. i Find the activation strength, i.e., the membership degree y, obtained when inputting X1 in all single-rule generator antecedents. i The calculation formula is:

[0121]

[0122] Take μ i The single rule corresponding to the largest μ1 in the middle, and the one-dimensional cloud parameters of the consequent given by this rule. Randomly generated For the expectation, A one-dimensional normal random value with variance (En) 2i The formula is used to calculate the value at y1, En. 2i The formula for calculating the T-value (i.e., the degree of congestion) under the given conditions is shown below.

[0123]

[0124] After multiple iterations (more than 1000 times), the average of all expected values ​​is finally output as the prediction result.

[0125] Similarly, a one-dimensional cloud congestion discrimination model is established for congestion discrimination indicators X2 and X3, and the congestion prediction value is calculated.

[0126] Step 7: Based on the one-dimensional fuzzy cloud multi-rule generator of each discrimination index obtained in Step 6, determine the comprehensive congestion level value to obtain the final congestion level.

[0127] The specific process of step 7 is as follows:

[0128] Let the output of the one-dimensional fuzzy cloud multi-rule generator of congestion discrimination index X1 be the congestion level value 1, i.e., T1; the output of the one-dimensional fuzzy cloud multi-rule generator of congestion discrimination index X2 be the congestion level value 2, i.e., T1; and the output of the one-dimensional fuzzy cloud multi-rule generator of congestion discrimination index X3 be the congestion level value 3, i.e., T3.

[0129] The overall congestion level is:

[0130] T = k1 × T1 + k2 × T2 + k3 × T3 (14)

[0131] Substituting the specific value of T into the threshold boundaries of the five congestion levels in step 4 yields the final congestion level.

[0132] Example 2

[0133] This embodiment also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0134] Example 3

[0135] This embodiment also discloses a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0136] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for identifying highway traffic congestion based on one-dimensional fuzzy clouds, characterized in that, Includes the following steps: By integrating historical traffic data with real-time traffic data, a congestion index database for a specific section of a highway can be constructed. Based on the aforementioned congestion indicator library, the top three key indicators are selected as congestion discrimination indicators using the entropy weight method, and the weight of each indicator is determined. The congestion identification indicators are fuzzified and divided into multiple levels; The degree of congestion is quantified by weighting traffic density and human perception scores, and then divided into multiple congestion levels. Based on the threshold values ​​for congestion discrimination indicators and congestion levels, a congestion discrimination rule with a single indicator input is constructed using expert experience. A one-dimensional fuzzy cloud model of the single indicator is then built using the inverse cloud algorithm. The process of constructing the congestion discrimination rule with a single indicator input and building the one-dimensional fuzzy cloud model using the inverse cloud algorithm includes: constructing a congestion discrimination rule based on the threshold values ​​for congestion discrimination indicators and congestion levels; extracting expected value, entropy, and hyperentropy from historical data using the inverse cloud algorithm as feature parameters of the one-dimensional fuzzy cloud model; and constructing the one-dimensional fuzzy cloud model of the single indicator based on the congestion discrimination rule and the feature parameters. Based on the aforementioned one-dimensional fuzzy cloud model, a one-dimensional fuzzy cloud multi-rule generator for each discriminant indicator is constructed. The process of constructing the one-dimensional fuzzy cloud multi-rule generator for each discriminant indicator includes: constructing multiple single-rule generators for each congestion discriminant indicator; calculating the activation intensity for each single rule based on the new input indicator value; iterating the activation intensity multiple times to generate random values; and using the average of all random values ​​as the congestion degree value for each congestion discriminant indicator. The congestion level of each indicator is output based on the one-dimensional fuzzy cloud multi-rule generator. The congestion level of each indicator is combined with the weight to calculate the comprehensive congestion level. The comprehensive congestion level is matched with the congestion level to obtain the final judgment result.

2. The method according to claim 1, characterized in that, The process of constructing a congestion index database for a specific section of a highway includes: Extract traffic volume indicators, congestion duration indicators, maximum queue length indicators, travel time ratio indicators, average travel time indicators, and average travel speed indicators from historical data; Extract traffic flow indicators, speed indicators, density indicators, number of blocked vehicles indicators, and space occupancy indicators from real-time data; The extracted indicators are processed for missing values ​​and outliers are removed. Based on the processed indicators, a congestion indicator database for a certain section of the highway is constructed.

3. The method according to claim 1, characterized in that, The process of selecting the top three key indicators as congestion assessment indicators using the entropy weight method includes: The indicators in the congestion indicator library are standardized to obtain standardized formulas. Using the standardized formula, the entropy value and entropy weight of each index are calculated using the entropy weight method. The top three indicators in terms of entropy weight were selected as congestion indicators.

4. The method according to claim 1, characterized in that, The process of fuzzifying the congestion identification indicators and dividing them into multiple levels includes: Based on ETC gantry detection data, a cumulative frequency curve of congestion discrimination indicators is plotted. Based on the frequency cumulative curve, each indicator is divided into five levels according to a preset threshold.

5. The method according to claim 1, characterized in that, The process of calculating a quantitative value of congestion level by weighting traffic density and human perception scores, and classifying it into multiple congestion levels, includes: Traffic density is mapped to the first score, and human perception score is mapped to the second score; The first and second scores are weighted and summed to obtain a quantitative value of the congestion level. The congestion level is divided into five congestion levels based on preset threshold boundaries.

6. The method according to claim 1, characterized in that, The process of obtaining the final judgment result includes: The overall congestion level value is matched with the threshold range of congestion level, and the level that falls into the corresponding range is the final judgment result.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1-6.

8. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1-6.

Images