Road traffic safety hazard improvement evaluation method and system

Abstract

The present invention provides a method and system for evaluating the improvement of road traffic safety hazards, including: S1: obtaining traffic accident records during the implementation of a traffic safety hazard improvement plan; S2: analyzing a first safety hazard set based on traffic accidents occurring within an evaluation period; S3: analyzing a second safety hazard set based on road status information within the evaluation period; S4: summarizing the first safety hazard set and the second safety hazard set and performing correlation mining to obtain a comprehensive safety hazard set existing within the corresponding evaluation period; S5: sorting and recording the safety hazard sets of all evaluation periods that were most recently obtained during the implementation of the traffic safety hazard improvement plan in chronological order to obtain a hazard record thread; S6: evaluating the improvement evaluation value of the traffic safety hazard improvement plan in real time based on the hazard record thread; so as to achieve an accurate evaluation of the real-time improvement of traffic safety hazards during the implementation of the traffic safety hazard improvement plan.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic safety assessment, and in particular to a method and system for evaluating improvement of hidden dangers in road traffic safety. Background Art

[0002] At present, the direction of the measures taken by the Ministry of Traffic Safety Supervision to improve road safety has evolved from "post-event improvement" to "pre-event warning", and improving road traffic safety hazards is an important part of the "pre-event warning" of traffic accidents. The traffic safety supervision technical department analyzes the traffic safety hazards on the road from the real-time situation of the road or the traffic accident situation, and formulates corresponding improvement plans to improve the road traffic safety hazards.

[0003] However, the improvement plan for road traffic safety hazards involves many aspects. The improvement projects may be large or small and some improvement projects may last a long time, causing the road condition to change during the improvement project. In turn, the previously specified road traffic safety hazard improvement plan may not be sufficient to improve the safety hazards caused by the road during the improvement project. Therefore, real-time evaluation of the road traffic safety hazard improvement plan during the improvement project is extremely necessary.

[0004] For example, the patent with publication number CN113095667B in the prior art discloses a method for evaluating the improvement of road traffic safety hazards, which includes: obtaining high-risk points for road traffic accidents based on historical accident data; establishing a knowledge base of safety hazard improvement measures based on high-risk points for road traffic accidents, wherein the safety hazard improvement measures knowledge base includes a variety of improvement plans; selecting at least two improvement plans from the safety hazard improvement measures knowledge base, and sorting the selected improvement plans according to evaluation rules; and determining an optimal improvement plan from the sorted improvement plans according to optimization principles. The present invention also discloses a road traffic safety hazard improvement evaluation system. The road traffic safety hazard improvement evaluation method provided by the present invention can select the best improvement plan for road safety hazards, which is convenient for traffic management departments to formulate scientific and economical improvement methods for road traffic safety hazard points.

[0005] The above patent also selects the best improvement plan for road safety hazards before the improvement project. It lacks real-time evaluation of the improvement plan for road traffic safety hazards during the improvement project, and cannot achieve real-time optimization of the improvement plan, so the improvement effect of road traffic safety hazards needs to be improved.

[0006] Therefore, the present invention proposes a method and system for evaluating improvement of road traffic safety hazards. Summary of the invention

[0007] The present invention provides a road traffic safety hazard improvement evaluation method and system, which are used to achieve accurate evaluation of the real-time improvement of traffic safety hazards during the implementation of traffic safety hazard improvement plans, provide an important reference for the real-time optimization of road traffic safety hazard improvement plans, and meet the needs of real-time optimization of road traffic safety hazard improvement plans.

[0008] The present invention provides a method for evaluating improvement of hidden dangers in road traffic safety, comprising:

[0009] S1: Obtain traffic accident records during the implementation of traffic safety hazard improvement plans;

[0010] S2: Analyze the first safety hazard set based on the traffic accidents that occurred within the evaluation period in the traffic accident records;

[0011] S3: Analyze a second set of safety hazards based on the road status information within the evaluation period;

[0012] S4: Summarize and associate the first set of safety hazards and the second set of safety hazards to obtain a comprehensive set of safety hazards existing in a corresponding assessment period;

[0013] S5: sorting and recording the safety hazard sets of all evaluation periods that are newly obtained in the process of executing the traffic safety hazard improvement plan in chronological order to obtain a hazard recording thread;

[0014] S6: Based on the hidden danger record thread, the improvement evaluation value of the traffic safety hidden danger improvement plan is evaluated in real time.

[0015] Preferably, the road traffic safety hazard improvement evaluation method, S2: analyzing a first safety hazard set based on traffic accidents occurring within an evaluation period in traffic accident records, comprises:

[0016] S201: Obtain a sub-record thread in the traffic accident record within an evaluation period, and determine a target hidden danger type and hidden danger weight corresponding to the target traffic accident based on the accident scale and accident type of the target traffic accident in the sub-record thread;

[0017] S202: Calculating the comprehensive weight of the target hidden danger type in the time dimension based on the interval time of the target traffic accidents of the same target hidden danger type in the sub-recording thread and the hidden danger weight;

[0018] S203: determining a target derived hazard type and a corresponding derived hazard weight of the target hazard type based on the derived hazard table of the target hazard type and the comprehensive weight;

[0019] S204: Sort all target hidden danger types and all derived hidden danger weights based on the comprehensive weights and derived hidden danger weights to obtain a first safety hidden danger set.

[0020] Preferably, the road traffic safety hazard improvement evaluation method, S203: based on the derived hazard table of the target hazard type and the comprehensive weight, determining the target derived hazard type and the corresponding derived hazard weight of the target hazard type, comprises:

[0021] Determine the target derivative depth based on the comprehensive weight of the target hidden danger type, and determine the target derivative hidden danger type whose derivative depth is within the target derivative depth in the derivative hidden danger table of the target hidden danger type;

[0022] Based on the comprehensive weight of the target hidden danger type and the derived weight of the target derived hidden danger type, the derived hidden danger weight of the corresponding target derived hidden danger type is calculated.

[0023] Preferably, the road traffic safety hazard improvement evaluation method, S3: analyzing a second safety hazard set based on road state information within an evaluation period, comprises:

[0024] Obtaining road status information within the latest evaluation period, wherein the road status information includes: road status monitoring images and road status inspection information;

[0025] Based on the hidden danger deep learning model, all road status monitoring images in the latest assessment period are identified to obtain the first hidden danger type and the corresponding first hidden danger weight;

[0026] Determine a second hidden danger type and a corresponding second hidden danger weight based on the hidden danger keyword in the road state inspection information;

[0027] Sort the first hidden danger category and the second hidden danger category based on the first hidden danger weight and the second hidden danger weight to obtain a hidden danger category table;

[0028] Obtaining a road normal hazard table for the target road based on the time variation characteristics of the second safety hazard set of the target road in all evaluation periods before the current evaluation period;

[0029] Analyze the road state deviation of each second safety hazard set of all evaluation periods before the current evaluation period based on the road normal hazard table, and use the average value of all road state deviations as the state deviation tolerance;

[0030] The hidden danger category table is corrected based on the state deviation tolerance to obtain a second safety hidden danger set.

[0031] Preferably, the road traffic safety hazard improvement evaluation method obtains a road normal hazard table of the target road based on the time variation characteristics of the second safety hazard set of all evaluation periods before the current evaluation period of the target road, including:

[0032] Obtain a historical hazard category table corresponding to the second safety hazard set of all evaluation periods before the current evaluation period for the target road, and generate a first ordinal change curve and a first weight change curve of each historical hazard category as a time change feature based on the ordinal number and the historical hazard weight of the historical hazard category in each historical hazard category table;

[0033] The first ordinal change curves and the first weight change curves of all historical hazard types in the time-varying characteristics are aligned and analyzed to determine the normal ordinal and normal weight of each historical hazard type. The historical hazard types whose normal weights are not less than the normal threshold are sorted according to the normal ordinal to obtain the road normal hazard table.

[0034] Preferably, the road traffic safety hazard improvement evaluation method corrects the hazard category table based on the state deviation tolerance to obtain the second safety hazard set, including:

[0035] Based on the state deviation tolerance and the road normality hidden danger table, calculate the weight range of each hidden danger type in the hidden danger type table;

[0036] Determine the corrected hazard weight of the corresponding hazard type based on the hazard weight and weight range of each hazard type in the hazard type table;

[0037] Reorder all hidden danger weights based on the corrected hidden danger weights to obtain a second safety hidden danger set;

[0038] Among them, the hidden danger weight includes a first hidden danger weight and a second hidden danger weight.

[0039] Preferably, the road traffic safety hazard improvement evaluation method, S4: aggregating and associating the first safety hazard set and the second safety hazard set to obtain a comprehensive safety hazard set existing in the corresponding evaluation period, includes:

[0040] Aggregating the first safety hazard set and the second safety hazard set to obtain a hazard type set;

[0041] Determine the calculation weight of each potential hazard type in the first set of potential safety hazards and the second set of potential safety hazards;

[0042] Based on the hidden danger type association library, the hidden danger type set is re-divided to determine multiple hidden danger association mining subsets;

[0043] Based on the standard correlation between each hidden danger type in the hidden danger correlation mining subset and the remaining hidden danger types in the hidden danger correlation mining subset and the calculation weight of the hidden danger type, the evaluation value of the corresponding hidden danger correlation mining subset is calculated;

[0044] Based on the calculated weights of the hidden danger types included in the hidden danger association mining subset whose evaluation value is not less than the evaluation threshold, determine the associated mining hidden danger types and the corresponding associated hidden danger weights;

[0045] Based on the comprehensive weight, the derived hidden danger weight, the first hidden danger weight, the second hidden danger weight and the associated hidden danger weight, all hidden danger types included in the first safety hidden danger set and the second safety hidden danger set and all associated mined hidden danger types are reordered to obtain a comprehensive safety hidden danger set.

[0046] Preferably, the road traffic safety hazard improvement evaluation method, S6: based on the hazard record thread, real-time evaluation of the improvement evaluation value of the traffic safety hazard improvement plan includes:

[0047] Generate a second ordinal change curve and a second weight change curve for each hidden danger type in the comprehensive safety hidden danger set based on the hidden danger record thread;

[0048] Perform sudden change division and alignment analysis on the second ordinal change curve and the second weight change curve to determine the comprehensive reduction of the corresponding hidden danger types;

[0049] The final hidden danger weight of the corresponding hidden danger type is determined based on the second weight change curve, and the improvement evaluation value of the traffic safety hidden danger improvement plan is calculated based on the final hidden danger weights and comprehensive reduction of all hidden danger types.

[0050] Preferably, the road traffic safety hidden danger improvement evaluation method performs sudden change division and alignment analysis on the second ordinal change curve and the second weight change curve to determine the comprehensive reduction of the corresponding hidden danger type, including:

[0051] Dividing the second ordinal change curve into a first curve segment and a second curve segment based on a sudden change point in the second ordinal change curve, aligning and analyzing the first curve segment and the second curve segment to obtain a first reduction corresponding to the hidden danger type;

[0052] Dividing the second weight change curve into a third curve segment and a fourth curve segment based on a sudden change point in the second weight change curve, aligning and analyzing the third curve segment and the fourth curve segment to obtain a second reduction corresponding to the hidden danger type;

[0053] The average of the first reduction and the second reduction is taken as the comprehensive reduction of the corresponding hidden danger type.

[0054] The present invention proposes a road traffic safety hazard improvement evaluation system, comprising:

[0055] An acquisition module is used to acquire traffic accident records during the implementation of the traffic safety hazard improvement plan;

[0056] A first analysis module, configured to analyze a first safety hazard set based on traffic accidents occurring within an evaluation period in traffic accident records;

[0057] A second analysis module, configured to analyze a second set of potential safety hazards based on the road status information within an evaluation period;

[0058] An association mining module is used to aggregate and associate the first set of potential safety hazards with the second set of potential safety hazards to obtain a set of comprehensive potential safety hazards existing in a corresponding evaluation period;

[0059] A sorting module is used to sort and record the safety hazard sets of all evaluation cycles that are newly obtained in the process of executing the traffic safety hazard improvement plan in chronological order to obtain a hazard recording thread;

[0060] The evaluation module is used to evaluate the improvement value of the traffic safety hazard improvement plan in real time based on the hazard recording thread.

[0061] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0062] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0064] Figure 1 A flow chart of a method for evaluating improvement of road traffic safety hazards in an embodiment of the present invention;

[0065] Figure 2 A flow chart of another method for evaluating improvement of road traffic safety hazards in an embodiment of the present invention;

[0066] Figure 3 Schematic diagram of a road traffic safety hazard improvement evaluation system in an embodiment of the present invention. DETAILED DESCRIPTION

[0067] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0068] Example 1

[0069] The present invention provides a method for evaluating the improvement of road traffic safety hazards, referring to Figure 1 ,include:

[0070] S1: Obtain traffic accident records during the implementation of traffic safety hazard improvement plans;

[0071] S2: Analyze the first safety hazard set based on the traffic accidents that occurred within the evaluation period in the traffic accident records;

[0072] S3: Analyze a second set of safety hazards based on the road status information within the evaluation period;

[0073] S4: Summarize and associate the first set of safety hazards and the second set of safety hazards to obtain a comprehensive set of safety hazards existing in a corresponding assessment period;

[0074] S5: sorting and recording the safety hazard sets of all evaluation periods that are newly obtained in the process of executing the traffic safety hazard improvement plan in chronological order to obtain a hazard recording thread;

[0075] S6: Based on the hidden danger record thread, the improvement evaluation value of the traffic safety hidden danger improvement plan is evaluated in real time.

[0076] In this embodiment, traffic safety hazards include, for example, poor road construction, incomplete traffic facilities, loopholes in the screening of illegal behaviors, etc.

[0077] In this embodiment, the traffic safety hazard improvement plan is a plan for improving traffic safety hazards, such as: adjusting the duration of traffic lights, road maintenance, adding cameras, strengthening vehicle screening plans, etc.

[0078] In this embodiment, the traffic accident record is a record containing information about all traffic accidents that occurred on the target road during the implementation of the traffic safety hazard improvement plan (such as the scale of the accident, the safety hazards caused by the accident, etc.), and the target road is the road whose traffic safety hazards are improved based on the traffic safety hazard improvement plan.

[0079] In this embodiment, the evaluation cycle is a period of multiple improvement evaluation values ​​of the traffic safety hazard improvement plan included in the process of executing the traffic safety hazard improvement plan. It can also be preset, but the evaluation cycle must be smaller than the total period of the process of executing the traffic safety hazard improvement plan.

[0080] In this embodiment, the first safety hazard set is a set of safety hazards analyzed based on traffic accidents occurring within an evaluation period in traffic accident records.

[0081] In this embodiment, the second safety hazard set is a set of safety hazards analyzed based on road status information within an evaluation period.

[0082] In this embodiment, the road state information is information reflecting the real-time state of the target road, such as a road state monitoring image, road state inspection information, and the like.

[0083] In this embodiment, the comprehensive safety hazard set is a set of all safety hazards existing on the target road in a corresponding evaluation period, obtained by aggregating and associating the first safety hazard set and the second safety hazard set and mining them.

[0084] In this embodiment, the hidden danger recording thread is a thread used to record the latest set of safety hidden dangers of all evaluation cycles obtained during the execution of the traffic safety hidden danger improvement plan, and is also a thread used to record the process of safety hidden dangers changing with the evaluation cycle during the execution of the traffic safety hidden danger improvement plan.

[0085] In this embodiment, the improvement evaluation value is an evaluation value of the improvement of the traffic safety hazard improvement plan on the target road under the current execution condition, which is evaluated based on the hazard record thread. The larger the improvement evaluation value, the better the improvement of the traffic safety hazard improvement plan on the target road under the current execution condition, and vice versa.

[0086] The beneficial effects of the above technology are: through the association mining of the first safety hazard set based on the analysis of traffic accident records within the evaluation period in the process of implementing the traffic safety hazard improvement plan and the second safety hazard set based on the analysis of road status information within the evaluation period, it is possible to realize real-time analysis and mining of safety hazards currently existing in the target road during the implementation of the traffic safety hazard improvement plan, and then based on the evaluation of the obtained hazard record thread, it is possible to realize an accurate evaluation of the real-time improvement of traffic safety hazards existing in the target road during the implementation of the traffic safety hazard improvement plan, which provides an important reference for the real-time optimization of the road traffic safety hazard improvement plan and meets the needs of real-time optimization of the road traffic safety hazard improvement plan.

[0087] Embodiment 2:

[0088] Based on Example 1, the road traffic safety hazard improvement evaluation method, S2: based on the traffic accident records that occurred within the evaluation period, analyze the first safety hazard set, refer to Figure 2 ,include:

[0089] S201: Obtain a sub-record thread in the traffic accident record within an evaluation period, and determine a target hidden danger type and hidden danger weight corresponding to the target traffic accident based on the accident scale and accident type of the target traffic accident in the sub-record thread;

[0090] S202: Calculating the comprehensive weight of the target hidden danger type in the time dimension based on the interval time of the target traffic accidents of the same target hidden danger type in the sub-recording thread and the hidden danger weight;

[0091] S203: determining a target derived hazard type and a corresponding derived hazard weight of the target hazard type based on the derived hazard table of the target hazard type and the comprehensive weight;

[0092] S204: Sort all target hidden danger types and all derived hidden danger weights based on the comprehensive weights and derived hidden danger weights to obtain a first safety hidden danger set.

[0093] In this embodiment, the sub-recording thread is a thread for recording a portion of traffic accidents extracted from the traffic accident records that include the traffic accidents occurring on the target road within the evaluation period.

[0094] In this embodiment, the target traffic accident is the traffic accident included in the sub-recording thread.

[0095] In this embodiment, based on the accident scale and accident type of the target traffic accident in the sub-recording thread, the target hidden danger type and hidden danger weight corresponding to the target traffic accident are determined, namely:

[0096] Based on the accident scale and accident type of the target traffic accident, the preset hidden danger correspondence table (a table containing hidden danger types and hidden danger weights corresponding to the accident scale of each traffic accident) is queried to obtain the target hidden danger type and hidden danger weight.

[0097] In this embodiment, the target hidden danger type is the type of traffic safety hidden danger existing on the target road analyzed by querying a preset hidden danger correspondence table based on the accident scale and accident type of the target traffic accident.

[0098] In this embodiment, the hidden danger weight is a value representing the severity of the corresponding target hidden danger type existing in the target road.

[0099] In this embodiment, based on the interval time of target traffic accidents of the same target hidden danger type in the sub-recording thread and the hidden danger weight, the comprehensive weight of the target hidden danger type in the time dimension is calculated, including:

[0100]

[0101] In the formula, is the comprehensive weight of the target hidden danger type in the time dimension, is the total number of target traffic accidents of the target hidden danger type currently calculated in the sub-recording thread, The sub-record thread contains The target traffic accident of the target hidden danger type currently calculated and the The interval time between the target traffic accidents of the target hidden danger type currently calculated, The total duration of the child record thread, The sub-record thread contains The hidden danger weight of the target traffic accident of the currently calculated target hidden danger type, The sub-record thread contains The hidden danger weight of the target traffic accident of the target hidden danger type currently calculated;

[0102] Based on the above formula, we can accurately calculate the comprehensive weight of the target hidden danger type in the time dimension based on the interval time and hidden danger weight of the target traffic accidents of the same target hidden danger type in the sub-recording thread, by considering the characteristics of the hidden danger weight of the target traffic accidents of the same target hidden danger type in the time dimension.

[0103] In this embodiment, the comprehensive weight is a value representing the severity of the corresponding target hidden danger type in the target road, which is determined by analyzing the characteristics of the occurrence time of the corresponding target hidden danger type in the sub-recording thread in the time dimension.

[0104] In this embodiment, the derived risk table is a list containing other risk types that can be derived and mined based on the corresponding target risk type and their corresponding derived risk weights.

[0105] In this embodiment, the target derived hidden danger type is other hidden danger types that can be derived and mined based on the corresponding target hidden danger type and are determined based on the derived hidden danger table and the comprehensive weight.

[0106] In this embodiment, the derived risk weight is a measure of the correlation between the target derived risk type and the corresponding target risk type.

[0107] In this embodiment, the types of hidden dangers (i.e., traffic safety hidden dangers) include, for example, poor road construction, incomplete traffic facilities, loopholes in the screening of illegal behaviors, etc.

[0108] In this embodiment, all target potential hazard types and all derived potential hazard weights are sorted based on the comprehensive weight and the derived potential hazard weight to obtain a first safety potential hazard set, which is:

[0109] All target hidden danger types and all derived hidden danger weights are sorted according to the principle of comprehensive weight and derived hidden danger weight from large to small to obtain the first safety hidden danger set.

[0110] In this embodiment, the first safety hazard set is a set obtained by sorting all target hazard types and all derived hazard weights based on the comprehensive weights and derived hazard weights.

[0111] The beneficial effects of the above technology are: by analyzing the accident scale, accident type and characteristics in the time dimension of traffic accidents that occurred on the target road during the evaluation period, the comprehensive weight that characterizes the severity of the corresponding target hidden danger type in the target road is determined, and then based on the comprehensive weight, the target derivative hidden danger type and the corresponding derivative hidden danger weight of the target hidden danger type are determined in the derivative hidden danger table, which not only realizes the direct mining of safety hazards based on traffic accidents, but also realizes the derivative mining of safety hazards, and realizes the comprehensive mining of traffic safety hazards on the road.

[0112] Embodiment 3:

[0113] On the basis of Example 2, the road traffic safety hazard improvement evaluation method, S203: based on the derived hazard table of the target hazard type and the comprehensive weight, determining the target derived hazard type and the corresponding derived hazard weight of the target hazard type, includes:

[0114] Determine the target derivative depth based on the comprehensive weight of the target hidden danger type, and determine the target derivative hidden danger type whose derivative depth is within the target derivative depth in the derivative hidden danger table of the target hidden danger type;

[0115] Based on the comprehensive weight of the target hidden danger type and the derived weight of the target derived hidden danger type, the derived hidden danger weight of the corresponding target derived hidden danger type is calculated.

[0116] In this embodiment, the target derivation depth is determined based on the comprehensive weight of the target hidden danger types, for example:

[0117] Based on the comprehensive weight, the derived depth table of the target hidden danger type (including a table of target derived depths corresponding to different ranges of the comprehensive weight of the corresponding target hidden danger type) is queried to determine the corresponding target derived depth.

[0118] In this embodiment, the target derivation depth is a value obtained by querying the derivation depth table of the target hidden danger type based on the comprehensive weight of the target hidden danger type, which represents the degree of derivation of the target derived hidden danger type based on the target hidden danger type.

[0119] In this embodiment, based on the comprehensive weight of the target hidden danger type and the derived weight of the target derived hidden danger type, the derived hidden danger weight corresponding to the target derived hidden danger type is calculated, that is,

[0120] The product of the comprehensive weight of the target hazard type and the derived weight of the target derived hazard type is taken as the derived hazard weight of the corresponding target derived hazard type.

[0121] The beneficial effects of the above technology are: based on the target derivative depth determined by the comprehensive weight, the target derived hazard type is determined in the derived hazard table of the target hazard type, and the target hazard type is derived by using the comprehensive weight representing the severity of the corresponding target hazard type in the target road as the derived depth, thereby obtaining the target derived hazard type and its derived hazard weight.

[0122] Embodiment 4:

[0123] Based on Example 1, the road traffic safety hazard improvement evaluation method, S3: analyzing a second safety hazard set based on road state information within an evaluation period, includes:

[0124] Obtaining road status information within the latest evaluation period, wherein the road status information includes: road status monitoring images and road status inspection information;

[0125] Based on the hidden danger deep learning model, all road status monitoring images in the latest assessment period are identified to obtain the first hidden danger type and the corresponding first hidden danger weight;

[0126] Determine a second hidden danger type and a corresponding second hidden danger weight based on the hidden danger keyword in the road state inspection information;

[0127] Sort the first hidden danger category and the second hidden danger category based on the first hidden danger weight and the second hidden danger weight to obtain a hidden danger category table;

[0128] Obtaining a road normal hazard table for the target road based on the time variation characteristics of the second safety hazard set of the target road in all evaluation periods before the current evaluation period;

[0129] Analyze the road state deviation of each second safety hazard set of all evaluation periods before the current evaluation period based on the road normal hazard table, and use the average value of all road state deviations as the state deviation tolerance;

[0130] The hidden danger category table is corrected based on the state deviation tolerance to obtain a second safety hidden danger set.

[0131] In this embodiment, the road state monitoring image is a monitoring image including the object scene of the target road.

[0132] In this embodiment, the road status inspection information includes information obtained from road inspections, such as manual inspection information, inspection information of an automatic inspection system, and the like.

[0133] In this embodiment, the hidden danger deep learning model is a model that is trained in advance based on a large number of road status monitoring images and is used to identify the types of traffic safety hidden dangers and their hidden danger weights in the road status monitoring images.

[0134] In this embodiment, the first hidden danger type is the hidden danger type existing in the target road within the evaluation period, which is identified in all road status monitoring images within the latest evaluation period based on the hidden danger deep learning model.

[0135] In this embodiment, the first hidden danger weight is a value representing the severity of the first hidden danger type existing in the target road within the evaluation period.

[0136] In this embodiment, the hidden danger keywords are keywords contained in the road status inspection information that can determine the types of safety hazards existing on the target road, such as uneven road surface, traffic light device failure, etc.

[0137] In this embodiment, the second hidden danger type is the type of safety hidden danger existing in the target road during the assessment period determined based on the hidden danger keyword in the road status inspection information.

[0138] In this embodiment, the second hidden danger weight is a value representing the severity of the second hidden danger type existing in the target road within the evaluation period.

[0139] In this embodiment, the hidden danger category table is a table obtained by sorting the first hidden danger category and the second hidden danger category based on the first hidden danger weight and the second hidden danger weight.

[0140] In this embodiment, the time variation feature is the feature of the second potential safety hazard set in all evaluation periods before the current evaluation period that varies with time.

[0141] In this embodiment, the road normal hazard table is a table containing the safety hazards that stably exist on the target road under normal conditions (i.e., before the implementation of the road traffic safety hazard improvement plan) and their corresponding hazard weights, which is determined based on the time-varying characteristics of the second safety hazard set of all evaluation periods of the target road before the current evaluation period.

[0142] In this embodiment, the road state deviation is a numerical value representing the degree of deviation of the safety hazards of the second safety hazard set in the evaluation period relative to the road normal hazard table.

[0143] In this embodiment, the state deviation tolerance is a value representing the tolerable degree of deviation of the second safety hazard set relative to the road normal hazard table.

[0144] The beneficial effects of the above technology are: based on the time change characteristics of the second safety hazard set of all evaluation cycles before the current evaluation cycle, the first hazard type and the corresponding first hazard weight as well as the second hazard type and the corresponding second hazard weight determined by respectively analyzing the road state monitoring image and the road state inspection information in the road state information are corrected, so as to realize the mining of traffic safety hazard types based on road state information, and also realize the correction of the analysis and mining results, thereby ensuring the comprehensiveness and accuracy of the mining of safety hazard types.

[0145] Embodiment 5:

[0146] On the basis of Example 4, the road traffic safety hazard improvement evaluation method obtains a road normal hazard table of the target road based on the time variation characteristics of the second safety hazard set of all evaluation periods before the current evaluation period of the target road, including:

[0147] Obtain a historical hazard category table corresponding to the second safety hazard set of all evaluation periods before the current evaluation period for the target road, and generate a first ordinal change curve and a first weight change curve of each historical hazard category as a time change feature based on the ordinal number and the historical hazard weight of the historical hazard category in each historical hazard category table;

[0148] The first ordinal change curves and the first weight change curves of all historical hazard types in the time-varying characteristics are aligned and analyzed to determine the normal ordinal and normal weight of each historical hazard type. The historical hazard types whose normal weights are not less than the normal threshold are sorted according to the normal ordinal to obtain the road normal hazard table.

[0149] In this embodiment, the historical hidden danger category table is a hidden danger category table (corrected) corresponding to the second safety hidden danger set in the evaluation period before the current evaluation period.

[0150] In this embodiment, the historical hidden danger weight is the hidden danger weight (first hidden danger weight or second hidden danger weight) corresponding to each hidden danger type (first hidden danger type or second hidden danger type) in the historical hidden danger type table.

[0151] In this embodiment, the first ordinal number change curve is a curve obtained by connecting the ordinals of each historical hidden danger type in the historical hidden danger type table of all evaluation periods before the current evaluation period in time sequence.

[0152] In this embodiment, the first weight change curve is a curve obtained by connecting the historical hazard weights of each historical hazard type in the historical hazard type table of all evaluation periods before the current evaluation period in time series.

[0153] In this embodiment, the first ordinal change curves and the first weight change curves of all historical hidden danger types in the time change characteristics are aligned and analyzed to determine the normal ordinal and normal weight of each historical hidden danger type, which are:

[0154] The first ordinal change curve and the first weight change curve of all historical hidden danger types in the time change characteristics are aligned to determine the sudden change points in the aligned curve combination, and the average value of all ordinals between the sudden change points in the first ordinal change curve is used as the normal coefficient, and the average value of all hidden danger weights between the sudden change points in the first weight change curve is used as the normal weight.

[0155] In this embodiment, the normal weight is the hidden danger weight corresponding to the hidden danger type in the hidden danger type table of the target road under normal conditions.

[0156] In this embodiment, the normal ordinal number is the ordinal number of the corresponding hidden danger type in the hidden danger type table of the target road under normal conditions.

[0157] In this embodiment, the normal threshold is a preset normal ordinal threshold used when screening out the types of hidden dangers included in the road normal hidden danger table.

[0158] The beneficial effect of the above technology is: by analyzing the changes in the ordinal numbers and hazard weights of historical hazard types in the historical hazard type table over time, the normal ordinal number and normal weight of each historical hazard type are determined, and then a road normal hazard table is obtained that contains the safety hazards that stably exist on the target road under normal conditions (that is, before the implementation of the road traffic safety hazard improvement plan) and their corresponding hazard weights.

[0159] Embodiment 6:

[0160] On the basis of Example 4, the road traffic safety hazard improvement evaluation method corrects the hazard category table based on the state deviation tolerance to obtain a second safety hazard set, including:

[0161] Based on the state deviation tolerance and the road normality hidden danger table, calculate the weight range of each hidden danger type in the hidden danger type table;

[0162] Determine the corrected hazard weight of the corresponding hazard type based on the hazard weight and weight range of each hazard type in the hazard type table;

[0163] Reorder all hidden danger weights based on the corrected hidden danger weights to obtain a second safety hidden danger set;

[0164] Among them, the hidden danger weight includes a first hidden danger weight and a second hidden danger weight.

[0165] In this embodiment, based on the state deviation tolerance and the road normality hidden danger table, the weight range of each hidden danger type in the hidden danger type table is calculated, that is,

[0166] The product of the difference between 1 and the state deviation tolerance and the hidden danger weight of the corresponding hidden danger type in the road normal hidden danger table is taken as the weight fluctuation value of the corresponding hidden danger type, the difference between the hidden danger weight and the weight fluctuation value of the hidden danger type in the hidden danger type table is taken as the lower limit of the weight range, and the sum of the hidden danger weight and the weight fluctuation value of the hidden danger type in the hidden danger type table is taken as the upper limit of the weight range, so as to obtain the weight range of the hidden danger type.

[0167] In this embodiment, the weight range is the value range of the corresponding hidden danger type in the hidden danger type table determined based on the state deviation tolerance.

[0168] In this embodiment, the corrected hidden danger weight is the hidden danger weight of the hidden danger type after correction.

[0169] In this embodiment, based on the risk weight and weight range of each risk type in the risk type table, the corrected risk weight of the corresponding risk type is determined, that is:

[0170] The hidden danger weight that does not exceed the weight range is used as the corrected hidden danger weight, the hidden danger weight that is less than the lower limit of the weight range is set to the lower limit of the weight range, and the hidden danger weight that is greater than the upper limit of the weight range is set to the upper limit of the weight range.

[0171] The beneficial effects of the above technology are: based on the state deviation tolerance and the road normal hazard table, the weight range of each hazard type is calculated, and then the hazard weight is controlled within the weight range, thereby achieving debiasing correction of the hazard type table.

[0172] Embodiment 7:

[0173] Based on Example 1, the road traffic safety hazard improvement evaluation method, S4: Summarize and associate the first safety hazard set and the second safety hazard set to obtain a comprehensive safety hazard set existing in the corresponding evaluation period, including:

[0174] Aggregating the first safety hazard set and the second safety hazard set to obtain a hazard type set;

[0175] Determine the calculation weight of each potential hazard type in the first set of potential safety hazards and the second set of potential safety hazards;

[0176] Based on the hidden danger type association library, the hidden danger type set is re-divided to determine multiple hidden danger association mining subsets;

[0177] Based on the standard correlation between each hidden danger type in the hidden danger correlation mining subset and the remaining hidden danger types in the hidden danger correlation mining subset and the calculation weight of the hidden danger type, the evaluation value of the corresponding hidden danger correlation mining subset is calculated;

[0178] Based on the calculated weights of the hidden danger types included in the hidden danger association mining subset whose evaluation value is not less than the evaluation threshold, determine the associated mining hidden danger types and the corresponding associated hidden danger weights;

[0179] Based on the calculated weights and associated hidden danger weights, the hidden danger category set and all associated mined hidden danger categories are reordered to obtain a comprehensive safety hidden danger set.

[0180] In this embodiment, the potential hazard type set is a set obtained by combining the first security potential hazard set and the second security potential hazard set.

[0181] In this embodiment, the calculation weight of each potential hazard type in the first potential hazard set and the second potential hazard set is determined as follows:

[0182] The comprehensive weight, derived hidden danger weight, first hidden danger weight and second hidden danger weight are used as the calculation weights for the corresponding hidden danger types. When a hidden danger type has multiple weights (that is, it has two or more values ​​of the comprehensive weight, derived hidden danger weight, first hidden danger weight and second hidden danger weight at the same time), the maximum value of the multiple weights is used as the calculation weight for the corresponding hidden danger type.

[0183] In this embodiment, the hidden danger type association library is a preset database containing multiple associated hidden danger type sets, and the associated hidden danger type set is a set consisting of multiple hidden danger types with associated relationships. Each associated hidden danger type set has a corresponding preset associated mining hidden danger type.

[0184] In this embodiment, the hidden danger association mining subset is a set obtained by re-dividing the hidden danger type set based on the associated hidden danger type set contained in the hidden danger type association library, and the hidden danger types contained in the hidden danger association mining subset are consistent with the hidden danger types contained in the preset associated hidden danger type set.

[0185] In this embodiment, the standard correlation degree is a value preset in the hidden danger type correlation library, which represents the correlation degree between hidden danger types.

[0186] In this embodiment, based on the standard correlation between each hidden danger type in the hidden danger association mining subset and the remaining hidden danger types in the hidden danger association mining subset and the calculation weight of the hidden danger type, the evaluation value of the corresponding hidden danger association mining subset is calculated, that is:

[0187] The average value of the standard correlation between the hidden danger type and all the remaining hidden danger types in the hidden danger association mining subset is taken as the exclusive correlation, and the average value of the product of the average value of the exclusive correlation of all hidden danger types in the hidden danger association mining subset and the corresponding calculation weight is taken as the evaluation value of the corresponding hidden danger association mining subset.

[0188] In this embodiment, the evaluation threshold is a preset evaluation value threshold used to filter out a hidden danger association mining subset in which the degree of association between hidden danger types meets the requirements.

[0189] In this embodiment, based on the calculated weights of the hidden danger types included in the hidden danger association mining subset whose evaluation value is not less than the evaluation threshold, the associated mining hidden danger types and the corresponding associated hidden danger weights are determined, namely:

[0190] Determine the associated mining hidden danger category corresponding to the associated hidden danger category set corresponding to the hidden danger associated mining subset whose evaluation value is not less than the evaluation threshold;

[0191] The product of the average value of the calculated weights of the hidden danger types included in the hidden danger association mining subset and the degree of association between the corresponding associated hidden danger type set and the corresponding associated mining hidden danger type is taken as the corresponding associated hidden danger weight.

[0192] The beneficial effects of the above technology are: by re-dividing the hazard type set obtained by aggregating the first safety hazard set and the second safety hazard set, a hazard association mining subset is obtained, and based on the standard correlation between different hazard types in the hazard association mining subset and the calculated weight of the hazard type, an evaluation value of the hazard association mining subset is calculated, and then the hazard association mining subset with load requirements is selected based on the calculated weight of the hazard weight on the target road, and then the corresponding associated mining hazard type and the corresponding associated hazard weight are determined based on the hazard association mining subset, so as to realize the associated mining of the first safety hazard set and the second safety hazard set, so that the scope of safety hazard mining is wider.

[0193] Embodiment 8:

[0194] Based on Example 1, the road traffic safety hazard improvement evaluation method, S6: based on the hazard record thread, real-time evaluation of the improvement evaluation value of the traffic safety hazard improvement plan includes:

[0195] Generate a second ordinal change curve and a second weight change curve for each hidden danger type in the comprehensive safety hidden danger set based on the hidden danger record thread;

[0196] Perform sudden change division and alignment analysis on the second ordinal change curve and the second weight change curve to determine the comprehensive reduction of the corresponding hidden danger types;

[0197] The final hidden danger weight of the corresponding hidden danger type is determined based on the second weight change curve, and the improvement evaluation value of the traffic safety hidden danger improvement plan is calculated based on the final hidden danger weights and comprehensive reduction of all hidden danger types.

[0198] In this embodiment, the second ordinal number change curve is a curve obtained by sorting the ordinals of all comprehensive safety potential hazards sets included in the potential hazard type in the potential hazard recording thread in chronological order.

[0199] In this embodiment, the second weight change curve is a curve obtained by sorting the calculated weights or associated hidden danger weights corresponding to all comprehensive safety hidden danger sets contained in the hidden danger recording thread in chronological order.

[0200] In this embodiment, the comprehensive reduction is the value of the comprehensive reduction in the weight representing the hidden danger type from the beginning of the implementation of the road traffic safety hidden danger improvement plan to the present, which is determined after sudden change division and alignment analysis of the second ordinal change curve and the second weight change curve.

[0201] In this embodiment, the final hidden danger weight is the average value of the weight values ​​(calculated weight or associated hidden danger weight) of all evaluation cycles before the sudden change point of the hidden danger type in the second weight change curve, and is also a numerical value that represents the comprehensive severity of the corresponding hidden danger type on the target road before it is improved.

[0202] In this embodiment, based on the final hazard weights and comprehensive reductions of all hazard types, the improvement evaluation value of the traffic safety hazard improvement plan is calculated, which is:

[0203] The average value of the product of the final hazard weights and the comprehensive reduction of all hazard types is taken as the improvement evaluation value of the traffic safety hazard improvement plan, which is:

[0204]

[0205] In the formula, is the improvement evaluation value of the traffic safety hazard improvement plan, is the total number of all hidden danger types that have appeared in the second ordinal change curve (or the second weight change curve), For the The final hazard weight of each hazard type, For the The comprehensive reduction of various types of hidden dangers;

[0206] Based on the above formula, the evaluation value representing the improvement of safety hazards existing in the target road under the current implementation progress of the traffic safety hazard improvement plan can be accurately calculated from the comprehensive reduction of hazard types and the final hazard weight before improvement.

[0207] The beneficial effect of the above technology is: through the sudden change division and alignment analysis of the curves of the ordinal numbers and hazard weights of the hazard types in the comprehensive safety hazard set in different evaluation cycles during the implementation of the traffic safety hazard improvement plan, the comprehensive reduction in the ordinal numbers and weights is determined, and then combined with the hazard weights before the improvement, the evaluation value representing the improvement of the safety hazards existing in the target road under the current implementation progress of the traffic safety hazard improvement plan is accurately calculated.

[0208] Embodiment 9:

[0209] On the basis of Example 8, the road traffic safety hidden danger improvement evaluation method performs sudden change division and alignment analysis on the second ordinal change curve and the second weight change curve to determine the comprehensive reduction of the corresponding hidden danger type, including:

[0210] Dividing the second ordinal change curve into a first curve segment and a second curve segment based on a sudden change point in the second ordinal change curve, aligning and analyzing the first curve segment and the second curve segment to obtain a first reduction corresponding to the hidden danger type;

[0211] Dividing the second weight change curve into a third curve segment and a fourth curve segment based on a sudden change point in the second weight change curve, aligning and analyzing the third curve segment and the fourth curve segment to obtain a second reduction corresponding to the hidden danger type;

[0212] The average of the first reduction and the second reduction is taken as the comprehensive reduction of the corresponding hidden danger type.

[0213] In this embodiment, the sudden change point is a data point that suddenly increases or suddenly decreases in the second ordinal change curve or the second weight change curve.

[0214] In this embodiment, the first curve segment and the second curve segment are two curve segments obtained by dividing the second ordinal number change curve based on the sudden change point in the second ordinal number change curve.

[0215] In this embodiment, the first curve segment and the second curve segment are aligned and analyzed to obtain a first reduction corresponding to the hidden danger type, including:

[0216] The first curve segment is aligned with the second curve segment, and the ratio of the difference between the values ​​of the data points in the curve segment after the sudden change point in the second ordinal change curve and the values ​​of the data points in the curve segment before the sudden change point to the data points in the curve segment before the sudden change point is determined as the first reduction.

[0217] In this embodiment, the third curve segment and the fourth curve segment are two curve segments obtained by dividing the second weight change curve based on the sudden change point in the second weight change curve.

[0218] In this embodiment, the third curve segment and the fourth curve segment are aligned and analyzed to obtain the second reduction amplitude corresponding to the hidden danger type, which is:

[0219] The third curve segment and the fourth curve segment are aligned, and the ratio of the difference between the values ​​of the data points in the curve segment after the sudden change point in the second weight change curve and the values ​​of the data points in the curve segment before the sudden change point to the data points in the curve segment before the sudden change point is taken as the second reduction.

[0220] The beneficial effects of the above technology are: taking the sudden change point as the time point when the corresponding hidden danger type is improved, based on the data reduction in the curve segments before and after the sudden change point, the first reduction in the ordinal number of the hidden danger type and the second reduction in the hidden danger weight are determined, and then a more reasonable comprehensive reduction is obtained by taking the average, and the accurate comprehensive reduction of the hidden danger type is determined by analyzing the ordinal number of the hidden danger weight and the hidden danger weight.

[0221] Embodiment 10:

[0222] The present invention provides a road traffic safety hazard improvement evaluation system, referring to Figure 3 ,include:

[0223] An acquisition module is used to acquire traffic accident records during the implementation of the traffic safety hazard improvement plan;

[0224] A first analysis module, configured to analyze a first safety hazard set based on traffic accidents occurring within an evaluation period in traffic accident records;

[0225] A second analysis module, configured to analyze a second set of potential safety hazards based on the road status information within an evaluation period;

[0226] An association mining module is used to aggregate and associate the first set of potential safety hazards with the second set of potential safety hazards to obtain a set of comprehensive potential safety hazards existing in a corresponding evaluation period;

[0227] A sorting module is used to sort and record the safety hazard sets of all evaluation cycles that are newly obtained in the process of executing the traffic safety hazard improvement plan in chronological order to obtain a hazard recording thread;

[0228] The evaluation module is used to evaluate the improvement value of the traffic safety hazard improvement plan in real time based on the hazard recording thread.

[0229] The beneficial effects of the above technology are: through the association mining of the first safety hazard set based on the analysis of traffic accident records within the evaluation period in the process of implementing the traffic safety hazard improvement plan and the second safety hazard set based on the analysis of road status information within the evaluation period, it is possible to realize real-time analysis and mining of safety hazards currently existing in the target road during the implementation of the traffic safety hazard improvement plan, and then based on the evaluation of the obtained hazard record thread, it is possible to realize an accurate evaluation of the real-time improvement of traffic safety hazards existing in the target road during the implementation of the traffic safety hazard improvement plan, which provides an important reference for the real-time optimization of the road traffic safety hazard improvement plan and meets the needs of real-time optimization of the road traffic safety hazard improvement plan.

[0230] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for evaluating improvement of hidden dangers in road traffic safety, characterized in that: include: S1: Obtain traffic accident records during the implementation of traffic safety hazard improvement plans; S2: Analyze the first safety hazard set based on the traffic accidents that occurred within the evaluation period in the traffic accident records; S3: Analyze a second set of safety hazards based on the road status information within the evaluation period; S4: Summarize and associate the first set of safety hazards and the second set of safety hazards to obtain a comprehensive set of safety hazards existing in a corresponding assessment period; S5: sorting and recording the safety hazard sets of all evaluation periods that are newly obtained in the process of executing the traffic safety hazard improvement plan in chronological order to obtain a hazard recording thread; S6: Based on the hidden danger record thread, the improvement evaluation value of the traffic safety hidden danger improvement plan is evaluated in real time; Wherein, step S3: analyzing a second set of potential safety hazards based on the road status information within the evaluation period includes: Obtaining road status information within the latest evaluation period, wherein the road status information includes: road status monitoring images and road status inspection information; Based on the hidden danger deep learning model, all road status monitoring images in the latest assessment period are identified to obtain the first hidden danger type and the corresponding first hidden danger weight; Determine a second hidden danger type and a corresponding second hidden danger weight based on the hidden danger keyword in the road state inspection information; Sort the first hidden danger category and the second hidden danger category based on the first hidden danger weight and the second hidden danger weight to obtain a hidden danger category table; Based on the time variation characteristics of the second safety hazard set of the target road in all evaluation periods before the current evaluation period, a road normal hazard table of the target road is obtained; Analyze the road state deviation of each second safety hazard set of all evaluation periods before the current evaluation period based on the road normal hazard table, and use the average value of all road state deviations as the state deviation tolerance; The hidden danger category table is corrected based on the state deviation tolerance to obtain a second safety hidden danger set.

2. The road traffic safety hazard improvement evaluation method according to claim 1, characterized in that: S2: Analyze the first set of safety hazards based on the traffic accidents that occurred during the evaluation period in the traffic accident records, including: S201: Obtain a sub-record thread in the traffic accident record within an evaluation period, and determine a target hidden danger type and hidden danger weight corresponding to the target traffic accident based on the accident scale and accident type of the target traffic accident in the sub-record thread; S202: Calculating the comprehensive weight of the target hidden danger type in the time dimension based on the interval time of the target traffic accidents of the same target hidden danger type in the sub-recording thread and the hidden danger weight; S203: determining a target derived hazard type and a corresponding derived hazard weight of the target hazard type based on the derived hazard table of the target hazard type and the comprehensive weight; S204: Sort all target hidden danger types and all derived hidden danger weights based on the comprehensive weights and derived hidden danger weights to obtain a first safety hidden danger set.

3. The road traffic safety hazard improvement evaluation method according to claim 2 is characterized in that: The hidden danger category table is corrected based on the state deviation tolerance to obtain a second set of safety hidden dangers, including: Based on the state deviation tolerance and the road normality hidden danger table, calculate the weight range of each hidden danger type in the hidden danger type table; Determine the corrected hazard weight of the corresponding hazard type based on the hazard weight and weight range of each hazard type in the hazard type table; Reorder all hidden danger weights based on the corrected hidden danger weights to obtain a second safety hidden danger set; Among them, the hidden danger weight includes a first hidden danger weight and a second hidden danger weight.

4. The road traffic safety hazard improvement evaluation method according to claim 3 is characterized in that: S4: Summarize and correlate the first set of potential safety hazards and the second set of potential safety hazards to obtain a comprehensive set of potential safety hazards existing in the corresponding assessment period, including: Aggregating the first safety hazard set and the second safety hazard set to obtain a hazard type set; Determine the calculation weight of each potential hazard type in the first safety potential hazard set and the second safety potential hazard set; Based on the hidden danger type association library, the hidden danger type set is re-divided to determine multiple hidden danger association mining subsets; Based on the standard correlation between each hidden danger type in the hidden danger correlation mining subset and the remaining hidden danger types in the hidden danger correlation mining subset and the calculation weight of the hidden danger type, the evaluation value of the corresponding hidden danger correlation mining subset is calculated; Based on the calculated weights of the hidden danger types included in the hidden danger association mining subset whose evaluation value is not less than the evaluation threshold, determine the associated mining hidden danger types and the corresponding associated hidden danger weights; Based on the comprehensive weight, the derived hidden danger weight, the first hidden danger weight, the second hidden danger weight and the associated hidden danger weight, all hidden danger types included in the first safety hidden danger set and the second safety hidden danger set and all associated mined hidden danger types are reordered to obtain a comprehensive safety hidden danger set.

5. Road traffic safety hazard improvement evaluation system, characterized by: A method for improving and evaluating road traffic safety hazards according to any one of claims 1 to 4, comprising: An acquisition module is used to acquire traffic accident records during the implementation of the traffic safety hazard improvement plan; A first analysis module, configured to analyze a first safety hazard set based on traffic accidents occurring within an evaluation period in traffic accident records; A second analysis module, configured to analyze a second set of potential safety hazards based on the road status information within an evaluation period; An association mining module is used to aggregate and associate the first set of potential safety hazards with the second set of potential safety hazards to obtain a set of comprehensive potential safety hazards existing in a corresponding evaluation period; A sorting module is used to sort and record the safety hazard sets of all evaluation cycles that are newly obtained in the process of executing the traffic safety hazard improvement plan in chronological order to obtain a hazard recording thread; The evaluation module is used to evaluate the improvement value of the traffic safety hazard improvement plan in real time based on the hazard recording thread.

Images