Road accident risk black spot generation method, equipment, device and electronic equipment

Abstract

The invention provides a road accident risk black spot generation method, equipment, a device and electronic equipment, and the method comprises the steps: obtaining a plurality of pieces of historical accident data, and determining time-space parameters in each piece of historical accident data, the time-space parameters comprising accident coordinates; taking each piece of historical accident data as a reference point, performing clustering with other historical accident data according to a set aggregation distance, and determining a successfully clustered core reference point according to a set minimum aggregation number; based on the clustering result of each core reference point, determining a clustering result belonging to the accident black spots; and determining the area range of the accident black spots based on the accident coordinates in the historical accident data in the accident black spots. According to the method, detailed distribution and statistics of accident risks are finally obtained through multi-dimension statistical analysis of accident black spot generation, region splicing, high-incidence time periods and the like, and prediction and basis are provided for subsequent command decision and post deployment.

Description

Technical Field

[0001] The present application relates to the field of intelligent traffic control command, and in particular to a method, equipment, device and electronic equipment for generating road accident risk black spots. Background Art

[0002] In recent years, with the rapid development of the national economy and the continuous improvement of residents' living standards, people's travel has become more convenient, and the number of motor vehicles has increased rapidly. At the same time, the road load and the frequency of accidents have also been increasing, bringing chaos and challenges to the management of traffic order.

[0003] The shortage of grassroots police force has become a major factor that restricts and troubles traffic management work. How to scientifically allocate police resources, achieve precise intervention, alleviate the shortage of police force, improve police work efficiency, and provide the command center with closed-loop command scenario analysis and prediction of better control measures is a pain point that urgently needs to be solved. Summary of the invention

[0004] The embodiments of the present application provide a method, device, apparatus and electronic device for generating road accident risk black spots, which are used to alleviate the problem of insufficient allocation of police resources and achieve precise intervention in traffic management chaos.

[0005] In a first aspect, the present application provides a method for generating a road accident risk black spot, comprising:

[0006] Acquire a plurality of historical accident data, and determine the spatiotemporal parameters in each of the historical accident data, wherein the spatiotemporal parameters include accident coordinates;

[0007] Taking each historical accident data as the reference point, cluster it with the rest of the historical accident data according to the set aggregation distance, and determine the core reference point for successful clustering according to the set minimum aggregation number;

[0008] Based on the clustering results of each core benchmark point, determine the clustering results belonging to the accident black spot;

[0009] Based on the accident coordinates in the historical accident data of each accident black spot, the regional scope of the accident black spot is determined.

[0010] In some possible embodiments, based on the clustering results of each core reference point, determining the clustering results belonging to the accident black spot includes:

[0011] If any clustering result contains only one core reference point, the clustering result is determined to be an accident black spot;

[0012] If any clustering result includes at least two core reference points, it is determined that at least two clustering results corresponding to the at least two core reference points belong to the same accident black spot.

[0013] In some possible embodiments, determining the area range of the accident black spot based on the accident coordinates in the historical accident data of each accident black spot includes:

[0014] Determining peripheral accident coordinates based on accident coordinates in historical accident data in each accident black spot;

[0015] According to the connection of the peripheral accident coordinates, the regional range of the accident black spot is determined, and the area of ​​the accident black spot is calculated.

[0016] In some possible embodiments, the spatiotemporal parameters in the historical accident data also include the accident time. The method of taking each historical accident data as a reference point, clustering with the remaining historical accident data according to a set aggregation distance, and determining the core reference point for successful clustering according to a set minimum aggregation number includes:

[0017] For at least one set time parameter, each historical accident data whose accident time is within the time parameter is taken as a reference point, and clustered with the remaining historical accident data according to the set aggregation distance. According to the set minimum aggregation number, the core reference point for successful clustering corresponding to the time parameter is determined.

[0018] In some possible embodiments, based on the clustering results of each core reference point, determining the clustering results belonging to the accident black spot includes:

[0019] If any clustering result includes at least two core reference points, and if the at least two core reference points belong to the same time parameter, it is determined that the at least two clustering results belong to the same accident black spot.

[0020] In some possible embodiments, after determining the area range of the accident black spot, the method further includes:

[0021] For two accident black spots with overlapping time parameters, determine the area of ​​the overlapping portion of the area ranges corresponding to the two accident black spots;

[0022] Calculate the corresponding degree of overlap according to the ratio of the area of ​​the overlapped portion to the area of ​​the two accident black spots;

[0023] If any overlap degree is greater than a set value, the two accident black spots are merged into one accident black spot, and the time parameter of the merged accident black spot is updated.

[0024] In some possible embodiments, the obtaining of historical accident data includes:

[0025] Obtain the original historical accident data and filter out the historical accident data containing the accident coordinates;

[0026] The original historical accident data that does not contain the accident coordinates is used to generate the accident coordinates through the original historical accident data related position information to obtain the corresponding historical accident data.

[0027] In some possible embodiments, the time parameter includes each set time period corresponding to each hour on a working day and each set time period corresponding to each hour on a non-working day, and the method further includes:

[0028] Count the number of historical accident data contained in each accident black spot in each set time period;

[0029] The accident level of each accident black spot in each set time period is determined based on the relationship between the amount of historical accident data of each accident black spot and a set value, wherein the set value is the average amount of historical accident data per working day / non-working hour of each black spot.

[0030] In a second aspect, the present application provides a road accident risk black spot generation device, the device comprising:

[0031] An acquisition module, which acquires a plurality of historical accident data and determines the spatiotemporal parameters in each of the historical accident data, wherein the spatiotemporal parameters include accident coordinates;

[0032] The clustering module takes each historical accident data as the reference point, clusters it with the rest of the historical accident data according to the set aggregation distance, and determines the core reference point for successful clustering according to the set minimum aggregation number;

[0033] The first determination module determines the clustering results belonging to the accident black spots based on the clustering results of each core reference point;

[0034] The second determination module determines the area range of the accident black spots based on the accident coordinates in the historical accident data of each accident black spot.

[0035] In a third aspect, the present application provides an electronic device, the electronic device comprising:

[0036] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the road accident risk black spot generation method as described in any one of the first aspects.

[0037] In a fourth aspect, the present application provides a computer storage medium storing a computer program, wherein the computer program is used to enable a computer to execute a road accident risk black spot generation method as described in any one of the first aspects.

[0038] According to a road accident risk black spot generation method, equipment, device and electronic equipment provided by the present application, based on the accident data occurring in the area within the historical time period, the historical accident data that meets the conditions are screened out, and the accident data in the accident-intensive area are aggregated according to the set points to generate accident black spots, which can obtain detailed distribution and statistics of accident risks, and provide predictions and basis for subsequent command decisions and job deployment. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification are used to explain the principles of the present application, and do not constitute an improper limitation on the present application.

[0040] Figure 1 A step diagram of a method for generating a road accident risk black spot provided in an embodiment of the present application;

[0041] Figure 2 A flow chart of a method for generating a road accident risk black spot provided in an embodiment of the present application;

[0042] Figure 3 A flow chart of a historical accident data screening method provided in an embodiment of the present application;

[0043] Figure 4 A flow chart of the steps for aggregating accident black spots provided in an embodiment of the present application;

[0044] Figure 5 A flow chart of a method for calculating the area of ​​an accident black spot region provided in an embodiment of the present application;

[0045] Figure 6 An accident coordinate display diagram provided in an embodiment of the present application;

[0046] FIG. 7( a ) is a distribution diagram of an accident black spot map provided in an embodiment of the present application;

[0047] FIG. 7( b ) is a distribution diagram of an accident black spot map provided in an embodiment of the present application;

[0048] Figure 8 The accident black spot map distribution diagram provided in the embodiment of the present application;

[0049] Fig. 9 The accident black spot map distribution diagram provided in the embodiment of the present application;

[0050] Fig.10 An accident black spot splicing flow chart provided for an embodiment of the present application;

[0051] Fig.11 A schematic diagram of accident black spot distribution provided in an embodiment of the present application;

[0052] Fig.12 A schematic diagram of accident black spot distribution provided in an embodiment of the present application;

[0053] Fig.13 A schematic diagram of overlapping accident black spots provided in an embodiment of the present application;

[0054] Fig.14 A schematic diagram of the accident black spots after splicing provided in the embodiment of the present application;

[0055] Fig.15 A diagram of a road accident risk black spot generation device provided in an embodiment of the present application;

[0056] Fig.16 A road accident risk black spot generation equipment diagram provided in an embodiment of the present application;

[0057] Fig.17 A storage medium map for generating road accident risk black spots provided in an embodiment of the present application. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0059] In recent years, with the rapid development of the national economy and the continuous improvement of residents' living standards, people's travel has become more convenient, and the number of motor vehicles has increased rapidly. At the same time, the road load and the frequency of accidents have also been increasing, bringing chaos and challenges to the management of traffic order.

[0060] The shortage of grassroots police force has become a major factor that restricts and troubles traffic management work. How to scientifically allocate police resources, achieve precise intervention, alleviate the shortage of police force, improve police work efficiency, and provide the command center with closed-loop command scenario analysis and prediction of better control measures is a pain point that urgently needs to be solved.

[0061] In view of the above problems, the present application invents a method for generating road accident black spots, that is, based on a certain time and space, according to the accident data that occurred in the area within a historical period, data processing is performed, the accident data in the accident-intensive area is point-aggregated, accident black spots are generated, regional splicing is performed, and statistical analysis is performed on multiple dimensions such as high-incidence time periods, and finally the detailed distribution and statistics of accident risks are obtained, providing predictions and basis for subsequent command decisions and job deployment.

[0062] The basic concept of the present invention is as follows Figure 1As shown in the figure, the generation and data processing and analysis process of accident black spots mainly includes:

[0063] In the data preparation stage, the acquisition process of historical accidents mainly includes: obtaining original historical accident data, extracting historical accident data that meets the time and space parameter conditions, cleaning the historical accident data and completing the geographic coordinates;

[0064] In the geographic aggregation stage, based on the cleaned and completed historical accident data, the geographic coordinates are grouped into regions to form accident black spots. Secondly, the contour lines of the accident black spot regions are taken to calculate the area of ​​the black spots.

[0065] In the data splicing stage, the geographical locations of accident black spots with overlapping time parameter ranges are identified and their regional matching degree is calculated. If they are confirmed to be the same black spot, the accident black spot regional splicing is completed, and the calculated effective time of the black spot is the splicing of the overlapping time parameter range;

[0066] Analysis of high-incidence time periods: Based on the previous step, historical accident data within accident black spots is calculated on a weekly basis, and statistical analysis is performed during key time periods (such as the morning rush hour).

[0067] The road accident black spot generation method provided in the embodiment of the present application is as follows: Figure 2 As shown, including:

[0068] Step 201, obtaining a plurality of historical accident data, and determining the spatiotemporal parameters in each historical accident data, wherein the spatiotemporal parameters include accident coordinates;

[0069] The historical accident data is obtained from the database. For the aggregation process of accident black spots, the accident coordinate information contained in the historical accident data is necessary data.

[0070] In some possible embodiments, the acquisition of historical accident data, such as Figure 3 As shown, including:

[0071] Step 301, obtaining original historical accident data, and filtering out historical accident data containing accident coordinates;

[0072] Step 302 : Generate accident coordinates from original historical accident data that does not contain accident coordinates using relevant position information of the original historical accident data to obtain corresponding historical accident data.

[0073] First, the accident data within a historical time range (such as half a year) is obtained from the database as the original historical accident data, and invalid data and data with missing important information are screened out. If the original historical accident data contains the coordinates of the accident, the data is directly used as usable and valid historical accident data; if the original historical accident data does not have a specific coordinate location, but the data records the geographical description of the accident, such as the accident location, accident section and other language and text description information, then according to the natural language segmentation method, the description is mapped into the corresponding POI (point of interest) coordinates or intersection coordinates through geographical segmentation as the accident coordinates of the historical accident data.

[0074] Step 202, taking each historical accident data as a reference point, clustering with the rest of the historical accident data according to a set aggregation distance, and determining the core reference point for successful clustering according to a set minimum aggregation number;

[0075] In order to group the historical accident data obtained in step 201 with similar coordinates, realize the point aggregation in geographic space, and preliminarily construct the aggregation point (accident black spot), it is necessary to first screen out the core reference points in the historical accident data. The meaning of the core reference point is that the core reference point and the points whose distance from it is not greater than the aggregation distance can be used to aggregate the accident black spots; the standard for judging whether a historical accident data is a core reference point is that the number of all historical accident data whose distance from the historical accident data is not greater than the aggregation distance (including the historical accident data itself) is not less than the minimum aggregation number, then the historical accident data is a core reference point. Among them, the aggregation distance and the minimum aggregation number are manually set values, which are set according to the scene characteristics selected by the historical accident data, the actual needs, or the needs of traffic management to achieve the best aggregation effect.

[0076] Step 203, based on the clustering results of each core reference point, determine the clustering results belonging to the accident black spot;

[0077] In step 202, for the clustering results of the preliminary aggregation performed with each core reference point as the center, the clustering results with overlapping ranges that meet the conditions are merged to determine the final accident black spot.

[0078] Step 204: determining the area range of the accident black spots based on the accident coordinates in the historical accident data of each accident black spot.

[0079] For each generated accident black spot, the outer coordinates of the accident black spot are obtained, and the connecting line of the outer coordinates corresponds to the area range corresponding to each accident black spot.

[0080] In a possible embodiment, based on the clustering results of each core reference point, determining the clustering results belonging to the accident black spot includes:

[0081] If any clustering result contains only one core reference point, the clustering result is determined to be an accident black spot;

[0082] If any clustering result includes at least two core reference points, it is determined that at least two clustering results corresponding to the at least two core reference points belong to the same accident black spot.

[0083] When determining accident black spots based on the clustering results of each core benchmark point, there are two possible situations. When the clustering result contains only one core benchmark point (the core benchmark point at this time is the core benchmark point that generates the clustering result), the clustering result is determined as an accident black spot; when any clustering result contains at least two core benchmark points, the clustering results corresponding to the above core benchmark points are merged.

[0084] Specifically, Figure 4 As shown in the figure, the possible operation process of generating accident black spots is explained:

[0085] Step 401: for a historical accident data point p that has not been confirmed as a core reference point, check its neighborhood. The neighborhood of p is all historical accident data within (including E) the aggregation distance from it. Find all historical accident data within (including E) the aggregation distance from it. If the number of historical accident data included is not less than the minimum aggregation number minP, establish a clustering result C with the core reference point p, and add all points in it to the candidate set N.

[0086] Step 402: for all the unprocessed historical accident data q in the candidate set N, check its neighborhood. The neighborhood of q is all the historical accident data within the aggregation distance E (including E). If it contains at least minP historical accident data, then add these historical accident data to N; if q is not included in any clustering result, then add q to the clustering result C;

[0087] Step 403, repeat steps 401 and 402, and continue to check whether the historical accident data in N is confirmed to be the core reference point, until the current candidate set N is empty; finally, until all historical accident data are classified into a certain clustering result or marked as noise.

[0088] In a possible embodiment, based on the accident coordinates in the historical accident data of each accident black spot, the area range of the accident black spot is determined, such as Figure 5 As shown, including:

[0089] Step 501, determining peripheral accident coordinates based on the accident coordinates in the historical accident data of each accident black spot;

[0090] Step 502, determining the region range of the accident black spot based on the connection of the peripheral accident coordinates, and calculating the area of ​​the accident black spot.

[0091] When the accident black spots are aggregated, each accident black spot is composed of a series of historical accident coordinates. In order to display the accident black spots more clearly on the map, it is necessary to determine the regional scope of each accident black spot based on the outer accident coordinates of the accident black spots, and calculate the area of ​​the accident black spots, so as to facilitate the subsequent data splicing process of the accident black spots.

[0092] In a possible embodiment, the spatiotemporal parameters in the historical accident data also include the accident time. The method of taking each historical accident data as a reference point, clustering with the remaining historical accident data according to a set aggregation distance, and determining the core reference point for successful clustering according to a set minimum aggregation number includes:

[0093] For at least one set time parameter, each historical accident data whose accident time is within the time parameter is taken as a reference point, and clustered with the remaining historical accident data according to the set aggregation distance. According to the set minimum aggregation number, the core reference point for successful clustering corresponding to the time parameter is determined.

[0094] When obtaining historical accident data, the time parameter is also one of the necessary space-time parameters. Each time the accident black spot aggregation process is performed, all historical accident data within a certain time range are always selected for aggregation. When the time range is different, the results of black spot aggregation are also different.

[0095] In a possible embodiment, based on the clustering results of each core reference point, determining the clustering results belonging to the accident black spot includes:

[0096] If any clustering result includes at least two core reference points, and if the at least two core reference points belong to the same time parameter, it is determined that the at least two clustering results belong to the same accident black spot.

[0097] In the process of black spot aggregation, when the clustering results are merged, the clustering results that can be merged must belong to the same time parameter. Belonging to the same time parameter means that the historical accident data contained in each clustering result belongs to the accident data that occurred within the same time range.

[0098] Specifically, Figure 6 Taking the historical accident data in as an example, the aggregation process of accident black spots is explained in detail:

[0099] It should be noted that each historical accident data is a coordinate point, referred to as a point. The neighborhood of a point refers to the set of points that take the point as the reference point and do not exceed the aggregation distance E (including E). The clustering results C1, C2 and C3 are accident black spots.

[0100] 1. First scan the reference point P1(1,2) and calculate its E neighborhood, that is, find the points in the reference point whose distance from it is not greater than E. The E neighborhood of the reference point P1 is: {P1, P2, P3, P13};

[0101] 2. Set the minimum number of aggregations MinP = 3, the density of P1 is 4>MinP, so P1 is the core reference point, and the clustering result C1 is established based on P1. Let N be the set of points in the neighborhood of P1;

[0102] 3. For each point q in N, if q has not been processed and is a core reference point, merge the points in the neighborhood of q into N; if q is not a member of any clustering result, add q to C1;

[0103] 4. The neighborhood of P2 is {P1, P2, P3, P13}. P2 is a core reference point and has not been processed. The points in its neighborhood are merged into N, and P2 is added to C1.

[0104] 5. The neighborhood of P3 is {P1, P2, P3, P4, P13}. As above, merge the points in the neighborhood of P3 into N and add P3 to C1.

[0105] 6. For P13, the neighborhood is {P1, P2, P3, P4, P13}. As above, merge the points in the neighborhood of P13 into N and add P13 to C1.

[0106] 7. Continue to consider the unprocessed point P4 in N, whose neighborhood is {P3, P4, P13}, merge the points in the neighborhood of P4 into N, and add P4 to C1; at this time, all objects in the set N have been processed, and the clustering result C1 is obtained, which contains the points {P1, P2, P3, P4, P13};

[0107] 8. Continue scanning the reference point P5 (5, 8), and obtain the clustering result C2 as above, which contains {P5, P6, P7, P8}.

[0108] 9. For the reference point P9 (9, 5), its E neighborhood is calculated as {P9}, then P9 is a non-core reference point;

[0109] 10. Scan the reference point P10 (1, 12). The E neighborhood of the P10 reference point is {P10, P11}, so P10 is a non-core reference point.

[0110] 11. Continue to scan the reference point P11 (3, 12). The neighborhood of the P11 reference point E is {P10, P11, P12}. P11 is the core reference point. Use P11 to establish a new clustering result C3. Let N be the set of points in the neighborhood of P11.

[0111] 12. For each point q in N, if q has not been processed and is a core reference point, merge the points in the neighborhood of q into N; if q is not a member of any clustering result, add q to C3;

[0112] 13. For the non-core reference point P10, add P10 to C3. For the non-core reference point P12, the neighborhood is {P11, P12}, add P12 to C3. At this time, the objects in N are processed and the clustering result C3 is obtained.

[0113] 14. Continue scanning the reference points. P12 and P13 have been processed and the algorithm ends.

[0114] The final clustering results are: 3 clustering results C1{P1, P2, P3, P4, P13}, C2{P5, P6, P7, P8}, C3{P10, P11, P12}. The reference point P9 was not classified into any one and was finally judged to be noise.

[0115] It should be noted that for all points in the historical accident data, the inspection order will not have any effect on the final aggregation results and the generated black points.

[0116] After completing the aggregation of black spot geographic coordinates to form accident black spots, the peripheral point sets are calculated for C1, C2, and C3 respectively to obtain the aggregation area.

[0117] For example, for the accident black spot C1, the aggregation area S C1 for:

[0118] S c1 =S{lon p1 ,lat p1 ,lon p2 ,lat p2 ,lon p3 ,lat p3 ,lon p4 ,lat p4 ,lon p13 ,lat p13 ,}(Formula 1)

[0119] Among them, lon p1 Indicates the longitude geographic coordinates of P1, lat p1 Indicates the latitude geographic coordinates of P1, and the same applies to other coordinate points.

[0120] By aggregating regions, the regional area can be calculated and displayed on the map, which facilitates subsequent data splicing.

[0121] Specifically, the following takes a city as an example, and analyzes and displays the aggregated accident black spots according to different set aggregation distances and minimum aggregation quantities for historical accident data within the same time parameter range:

[0122] 1. Data preparation:

[0123] The original historical accident data was obtained, with a total of 6150; the data with a time range of 2021-12-09 to 2021-05-09 was screened, and 5307 data remained; the data without specific accident coordinates was deleted, and 4424 data remained; for the data with geographic location description information, after natural language coordinates were supplemented, there were 4702 data in total; after removing obviously abnormal points, a total of 4118 valid historical accident data were screened out.

[0124] 2. Clustering effect:

[0125] As shown in Figure 7(a), the map in Figure 7(a) shows the historical accident data in a certain area and a certain time range in step 1 in three dimensions: heat, accident black spots, and noise. The heat map is generated based on a single historical accident data to represent the density of accidents. The shaded part in Figure 7(a) is the heat map, and the lighter the shade, the denser the accidents. It can be seen that the number of accident black spots in the figure is consistent with the trend of the heat map, thus verifying the effectiveness of the generated accident black spots.

[0126] As shown in Figure 7(b), the aggregation distance is set to 200 meters, the minimum aggregation number is MinP=10, and the number of accident black spots after aggregation is 67. After calculation, the number of historical accident data contained in the accident black spots accounts for the total number of historical accident data (4118) The ratio is: 0.4.

[0127] like Figure 8 As shown, the aggregation distance is set to 100 meters, the minimum aggregation number is MinP=10, and the number of accident black spots after aggregation is 32. After calculation, the number of historical accident data contained in the accident black spots accounts for the total number of historical accident data (4118) The ratio is: 0.31.

[0128] like Fig. 9 As shown, the aggregation distance is set to 50 meters, the minimum aggregation number is MinP=5, and the number of accident black spots after aggregation is 66. After calculation, the number of historical accident data contained in the accident black spots accounts for the total number of historical accident data (4118) The ratio is: 0.33.

[0129] The aggregation distance and the minimum aggregation number jointly control the clustering results, that is, the size and number of accident black spots generated. Generally speaking, if the aggregation distance is too large, the area of ​​the class in the clustering result will be too large, that is, the black spot area will be too large. If the minimum aggregation number is too large, accidents must occur very densely within the range to form black spots. If the minimum aggregation number is too small, the number of accidents formed will be too large. Therefore, when setting the aggregation distance and the minimum aggregation number, it is necessary to make reasonable settings based on the specific accident environment and needs, so as to achieve the best prediction of historical accident data.

[0130] In some possible embodiments, after determining the area range of the accident black spot, such as Fig.10 As shown, it also includes:

[0131] Step 1001, for two accident black spots with overlapping time parameters, determine the area of ​​the overlapping portion of the region ranges corresponding to the two accident black spots;

[0132] Accident black spots only use a fixed time period as the statistical cycle, which can only reflect the accident situation in the area during that time period. It is impossible to effectively conduct historical trend analysis over a longer period of time, that is, the changes in accidents. Therefore, adding a historical data comparison method, that is, splicing accident black spots with overlapping time ranges, can effectively conduct long-term black spot statistics.

[0133] Step 1002, calculating the corresponding overlap degree according to the ratio of the area of ​​the overlapped portion to the area of ​​the two accident black spots;

[0134] For the black dots that overlap in two coordinate regions with different time parameter ranges, the area of ​​the overlapping portion is calculated. Since the areas of the two overlapping black dots are generally different in size, it is necessary to calculate the area overlap of each black dot separately.

[0135] Step 1003: If any overlap degree is greater than a set value, the two accident black spots are merged into one accident black spot, and the time parameter of the merged accident black spot is updated.

[0136] If the degree of overlap of two black spots with overlapping areas is greater than a set value, usually 50%, the two black spots are confirmed to be the same accident black spot, and the area range of the black spots is merged, and the time range of the black spots is also merged.

[0137] Specifically, the splicing process of accident black spots is described in detail:

[0138] (1) Fig.11 As shown, the data for 6 months from 2023.1.1 to 2023.6.30 are counted and aggregated to form black dots, and a total of 3 black dots are obtained; Fig.10As shown, the black spots from 2023.2.1 to 2023.7.31 are counted in sequence, and 3 black spots are obtained. Fig.11 and Fig.12 ,It is found that some black spots have similar geographical coordinates locations, while some black spots are completely different;

[0139] (2) In order to analyze the specific correlation of black spots, black spots [A1, A2, A3] and black spots [B1, B2, B3] are placed in a unified longitude and latitude coordinate system, as follows: Fig.13 As shown, calculate the area of ​​each black point [S-a1, S-a2, S-a3], [S-b1, S-b2, S-b3] respectively;

[0140] (3) Take the black points in group A as the reference black points, loop through the black points in group A, calculate the geographic coordinate area overlap between each black point in group A and the black points in the control group B, and sort them [in1, in2, in3]. Take the area in1 with the largest overlap area to calculate the area S-in1, and compare its area with the original control group black points S-b1 and S-in1 with S-a1 and S-b1 respectively;

[0141]

[0142] Where σ is the area ratio threshold. If they are all greater than this threshold (such as 50%), the overlapping area is greater than 50%, that is, S-a1 and S-b1 are considered to be the same group of black spots with different time spans. The historical time can be merged, and the change in area reflects the change in time and trend. For example, if the overlapping areas of the black spots in the subsequent time range (2023.3.1-2023.8.30, 2023.4.1-2023.9.31...) are all greater than 50% of the overlapping areas of the previous black spots, then a black spot set [S-a1, S-b1, S-c1,..., S-n1] is formed. The start time of the merged black spot is the start time of the S-a1 black spot, and the end time is the end time of S-n1 (temporarily weakening the influence of other time periods).

[0143] 4) If the area and proportion of S-a2 and S-b2 to S-in2 are less than σ, they are considered to be two completely different accident black spots, that is, there is no time correlation. The start and end times of S-a2 and S-b2 are their respective start and end times.

[0144] 5) The final data splicing result is as follows Fig.14As shown, the number of accidents of S-a3 decreased during 2023.2.1-7.31, and the aggregation threshold was not reached, so the point aggregation could not be completed, so the black point disappeared, and the existence period of S-a3 was 1.1-6.30; S-b3 did not reach the aggregation threshold between 1.1-6.30, and the point aggregation could not be completed, and no black point was generated, but a large number of accidents were added between 2.1-7.31, which met the aggregation conditions and generated new black points. The period of the black point existence was 2.1-7.31.

[0145] By comparing historical data, unrelated black spots can be effectively temporally correlated, providing a time dimension guarantee for trend analysis of black spots.

[0146] In some possible embodiments, the time parameter includes each set time period corresponding to each hour on a working day and each set time period corresponding to each hour on a non-working day, and the method further includes:

[0147] Count the number of historical accident data contained in each accident black spot in each set time period;

[0148] The accident level of each accident black spot in each set time period is determined based on the relationship between the amount of historical accident data of each accident black spot and a set value, wherein the set value is the average amount of historical accident data per working day / non-working hour of each black spot.

[0149] When analyzing the high-incidence periods of historical accident data, based on the comparison of historical data, the overall data of accident black spots including historical trends are divided into frequent occurrence periods at a weekly granularity. The division steps include:

[0150] 1) First divide the overall data into working days and non-working days from Monday to Friday and Saturday to Sunday.

[0151] 2) Divide the data into 24 groups according to each hour (0-1 o'clock, 1-2 o'clock, etc.), and count the number of accidents in a single black spot in each time period, and calculate

[0152] w = total / 24

[0153] Among them, total is the total number of black spot accidents (working days, non-working days), and w is the average number of accidents (per hour). If total1h>w, it is a general level, total1h>2w, it is a medium level, and total1h>4w, it is a serious level and needs special attention. It usually occurs during peak hours in the morning and evening (7:00-9:00, 17:00-19:00). Warnings can also be issued for other abnormal time periods. If the distribution of a single black spot time period is not obvious, other accident influencing factors should be considered. Through further detailed division of time periods, detailed analysis and summary of black spot formation can be carried out for subsequent trend prediction and system warning.

[0154] Through the above steps, the present invention can screen out road accident risk black spots and perform relevant prediction and data display work. In addition, the present invention is adaptable to a variety of data types. For example, it can also be used for similar business data such as violations and passing vehicles, and focuses on statistical analysis of data, providing the command center with closed-loop command scenario analysis and prediction of control measures.

[0155] Based on the same inventive concept, the present application also provides a road accident risk black spot generating device, such as Fig.15 As shown, including:

[0156] An acquisition module 1501 acquires a plurality of historical accident data and determines a spatiotemporal parameter in each historical accident data, wherein the spatiotemporal parameter includes an accident coordinate;

[0157] Clustering module 1502, taking each historical accident data as a reference point, clustering with other historical accident data according to a set aggregation distance, and determining a core reference point for successful clustering according to a set minimum aggregation number;

[0158] The first determination module 1503 determines the clustering results belonging to the accident black spots based on the clustering results of each core reference point;

[0159] The second determination module 1504 determines the area range of the accident black spots based on the accident coordinates in the historical accident data of each accident black spot.

[0160] In some possible embodiments, the first determining module 1503 determines the clustering results belonging to the accident black spots based on the clustering results of each core reference point, including:

[0161] If any clustering result contains only one core reference point, the clustering result is determined to be an accident black spot;

[0162] If any clustering result includes at least two core reference points, it is determined that at least two clustering results corresponding to the at least two core reference points belong to the same accident black spot.

[0163] In some possible embodiments, the second determining module 1504 determines the area range of the accident black spot based on the accident coordinates in the historical accident data of each accident black spot, including:

[0164] Determining peripheral accident coordinates based on accident coordinates in historical accident data in each accident black spot;

[0165] According to the connection of the peripheral accident coordinates, the regional range of the accident black spot is determined, and the area of ​​the accident black spot is calculated.

[0166] In some possible embodiments, the spatiotemporal parameters in the historical accident data also include the accident time. The clustering module 1502 uses each historical accident data as a reference point, clusters it with the remaining historical accident data according to a set aggregation distance, and determines the core reference point for successful clustering according to a set minimum aggregation number, including:

[0167] For at least one set time parameter, each historical accident data whose accident time is within the time parameter is taken as a reference point, and clustered with the remaining historical accident data according to the set aggregation distance. According to the set minimum aggregation number, the core reference point for successful clustering corresponding to the time parameter is determined.

[0168] In some possible embodiments, the first determining module 1503 determines the clustering results belonging to the accident black spots based on the clustering results of each core reference point, including:

[0169] If any clustering result includes at least two core reference points, and if the at least two core reference points belong to the same time parameter, it is determined that the at least two clustering results belong to the same accident black spot.

[0170] In some possible embodiments, after determining the area range of the accident black spot, the second determination module 1504 further executes:

[0171] For two accident black spots with overlapping time parameters, determine the area of ​​the overlapping portion of the area ranges corresponding to the two accident black spots;

[0172] Calculate the corresponding degree of overlap according to the ratio of the area of ​​the overlapped portion to the area of ​​the two accident black spots;

[0173] If any overlap degree is greater than a set value, the two accident black spots are merged into one accident black spot, and the time parameter of the merged accident black spot is updated.

[0174] In some possible embodiments, the acquisition module 1501 acquires historical accident data, including:

[0175] Obtain the original historical accident data and filter out the historical accident data containing the accident coordinates;

[0176] The original historical accident data that does not contain the accident coordinates is used to generate the accident coordinates through the original historical accident data related position information to obtain the corresponding historical accident data.

[0177] In some possible embodiments, the time parameters include each set time period corresponding to each hour on a working day and each set time period corresponding to each hour on a non-working day, and the first determining module 1503 further executes:

[0178] Count the number of historical accident data contained in each accident black spot in each set time period;

[0179] The accident level of each accident black spot in each set time period is determined based on the relationship between the amount of historical accident data of each accident black spot and a set value, wherein the set value is the average amount of historical accident data per working day / non-working hour of each black spot.

[0180] Based on the same inventive concept, the present application also provides an electronic device, comprising at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a road accident risk black spot generation method in the above-mentioned embodiment.

[0181] like Fig.16 As shown, the device includes a processor 1601, a memory 1602, a communication interface 1603 and a bus 1604. The processor 1601, the memory 1602 and the communication interface 1603 are connected to each other via the bus 1604.

[0182] The processor 1601 is used to read and execute instructions in the memory 1602, so that at least one processor can execute the road accident risk black spot generation method provided in the above embodiment.

[0183] The memory 1602 is used to store various instructions and programs of the road accident risk black spot generation method provided in the above embodiment.

[0184] The bus 1604 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.16 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0185] Processor 1601 may be a central processing unit (CPU), a network processor (NP), a graphic processing unit (GPU), or any combination of CPU, NP, and GPU. It may also be a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0186] Based on the same inventive concept, the present application also provides a readable storage medium, such as Fig.17 As shown, the computer storage medium stores a computer program, and the computer program is used to enable a computer to execute any one of the methods in the above embodiments.

[0187] The memory may include readable media in the form of volatile memory, such as random access memory (RAM) 171 and / or cache memory 1722 , and may further include read-only memory (ROM) 1723 .

[0188] The memory may also include a program / utility 1725 having a set (at least one) of program modules 1724, such program modules 1724 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0189] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0190] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0191] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0192] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0193] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for generating road accident risk black spots, characterized in that: The method comprises: Acquire a plurality of historical accident data, and determine the spatiotemporal parameters in each of the historical accident data, wherein the spatiotemporal parameters include accident coordinates; Taking each historical accident data as the reference point, cluster it with the rest of the historical accident data according to the set aggregation distance, and determine the core reference point for successful clustering according to the set minimum aggregation number; Based on the clustering results of each core benchmark point, determine the clustering results belonging to the accident black spot; Based on the accident coordinates in the historical accident data of each accident black spot, the regional scope of the accident black spot is determined.

2. The method according to claim 1, characterized in that: Based on the clustering results of each core benchmark point, the clustering results belonging to the accident black spot are determined, including: If any clustering result contains only one core reference point, the clustering result is determined to be an accident black spot; If any clustering result includes at least two core reference points, it is determined that at least two clustering results corresponding to the at least two core reference points belong to the same accident black spot.

3. The method according to claim 1 or 2, characterized in that: Based on the accident coordinates in the historical accident data of each accident black spot, the regional scope of the accident black spot is determined, including: Determining peripheral accident coordinates based on accident coordinates in historical accident data in each accident black spot; According to the connection of the peripheral accident coordinates, the regional range of the accident black spot is determined, and the area of ​​the accident black spot is calculated.

4. The method according to claim 1 or 2, characterized in that: The spatiotemporal parameters in the historical accident data also include the accident time. The historical accident data are used as reference points, clustered with the remaining historical accident data according to the set aggregation distance, and the core reference points for successful clustering are determined according to the set minimum aggregation number, including: For at least one set time parameter, each historical accident data whose accident time is within the time parameter is taken as a reference point, and clustered with the remaining historical accident data according to the set aggregation distance. According to the set minimum aggregation number, the core reference point for successful clustering corresponding to the time parameter is determined.

5. The method according to claim 1, characterized in that: Based on the clustering results of each core benchmark point, the clustering results belonging to the accident black spot are determined, including: If any clustering result includes at least two core reference points, and if the at least two core reference points belong to the same time parameter, it is determined that the at least two clustering results belong to the same accident black spot.

6. The method according to claim 4, characterized in that After determining the area of ​​the accident black spot, it also includes: For two accident black spots with overlapping time parameters, determine the area of ​​the overlapping portion of the area ranges corresponding to the two accident black spots; Calculate the corresponding degree of overlap according to the ratio of the area of ​​the overlapped portion to the area of ​​the two accident black spots; If any overlap degree is greater than a set value, the two accident black spots are merged into one accident black spot, and the time parameter of the merged accident black spot is updated.

7. The method according to claim 1, characterized in that The obtaining of historical accident data includes: Obtain the original historical accident data and filter out the historical accident data containing the accident coordinates; The original historical accident data that does not contain the accident coordinates is used to generate the accident coordinates through the original historical accident data related position information to obtain the corresponding historical accident data.

8. The method according to claim 4, characterized in that The time parameters include each set time period corresponding to each hour on a working day and each set time period corresponding to each hour on a non-working day, and the method further includes: Count the number of historical accident data contained in each accident black spot in each set time period; The accident level of each accident black spot in each set time period is determined based on the relationship between the amount of historical accident data of each accident black spot and a set value, wherein the set value is the average amount of historical accident data per working day / non-working hour of each black spot.

9. A road accident risk black spot generation device, characterized in that: The device comprises: An acquisition module, which acquires a plurality of historical accident data and determines the spatiotemporal parameters in each of the historical accident data, wherein the spatiotemporal parameters include accident coordinates; The clustering module takes each historical accident data as the reference point, clusters it with the rest of the historical accident data according to the set aggregation distance, and determines the core reference point for successful clustering according to the set minimum aggregation number; The first determination module determines the clustering results belonging to the accident black spots based on the clustering results of each core reference point; The second determination module determines the area range of the accident black spots based on the accident coordinates in the historical accident data of each accident black spot.

10. An electronic device, characterized in that: The electronic device comprises: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the road accident risk black spot generation method as described in any one of claims 1-8.

Images