CN112966941B - Accident black spot identification method and system based on traffic accident big data - Google Patents
Accident black spot identification method and system based on traffic accident big data Download PDFInfo
- Publication number
- CN112966941B CN112966941B CN202110255857.8A CN202110255857A CN112966941B CN 112966941 B CN112966941 B CN 112966941B CN 202110255857 A CN202110255857 A CN 202110255857A CN 112966941 B CN112966941 B CN 112966941B
- Authority
- CN
- China
- Prior art keywords
- accident
- traffic
- data
- point
- points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 206010039203 Road traffic accident Diseases 0.000 title claims abstract description 92
- 238000000034 method Methods 0.000 title claims abstract description 30
- 206010027146 Melanoderma Diseases 0.000 title claims abstract description 17
- 238000003064 k means clustering Methods 0.000 claims abstract description 4
- 230000034994 death Effects 0.000 claims description 21
- 230000006378 damage Effects 0.000 claims description 19
- 208000027418 Wounds and injury Diseases 0.000 claims description 16
- 208000014674 injury Diseases 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 6
- 239000000463 material Substances 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000010438 heat treatment Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 abstract 1
- 238000007726 management method Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000011835 investigation Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Primary Health Care (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Security & Cryptography (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses an accident black spot identification method based on traffic accident big data, which comprises the following steps: s100: acquiring road network data and traffic accident data in a preset time period; s200: matching the positions of traffic accident data and road network data; s300: identifying accident black spots, the steps further comprise: s301: extracting accident high incidence points; s302: clustering the accident high-incidence points by using a K-means clustering method based on the position information of the accident high-incidence points; s303: calculating the center point of each cluster respectively for each cluster obtained in the sub-step S302; s304: and all accident points in the preset radius range form black point areas by taking the central points as the centers respectively. The invention extracts objective regularity information such as time-space distribution characteristics of traffic accidents based on the traffic accident data, and can accurately identify traffic accident black spots on the road.
Description
Technical Field
The invention belongs to the technical field of intelligent transportation, and particularly relates to an accident black spot identification method and system based on traffic accident big data.
Background
In recent years, with the rapid development of economic construction in China and the continuous improvement of living standard of people, the maintenance amount of urban motor vehicles is rapidly increased, and the life of people is also more and more convenient. But the contradiction between the rapidly growing number of motor vehicles and the limited road infrastructure is also becoming serious, and a series of road safety problems such as traffic jams, disordered traffic and traffic accidents frequently occur in urban road networks. According to statistics issued by world health organization, the casualties caused by road traffic accidents all over the world reach 125 ten thousand each year, and the total economic loss exceeds 5000 hundred million dollars. Road traffic accidents are always in a high-rise situation, so that not only is certain potential safety hazard brought to public travel and social stability, but also great influence is brought to healthy running of urban traffic. Therefore, traffic accidents have become one of the most important public safety problems, and the management and protection of traffic accidents have attracted close attention from governments around the world. In recent years, road safety is improved through science and technology, and foreseeable potential safety hazards and traffic accidents are actively handled and prevented, so that the maximization of the road utilization rate becomes a hot problem of academic research.
One of the important directions for researching road traffic safety is to identify traffic accident black spots of road networks. At present, a common traffic accident black spot recognition method mostly constructs a corresponding mathematical model based on various factors (such as weather, road conditions, illumination, road density and the like) influencing the occurrence of an accident, takes the constructed mathematical model as an input vector of a traffic accident prediction model, takes the occurrence of the traffic accident as an output vector, and trains the constructed traffic accident prediction model; or the prediction problem of the traffic accident is classified as a classification problem or a regression problem. Prediction of traffic accidents remains a challenging problem. First, the cause of traffic accidents is complex, and besides some common factors, traffic design of road sections or intersections is an important cause of accidents, and the causes of traffic accidents in each place may be different. For example, the cause of a motor vehicle accident is likely to be different from the cause of a non-motor vehicle accident. However, among the influencing factors, the most important reason is the traffic design of road sections or intersections, and unreasonable and irregular design is a direct cause of traffic accidents.
If the black points of the traffic accidents in the urban road network can be identified, corresponding investigation and repair are carried out on the relevant road sections or intersections based on the identified black points of the traffic accidents, and corresponding measures are adopted by the auxiliary traffic management department. The method not only can help to improve the road traffic environment and reduce the occurrence rate of traffic accidents; the traffic management department can be guided to comprehensively manage each influencing factor through feasible safety measures and precaution means, so that the occurrence probability of road traffic accidents is reduced on the whole, or the efficiency of processing the traffic accidents in real time is improved.
Disclosure of Invention
The invention aims to provide an accident black spot identification method and system based on traffic accident big data.
The invention provides an accident black spot identification method based on traffic accident big data, which comprises the following steps:
S100: acquiring road network data and traffic accident data of a preset time period, wherein the traffic accident data at least comprises time and position information of traffic accidents and accident degree, and the accident degree further comprises three types of material loss, injury and death;
s200: matching the positions of traffic accident data and road network data;
S300: identifying accident black spots, the steps further comprise:
s301: the method for extracting the accident high incidence point comprises the following steps:
Traversing each accident point, namely, a traffic accident occurrence point, obtaining other accident points in a preset range S k around the current accident point, and calculating an accident heating value H (S k) in a preset range S k around the current accident point:
Wherein: the weight values of the degrees of death, injury and physical damage are respectively represented by P (1), P (2) and P (3), and are all in the range of [0,1], and the weight values are artificially set; when P (1) =p (2) =p (3) =1, H (S k) is equivalent to the number of other accident points within a preset range around the current accident point; d r denotes the R-th death accident within the preset range S k, R denotes the total number of death accidents within the preset range S k, Representing a count of death incidents within a preset range S k; h j represents the jth injury incident within the preset range S k, J represents the total number of injuries within the preset range S k,/>Indicating a count of wound incidents within a preset range S k; m p represents the P-th loss accident within the preset range S k, P represents the total number of loss accidents within the preset range S k,/>A count indicating a loss of material accident within a predetermined range S k;
Judging whether the accident thermal value H (S k) of the current accident point in S k is larger than a preset thermal value threshold value, and if so, considering the current accident point as an accident high-incidence point;
completing traversal to obtain an accident high-occurrence point set;
S302: clustering the accident high-incidence points by using a K-means clustering method based on the position information of the accident high-incidence points;
S303: and (3) respectively calculating the center points of all clusters obtained in the substep S302 according to the following calculation formula:
Wherein (X centre、Ycentre) represents the center point coordinates of the cluster; (x s,ys) represents the coordinates of the s-th accident point in the cluster; n is the number of accident points in each cluster;
s304: and all accident points in the preset radius range form black point areas by taking the central points as the centers respectively.
In the invention, the weight values P (1), P (2) and P (3) are all in the range of [0,1], and can be flexibly set according to actual demands, and generally, larger weight values are given to the stressed accident degree. For example, P (1) =p (2) =p (3) =1 can be made without emphasis on the degree of the accident when identifying the accident black spot; when only the accident black spot with the accident degree of death needs to be identified, P (1) =1, P (2) =p (3) =0; when the accident black points are identified according to the casualties, P (1) can be larger than P (2) and P (3) when assigning values.
Further, before executing step S200, preprocessing is performed on the traffic accident data, where the preprocessing includes removing data including null points in the traffic accident data and removing data with deviation in position information.
Further, in sub-step S302, the number K of clusters is automatically determined by using a contour coefficient method.
Furthermore, the method of the invention can also perform black point identification according to the accident type, namely, the traffic accident data of a certain preset accident type is extracted, and the steps S200-S300 are executed based on the traffic accident data and road network data of the preset accident type, so as to finally obtain the black point corresponding to the preset accident type.
Compared with the prior art, the invention has the following advantages and beneficial effects:
The invention mainly considers that the unreasonable road section or intersection design is the direct cause of the accident, thus extracting objective regularity information such as time-space distribution characteristics of the traffic accident based on the occurred traffic accident data and accurately identifying the traffic accident black spots on the road. The traffic management department adopts precautionary measures or controls related road sections in advance according to the detected accident black spots, so that traffic accidents caused by unreasonable road section or road junction design can be limited, the occurrence probability of the traffic accidents is reduced, casualties caused by the traffic accidents are reduced, and the life safety of people is ensured.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention;
FIG. 2 is raw traffic accident data in an embodiment;
Fig. 3 is an effect diagram of traffic accident data after coordinate conversion and road network data matching in the specific embodiment;
FIG. 4 is a schematic view of an accident high incidence point extracted in the embodiment;
FIG. 5 is a schematic illustration of a center point of a cluster obtained by an embodiment;
fig. 6 is a schematic view of a certain black dot area in the embodiment.
Detailed Description
The present invention will be described in further detail with reference to the following examples in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.
FIG. 1 is a schematic flow chart of the method of the present invention, and the steps for carrying out the method of the present invention will be described with reference to FIG. 1.
1. Data acquisition
The data acquired in the step comprise traffic accident data and road network data. Comprehensive and accurate traffic accident data is the basis of accident black spot identification. The collected traffic accident data comprises time and position information, and also comprises attribute information related to the accident, such as accident degree, affiliated institutions, accident handling people, whether an alarm is given, accident type, accident weather, accident environment and the like. In recent years, with rapid development of wireless communication technology and embedded technology, wireless positioning technology has been developed, and smart phones as a portable data acquisition terminal have begun to be widely used as substitutes for traditional paper data acquisition methods. The App installed by the intelligent mobile phone can rapidly complete data acquisition work of traffic accident sites, and the accuracy of accident acquisition and the efficiency of processing are improved. The current traffic accident data acquisition method by adopting the mobile terminal of the mobile phone is widely applied.
In this embodiment, the state area yin in Ningbo city is selected as the experimental area, and the state area yin is located in the south of Ningbo city and is one of the three main urban areas of Ningbo city. The road networks of yin state areas are crisscrossed and bear the main traffic flow of Ningbo city.
The data collected in this embodiment specifically includes:
(1) Traffic accident data
Based on the prior art, the traffic accident data are more acquired, and in this embodiment, the traffic accident data from 1 st of 7 th month of 2020 to 30 th 11 th month of 2020 are acquired from traffic big data laboratory of Ningbo engineering institute, and the data include, besides time and location information of traffic accidents, attribute information related to the accidents, such as accident type, accident degree, accident environment, and the like. The accident types further include accidents between non-motor vehicles, accidents between motor vehicles and non-motor vehicles, accidents between motor vehicles and pedestrians, accidents between non-motor vehicles and pedestrians, and other accidents than the foregoing. The extent of the accident further includes physical damage, injury and death.
(2) Road network data
The current channels for obtaining road network data are more, for example, road network data of the yin state area of Ningbo city can be crawled from a hundred degree map or a Goldmap, and required road network data can be purchased from a mapping or traffic management department.
2. Preprocessing and matching of traffic accident data and road network data
In this embodiment, the original traffic accident data set collected for 5 months (2020.07.01-2020.11.30) has 40535 data records. Through manual inspection, the data records containing the null points have 204 pieces in total. The null point refers to an accident point without space position information, which cannot participate in subsequent calculation, so that the data record containing the null point is manually removed. And rendering accident points corresponding to the rest traffic accident data on a geographic map with administrative division, and performing point location matching with the administrative division data to find that 26 accident points have position deviation. These 26 accident points and the corresponding data records are also deleted. And finally, carrying out preprocessing to obtain 40305 pieces of traffic accident data with complete information and accurate position information.
In this embodiment, the mobile phone App is used to enter the traffic accident data into the system, and a hundred-degree map coordinate system is adopted, so that before the next analysis is performed, the traffic accident data is converted from the hundred-degree map coordinate system into a WGS1984 coordinate system and then into a projection coordinate system. Both the hundred degree map coordinate system and the WGS1984 coordinate system are common geographic coordinate systems. Similarly, the road network data is also subjected to coordinate conversion, the road network data and the traffic accident data are converted into the same set of coordinate system, the road network data and the accident points are drawn in the same coordinate system, and the matching of the traffic accident data and the road network data is realized. The accident point is herein a traffic accident point. Fig. 2 shows original traffic accident data, fig. 3 shows traffic accident data after coordinate conversion, namely an effect diagram after matching the traffic accident data with road network data, and black dots in fig. 2 to 3 represent accident points.
3. Accident black spot identification
The specific steps of calculating the black points of the traffic accidents are as follows:
(1) And traversing each accident point, recording the currently traversed accident point as a current accident point, and counting other accident points in a preset range around the current accident point, wherein the other accident points refer to accident points except the current accident point in the preset range. In this embodiment, the preset range around the current accident point refers to a range centered on the current accident point and having a radius of 100 meters. Judging whether the number of other accident points in the preset range is larger than a preset number threshold value, and if so, considering the current accident point as an accident high-incidence point; otherwise, the current accident point is not considered to be the accident high-incidence point. And (5) finishing traversal, and obtaining a set of accident high-incidence points. The black dots shown in fig. 4 are the extracted accident high incidence points.
The preset range is manually set according to the requirement, and the radius of the general range is set to be 100-300 meters. The number threshold is used for screening accident high occurrence points, and can be determined based on a preset basic number threshold, wherein the basic number threshold refers to the number threshold of accidents in the month, and is set manually. Specifically, the quantity threshold is determined according to the time span of the traffic accident data used and a preset basic quantity threshold, for example, the time span is n months, and n is multiplied by the basic quantity threshold, namely the quantity threshold. The traffic accident data of a certain month can be directly collected, and the corresponding quantity threshold value is a preset basic quantity threshold value. In general, the base number threshold for urban areas should be set to be greater than the base number threshold for suburban areas. In this embodiment, the threshold value of the basic number of urban areas is set to 15 years/month, and the threshold value of the basic number of suburban areas is set to 10 years/month.
As a preferred scheme, the accident severity, the damage caused by the accident and the urgent degree of transformation are considered to be different, and the specific embodiment also provides another accident high occurrence point determination method considering the accident severity. In the preferred scheme, different weight values P (t) are given to the accident degrees of object damage, injury and death, the greater the weight value is, the more serious the corresponding accident degree is, and the urgent degree of arranging personnel to conduct field investigation, troubleshooting and reconstruction is indicated. In general, the degree of risk of a death accident should be given the greatest weight.
In the specific embodiment, the death accident level is given a weight value of 1.0, and the death accident level accounts for the largest proportion in all accident levels, so that the greater the urgent level of the improvement and upgrading of the related road section (or crossing) is needed; the injury accident level and the object damage accident level are respectively given with weight values of 0.5 and 0.2.
The weight P (t) is assigned the formula:
in different cities, different weight values can be set according to actual needs, and when the hazard degree weight value of the death accident is set to be higher, the extracted accident high incidence point is the death accident high incidence point. Equation (1) provides only one possible weight configuration scheme provided by this embodiment.
When traversing each accident point, for each current accident point, taking other accident points in a preset range around the current accident point, and calculating an accident heating value H in a preset range S k around the current accident point (S k):
In the formula (2), d r represents the R-th death accident within the preset range S k, R represents the total number of death accidents within the preset range S k, Representing a count of death incidents within a preset range S k; h j represents the jth injury incident within the preset range S k, J represents the total number of injuries within the preset range S k,/>Indicating a count of wound incidents within a preset range S k; m p represents the P-th loss accident within the preset range S k, P represents the total number of loss accidents within the preset range S k,/>The count of the object loss accidents in the preset range S k is indicated.
Judging whether the accident thermal value H (S k) of the current accident point in the preset range S k is larger than a preset thermal value threshold, wherein the thermal value threshold is also used for screening accident high-incidence points and is manually set, and according to actual needs, the accident high-incidence points with a large number are adjusted, and when the accident high-incidence points with a large number are required to be acquired, a smaller thermal value threshold can be set; conversely, when a small number of accident high incidence points need to be acquired, a larger threshold of thermal value can be set. If the current accident point is larger than the threshold value of the thermal value, the current accident point is considered to be an accident high-incidence point; otherwise, the accident is not a high incidence point. And (5) after traversing is completed, obtaining the accident high-incidence point set.
When P (1) =p (2) =p (3) =1, then H (S k) is equivalent to the number of other accident points within the preset range around the current accident point.
(2) Clustering the discrete accident high-occurrence points extracted in the step (1) according to the space position by using a K-means clustering method, wherein the number of clusters is automatically determined by adopting a contour coefficient method, namely the K value is automatically determined.
(3) And (3) respectively calculating the center points of the clusters obtained in the step (2).
Assuming K clusters are obtained, the set of clusters is denoted (C 1,C2,…,CK), for each cluster C i (i=1, 2, … K), for each accident point C i(s) in the cluster, the sitting sign is (x s,ys), s=1, 2, …, n, n is the number of accident points in the cluster, the n values of the different clusters may be the same or different; the coordinates (X centre、Ycentre) of the center point center of the cluster are:
the black dot in fig. 5 is the center point of the cluster obtained in this embodiment.
(4) And (3) taking each central point obtained in the step (3) as a center, and forming a corresponding black point area by all original accident points in a preset range. In this embodiment, the preset range is set as a range of 100 meters radius around the center point, and fig. 6 shows a black dot area centered on a certain center point, where the black dot is an accident point of the preset range around the center point. The outputted black dot area can facilitate the traffic management department to conduct hidden trouble investigation and transformation on related intersections or road sections.
According to the invention, black spots can be identified according to accident types (such as motor vehicle and motor vehicle accidents, motor vehicle and non-motor vehicle accidents, motor vehicle and pedestrians and the like), namely, traffic accident data of a certain preset accident type are only extracted, and based on the traffic accident data of the preset accident type, the black spot identification is carried out to obtain black spot areas corresponding to the preset accident type. Therefore, the invention has higher portability and applicability.
The foregoing embodiments are provided to illustrate the present invention by specific terms, but not to limit the scope of the invention, so that those skilled in the art can make changes and modifications to the invention with the understanding of the spirit and principles of the invention, and such equivalent changes and modifications are intended to be covered by the scope of the appended claims.
Claims (4)
1. The accident black spot identification method based on the traffic accident big data is characterized by comprising the following steps:
S100: acquiring road network data and traffic accident data of a preset time period, wherein the traffic accident data at least comprises time and position information of traffic accidents and accident degree, and the accident degree further comprises three types of material loss, injury and death;
s200: matching the positions of traffic accident data and road network data;
S300: identifying accident black spots, the steps further comprise:
s301: the method for extracting the accident high incidence point comprises the following steps:
Traversing each accident point, namely, a traffic accident occurrence point, obtaining other accident points in a preset range S k around the current accident point, and calculating an accident heating value H (S k) in a preset range S k around the current accident point:
Wherein: the weight values of the degrees of death, injury and physical damage are respectively represented by P (1), P (2) and P (3), and are all in the range of [0,1], and the weight values are artificially set; d r denotes the R-th death accident within the preset range S k, R denotes the total number of death accidents within the preset range S k, Representing a count of death incidents within a preset range S k; h j represents the jth injury incident within the preset range S k, J represents the total number of injuries within the preset range S k,/>Indicating a count of wound incidents within a preset range S k; m p represents the P-th loss accident within the preset range S k, P represents the total number of loss accidents within the preset range S k,/>A count indicating a loss of material accident within a predetermined range S k;
Judging whether the accident thermal value H (S k) of the current accident point in S k is larger than a preset thermal value threshold value, and if so, considering the current accident point as an accident high-incidence point;
completing traversal to obtain an accident high-occurrence point set;
S302: clustering the accident high-incidence points by using a K-means clustering method based on the position information of the accident high-incidence points;
S303: and (3) respectively calculating the center points of all clusters obtained in the substep S302 according to the following calculation formula:
Wherein (X centre、Ycentre) represents the center point coordinates of the cluster; (x s,ys) represents the coordinates of the s-th accident point in the cluster; n is the number of accident points in each cluster;
s304: and all accident points in the preset radius range form black point areas by taking the central points as the centers respectively.
2. The accident black spot recognition method based on the traffic accident big data according to claim 1, wherein the method is characterized in that:
before executing step S200, preprocessing is performed on traffic accident data, where the preprocessing includes removing data including null points in the traffic accident data and removing data with deviation in position information.
3. The accident black spot recognition method based on the traffic accident big data according to claim 1, wherein the method is characterized in that:
In sub-step S302, the number K of clusters is automatically determined using a contour coefficient method.
4. The accident black spot recognition method based on the traffic accident big data according to claim 1, wherein the method is characterized in that:
And (3) performing black point identification according to the accident type, namely extracting traffic accident data of a certain preset accident type, and executing steps S200-S300 based on the traffic accident data and road network data of the preset accident type to finally obtain black points corresponding to the preset accident type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110255857.8A CN112966941B (en) | 2021-03-09 | 2021-03-09 | Accident black spot identification method and system based on traffic accident big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110255857.8A CN112966941B (en) | 2021-03-09 | 2021-03-09 | Accident black spot identification method and system based on traffic accident big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112966941A CN112966941A (en) | 2021-06-15 |
| CN112966941B true CN112966941B (en) | 2024-04-19 |
Family
ID=76277001
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110255857.8A Active CN112966941B (en) | 2021-03-09 | 2021-03-09 | Accident black spot identification method and system based on traffic accident big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112966941B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113704317B (en) * | 2021-07-12 | 2023-11-14 | 武汉众智数字技术有限公司 | Accident black point prediction method based on traffic accident feature analysis |
| CN114067566B (en) * | 2021-11-18 | 2023-09-19 | 安徽达尔智能控制系统股份有限公司 | Road accident black spot screening and accident impact feature analysis method and system |
| CN115424430B (en) * | 2022-06-09 | 2024-01-23 | 长沙理工大学 | Highway traffic accident black point road section identification method and computer device |
| CN115240407B (en) * | 2022-06-10 | 2024-01-12 | 深圳市综合交通与市政工程设计研究总院有限公司 | Method and device for identifying black spots of traffic accidents, electronic equipment and storage medium |
| CN115497293B (en) * | 2022-09-21 | 2024-05-14 | 浙江大学 | Dynamic traffic accident hidden trouble point identification method |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955596A (en) * | 2014-03-14 | 2014-07-30 | 安徽科力信息产业有限责任公司 | Accident hotspot comprehensive judging method based on traffic accident collection technology |
| CN106355881A (en) * | 2016-10-12 | 2017-01-25 | 同济大学 | Space-autocorrelation-based traffic accident blackspot identification method and device |
| CN107430006A (en) * | 2014-12-02 | 2017-12-01 | 凯文·孙林·王 | Avoid the method and system of accident |
| CN107730937A (en) * | 2017-10-26 | 2018-02-23 | 东南大学 | The tunnel gateway dynamic vehicle speed abductive approach that a kind of street accidents risks minimize |
| CN107784832A (en) * | 2016-08-25 | 2018-03-09 | 上海电科智能系统股份有限公司 | A kind of method and apparatus for being used to identify the accident black-spot in traffic route |
| CN108447265A (en) * | 2018-05-21 | 2018-08-24 | 东南大学 | Road traffic accident stain section discrimination method based on TOPSIS methods |
| CN108682149A (en) * | 2018-05-21 | 2018-10-19 | 东南大学 | The linear causation analysis method in highway accident stain section based on binary logistic regression |
| CN111179141A (en) * | 2019-12-04 | 2020-05-19 | 江苏大学 | Accident-prone road section identification method based on two-stage classification |
| CN111275957A (en) * | 2018-12-05 | 2020-06-12 | 杭州海康威视系统技术有限公司 | Traffic accident information acquisition method, system and camera |
| CN111552772A (en) * | 2020-04-22 | 2020-08-18 | 中国计量大学 | Real-time traffic road condition text data and traffic volume combined visual analysis method |
-
2021
- 2021-03-09 CN CN202110255857.8A patent/CN112966941B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955596A (en) * | 2014-03-14 | 2014-07-30 | 安徽科力信息产业有限责任公司 | Accident hotspot comprehensive judging method based on traffic accident collection technology |
| CN107430006A (en) * | 2014-12-02 | 2017-12-01 | 凯文·孙林·王 | Avoid the method and system of accident |
| CN107784832A (en) * | 2016-08-25 | 2018-03-09 | 上海电科智能系统股份有限公司 | A kind of method and apparatus for being used to identify the accident black-spot in traffic route |
| CN106355881A (en) * | 2016-10-12 | 2017-01-25 | 同济大学 | Space-autocorrelation-based traffic accident blackspot identification method and device |
| CN107730937A (en) * | 2017-10-26 | 2018-02-23 | 东南大学 | The tunnel gateway dynamic vehicle speed abductive approach that a kind of street accidents risks minimize |
| CN108447265A (en) * | 2018-05-21 | 2018-08-24 | 东南大学 | Road traffic accident stain section discrimination method based on TOPSIS methods |
| CN108682149A (en) * | 2018-05-21 | 2018-10-19 | 东南大学 | The linear causation analysis method in highway accident stain section based on binary logistic regression |
| CN111275957A (en) * | 2018-12-05 | 2020-06-12 | 杭州海康威视系统技术有限公司 | Traffic accident information acquisition method, system and camera |
| CN111179141A (en) * | 2019-12-04 | 2020-05-19 | 江苏大学 | Accident-prone road section identification method based on two-stage classification |
| CN111552772A (en) * | 2020-04-22 | 2020-08-18 | 中国计量大学 | Real-time traffic road condition text data and traffic volume combined visual analysis method |
Non-Patent Citations (7)
| Title |
|---|
| Statistical Analysis and Treatment of Accident Black Spots: A Case Study of Nandyal Mandal;Reddy B S 等;《International Conference on Materials, Alloys and Experimental Mechanics (ICMAEM)》;20171119;1-17 * |
| 一种改进的K-均值算法在交通事故黑点识别中的应用;舒玥等;《黑龙江交通科技》;20180115;第41卷(第1期);194-195 * |
| 城市交通事故黑点治理新模式研究;李泽炜等;《2022世界交通运输大会(WTC2022)论文集（交通工程与航空运输篇）》;20221104;2298-2304 * |
| 基于GIS的城市道路交通事故黑点鉴别;王丽雅等;《黑龙江交通科技》(第11期);213-215 * |
| 基于改进K-means算法的城市道路交通事故分析;郭璘等;《中国公路学报》;第31卷(第04期);270-279 * |
| 基于计算机视觉的事故现场三维重建技术研究;张凯;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑 (月刊)》;20200615(第6期);C034-827 * |
| 基于驾驶员心理与生理反应的草原二级公路交通事故分析;姚娜;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑 (月刊)》;20101215(第12期);C034-302 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112966941A (en) | 2021-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112966941B (en) | Accident black spot identification method and system based on traffic accident big data | |
| CN104318324B (en) | Shuttle Bus website and route planning method based on taxi GPS records | |
| Feng et al. | Towards big data analytics and mining for UK traffic accident analysis, visualization & prediction | |
| Zhang et al. | Using street view images to identify road noise barriers with ensemble classification model and geospatial analysis | |
| CN107766983B (en) | Method for setting emergency rescue parking point of urban rail transit station | |
| Liu et al. | Dynamic estimation system for fire station service areas based on travel time data | |
| CN116168356B (en) | Vehicle damage judging method based on computer vision | |
| CN112184282A (en) | Cinema site selection model establishing method, cinema site selection method and cinema site selection platform | |
| CN117708617B (en) | Atmospheric pollution tracing method based on multi-source big data and pollution characteristic space-time matching | |
| CN109409563B (en) | Method, system and storage medium for analyzing real-time number of people in public transport operation vehicle | |
| Liu et al. | Developing an extenics-based model for evaluating bus transit system | |
| Mou et al. | Spatial influence of engineering construction on traffic accidents, a case study of Jinan | |
| CN116013084A (en) | Traffic management and control scene determining method and device, electronic equipment and storage medium | |
| CN115438938A (en) | Urban TOD mode development suitability-oriented evaluation method and system | |
| CN108831182A (en) | A kind of Urban Transit Network OD matrix construction methods | |
| Sun et al. | Identification of recurrent congestion in main trunk road based on grid and analysis on influencing factors | |
| Debebe et al. | Assessment on the Current and Future Performance of Addis Ababa Light Rail Transit Service Using Mathematical Modeling | |
| CN111833229A (en) | Travel behavior space-time analysis method and device based on subway dependency | |
| CN113343781B (en) | City functional area identification method using remote sensing data and taxi track data | |
| CN117473398B (en) | Urban dust pollution source classification method based on slag transport vehicle activity | |
| Yaofang et al. | Nonlinear impact analysis of built environment on urban road traffic safety risk | |
| Bai et al. | Which Variables are Influential to Evaluate Rapid-Transit Systems in China’s Cities? | |
| CN116681341A (en) | Urban service facility evaluation method and system based on commute path prediction | |
| CN117610929A (en) | Method and system for identifying and evaluating risks in regions with severe wind disaster along railway | |
| Yu et al. | Optimization of Bus Stop Location based on Spatial Analysis and K-Means Clustering Method–A Case Study in Hohhot City |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |