Accident prediction device and accident prediction method

Abstract

To obtain accident prediction results in a highly-accurate and almost unchanged manner when predicting road accidents.SOLUTION: An accident prediction device of an embodiment comprises the steps of: as to each of multiple sections in a road, using at least previous traffic state information and previous accident data to create accident occurrence level information associated with an accident occurrence level representing a possibility of the occurrence of accidents for each of multiple road states; as to each of the multiple sections, obtaining current traffic state information; for each prescribed time unit, as to each of the multiple sections, calculating a current accident occurrence level based on the accident occurrence level information and the current traffic state information; statistically processing multiple calculated results in a time-series manner to compute a tendency of the accident occurrence level for each period of time as to each of the multiple sections; and based on the statistic processing results, as to each of the multiple sections, creating statistic accident prediction information indicating a statistic accident occurrence level for each period of time.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] An embodiment of the present invention relates to an accident prediction device and an accident prediction method. [Background technology]

[0002] Conventionally, there has been technology that predicts accidents on roads such as expressways by using machine learning techniques that use traffic conditions (such as congestion level) and past accident data, and displays warning information (such as so-called hazard maps and near-miss maps) to traveling vehicles.

[0003] In this case, the display information regarding accident prediction can be classified into static information and dynamic information. Static information is information that does not change for a specified long period of time (e.g., one year). In the case of static information, for example, areas where accidents frequently occurred in the past can be displayed, but this information is simply a compilation of past information and has not been subjected to information analysis. In addition, the information does not take into account the current traffic conditions, etc. Therefore, there is room for improvement in terms of accuracy.

[0004] On the other hand, dynamic information is information that changes every predetermined short period (for example, every 5 minutes). Dynamic information is created by, for example, judging whether the current traffic situation is similar to the traffic situation at the time of the past accident, and therefore has higher accuracy than static information. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2009-193212 A [Patent Document 2] JP 2018-198026 A [Patent Document 3] JP 2017-126122 A [Patent Document 4] JP 2018-505422 A Summary of the Invention [Problem to be solved by the invention]

[0006] However, in the case of using dynamic information in the conventional technology, for example, if an accident prediction is made every five minutes and the accident prediction results are displayed, the accident prediction results may change frequently for the same area of ​​the road, which may confuse the drivers of traveling vehicles, and is therefore undesirable.

[0007] Therefore, an object of an embodiment of the present invention is to provide an accident prediction device and an accident prediction method that can obtain accident prediction results with high accuracy and that do not change frequently when predicting accidents on roads. [Means for solving the problem]

[0008] The accident prediction device of the embodiment includes an accident occurrence rate information creation unit that uses at least information on past traffic conditions and past accident data to create accident occurrence rate information for each of a plurality of sections on a road, in which an accident occurrence rate indicating the likelihood of an accident occurring is associated with each of a plurality of road conditions; an acquisition unit that acquires information on current traffic conditions for each of the plurality of sections; a calculation unit that calculates a current accident occurrence rate for each of the plurality of sections for each predetermined time unit based on the accident occurrence rate information and the current traffic condition information; a statistical processing unit that determines a trend in the accident occurrence rate for each of the plurality of sections by statistically processing a plurality of calculation results in a time series by the calculation unit; and a statistical accident prediction information creation unit that creates statistical accident prediction information indicating the statistical accident occurrence rate for each time period for each of the plurality of sections based on the processing results by the statistical processing unit. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a configuration diagram of a traffic control system etc. according to the first embodiment. [Diagram 2] FIG. 2 is an explanatory diagram of road sections etc. in the first embodiment. [Diagram 3]FIG. 3 is a table showing an example of the accident occurrence rate for each section in the first embodiment. [Figure 4] FIG. 4 is a flowchart showing the processing in the traffic control system of the first embodiment. [Diagram 5] FIG. 5 is a diagram showing an example of a hazard map in the first embodiment. [Figure 6] FIG. 6 is a flowchart showing the processing in the traffic control system of the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, embodiments (first to third embodiments) of an accident prediction device and an accident prediction method of the present invention will be described with reference to the drawings. In each of the following embodiments, an example will be described in which the road is an expressway.

[0011] (First embodiment) First, the first embodiment will be described. 1 is a configuration diagram of a traffic control system 10 of the first embodiment. The traffic control system 10 is configured by a plurality of computer devices arranged in a traffic control center, for example, and recognizes traffic conditions including congestion levels and traffic accidents on expressways, and performs processing such as traffic restrictions according to the traffic conditions and providing users of the expressway with caution information. In the following, descriptions of technical contents similar to those of conventional technologies will be appropriately simplified or omitted.

[0012] The traffic control system 10 uses vehicle detectors 2, surveillance cameras 3, management vehicle PCs (on-board devices of management vehicles), emergency telephones 4, mobile information terminals 5 such as mobile phones 5A and smartphones 5B, and general vehicles 6 (on-board devices of general vehicles) as information collection means.

[0013] The vehicle detector 2 is a detector (traffic counter) that is installed, for example, on the roadside of a highway and collects information such as traffic volume [vehicles / h], average speed [km / h], vehicle density [vehicles / km], and occupancy [%], and transmits the detected information to the traffic control system 10.

[0014] Fig. 2 is an explanatory diagram of sections of road R in the first embodiment. Fig. 2 shows only one direction of road R. In section A between interchanges IC1 and IC2 of road R, traffic counters TC1 to TC4 are placed on the roadside. Section A1 is between traffic counters TC1 and TC2, section A2 is between traffic counters TC2 and TC3, and section A3 is between traffic counters TC3 and TC4.

[0015] That is, the traffic conditions in section A1 can be recognized from the information from the traffic counter TC1, the traffic conditions in section A2 can be recognized from the information from the traffic counter TC2, and the traffic conditions in section A3 can be recognized from the information from the traffic counter TC3. Note that the unit for creating the statistical accident prediction information by the statistical accident prediction information creating unit 215 of the accident prediction device 20 described later may be the unit of sections A1, A2, and A3, the unit of section A, or other units.

[0016] Returning to FIG. 1, the surveillance camera 3 is, for example, a camera installed on the roadside of a highway to capture images of the highway and transmits the captured images to the traffic control system 10.

[0017] The management vehicle PC uses a detection system such as a GPS (Global Positioning System) to collect probe information such as longitude information, latitude information, altitude information, position accuracy, speed information, driving direction, and acceleration / deceleration information (acceleration in X, Y, and Z directions) as time-series vehicle information, and transmits the probe information to the traffic control system 10.

[0018] The emergency telephones 4 are installed, for example, on the main line of the expressway, in tunnels, interchanges, service areas, parking areas, bus stops, emergency parking areas, etc. In the event of a traffic accident, road users or the like use the emergency telephones 4 to notify the traffic control center that has jurisdiction over the traffic control system 10.

[0019] The general vehicle 6 transmits traffic information to the traffic control system 10 by communicating with a road-to-vehicle communication device (not shown) installed on the roadside.

[0020] The mobile information terminal 5 is carried by, for example, a road user, etc. Then, the owner of the mobile information terminal 5 uses the mobile information terminal 5 to notify the traffic control center that has jurisdiction over the traffic control system 10 of the date and time of the traffic accident, individual location information, etc.

[0021] The traffic control system 10 includes an accident prediction device 20 and an information distribution device 30. The accident prediction device 20 and the information distribution device 30 are capable of communicating with each other.

[0022] The accident prediction device 20 is a computer device, and includes a processing unit 21, a storage unit 22, an input unit 23, and a display unit 24. The storage unit 22 is a storage device such as a hard disk drive (HDD) or a solid state drive (SSD), and stores various information. The storage unit 22 stores, for example, road information 221, acquired information 222, traffic condition information 223, accident occurrence rate information 224, and statistical accident prediction information 225.

[0023] The road information 221 is information on the number of lanes on an expressway, the locations of interchanges and parking areas, and the like.

[0024] The acquired information 222 is various information acquired by the acquisition unit 212 from the vehicle detector 2, the monitoring camera 3, the management vehicle PC, the general vehicle 6, and the like.

[0025] The traffic condition information 223 is information on the traffic condition calculated by the calculation unit 213 .

[0026] The accident occurrence rate information 224 is information on the accident occurrence rate generated by the accident occurrence rate information generating unit 211 .

[0027] The statistical accident prediction information 225 is statistical accident prediction information created by the statistical accident prediction information creation unit 215 .

[0028] The processing unit 21 includes, for example, a micro processing unit (MPU), a read only memory (ROM), and a random access memory (RAM).

[0029] The MPU comprehensively controls the operation of the accident prediction device 20. The ROM is a storage medium for storing various programs and data. The RAM is a storage medium for temporarily storing various programs and rewriting various data.

[0030] The MPU executes programs stored in the ROM, storage unit 22, etc., using the RAM as a work area. The processing unit 21 performs various information processing. The processing unit 21 includes, as its functional configuration, an accident occurrence information creation unit 211, an acquisition unit 212, a calculation unit 213, a statistical processing unit 214, a statistical accident prediction information creation unit 215, and a display control unit 216.

[0031] The accident occurrence rate information creation unit 211 creates accident occurrence rate information for each of a plurality of sections on a road (e.g., sections A1, A2, and A3 in FIG. 2) by using at least information on past traffic conditions and past accident data, in which an accident occurrence rate indicating the likelihood of an accident occurring is associated with each of a plurality of road conditions.

[0032] As an example of the accident occurrence rate information, the accident occurrence rate information creating unit 211 creates, for example, a self-organizing map described in Japanese Patent No. 6045846. However, the accident occurrence rate information is not limited to the self-organizing map, and may be information created using other algorithms such as Bayesian estimation.

[0033] 1, the acquisition unit 212 acquires information on the current traffic conditions for each of the multiple sections. The acquisition unit 212 acquires various pieces of information from, for example, each information collection means (such as the vehicle detector 2 shown in the figure, as well as a precipitation detector, rain gauge, road surface thermometer, and sediment detector (not shown)) and various external information servers (such as a disaster information server and weather information server), and stores the acquired various pieces of information in the storage unit 22 as acquired information 222.

[0034] The calculation unit 213 calculates the traffic conditions including the degree of congestion of vehicles on the expressway based on the acquired information 222 and the like, and stores the calculated information in the storage unit 22 as traffic condition information 223.

[0035] Furthermore, the calculation unit 213 calculates the current accident occurrence degree for each of a plurality of sections for each predetermined time unit (e.g., 5 minutes) based on the accident occurrence degree information and traffic condition information 223 (information on current traffic conditions). The calculation unit 213 calculates the accident occurrence degree by a method of learning accident cases for each condition such as position on a road, time, and weather using a self-organizing map (accident occurrence degree information), as described in Japanese Patent No. 6045846. The calculation unit 213 stores the calculated accident occurrence degree in the storage unit 22 as accident occurrence degree information 224.

[0036] 3 is a table showing an example of the accident occurrence rate for each section in the first embodiment. As shown in Fig. 3, the calculation unit 213 calculates the accident occurrence rate for each 5 minutes for each of the sections A1, A2, and A3 in Fig. 2.

[0037] Returning to Fig. 1, the statistical processing unit 214 statistically processes the calculation results of the multiple accident occurrence rates in a time series by the calculation unit 213, thereby determining the tendency of the accident occurrence rate for each time period for each of the multiple sections. For example, as shown in Fig. 3, the statistical processing unit 214 calculates the average value of six accident occurrence rates every five minutes for each section. In addition, the statistical processing unit 214 can also calculate the accident occurrence rate for section A, for example, by calculating the average value of the average values ​​of the accident occurrence rates for each of sections A1, A2, and A3.

[0038] Based on the processing results by the statistical processing unit 214, the statistical accident prediction information creation unit 215 creates statistical accident prediction information (such as a hazard map) that indicates the statistical occurrence rate of accidents for each time period (e.g., every 30 minutes or every hour) for each of multiple sections.

[0039] The display control unit 216 controls the display of various information on the display unit 24 .

[0040] The input unit 23 is an input device that accepts operations by a traffic controller with respect to the processing unit 21 of the traffic control system 10, and is, for example, a keyboard, a mouse, or the like.

[0041] The display unit 24 is realized by a liquid crystal display (LCD (Liquid Crystal Display)), an organic EL (Electro-Luminescence) display device, or the like, and displays various information.

[0042] The information distribution device 30 is a computer device, and includes a processing unit 31, a storage unit 32, an input unit 33, a display unit 34, and a communication unit 35.

[0043] The storage unit 32 is a storage device such as an HDD or SSD, and stores various information. The storage unit 32 stores, for example, distribution information 321. The distribution information 321 is statistical accident prediction information (such as a hazard map) that is instructed to be distributed by the accident prediction device 20.

[0044] The processing unit 21 includes, for example, an MPU, a ROM, and a RAM. The MPU comprehensively controls the operation of the information distribution device 30. The ROM is a storage medium for storing various programs and data. The RAM is a storage medium for temporarily storing various programs and rewriting various data.

[0045] The MPU executes programs stored in the ROM, storage unit 32, etc., using the RAM as a work area. The processing unit 31 performs various information processes. The processing unit 31 includes an information distribution unit 311 and a display control unit 312 as functional units.

[0046] The information distribution unit 311 distributes distribution information 321 to a distribution destination. In the example of Fig. 1, the distribution destinations include an information board DT1, a highway radio DT2, an information terminal DT3, a mobile information terminal DT4, and an in-vehicle device DT5.

[0047] The information boards DT1 are information display devices installed at various points on the expressway. The highway radios DT2 are radio devices that receive radio waves transmitted from transmitting antennas installed at various points on the expressway and convert them into sound.

[0048] The information terminal DT3 is a terminal device for providing information installed in service areas and parking areas on expressways. The mobile information terminal DT4 is a smartphone, a mobile phone, or the like carried by a pre-registered road user. The vehicle-mounted device DT5 is a car navigation device or the like mounted on a pre-registered general vehicle 6.

[0049] The display control unit 312 controls the display of various information on the display unit .

[0050] The input unit 33 is an input device that accepts operations by a traffic controller on the information distribution device 30, and is, for example, a keyboard, a mouse, or the like.

[0051] The display unit 34 is realized by a liquid crystal display device (LCD), an organic EL display device, or the like, and displays various information.

[0052] The communication unit 35 has a communication interface (not shown) and distributes various information (text data, voice data, image data, etc.) via communication paths corresponding to the information board DT1, the highway radio DT2, the information terminal DT3, the mobile information terminal DT4, and the vehicle-mounted device DT5, which are the information distribution targets. The communication unit 35 executes the above information distribution under the control of the information distribution unit 311.

[0053] FIG. 4 is a flowchart showing the processing in the traffic control system 10 of the first embodiment. In step S1, the acquisition unit 212 of the accident prediction device 20 acquires information on the current traffic situation from the vehicle sensors 2 and the like for each of a plurality of sections of the road.

[0054] Next, in step S2, the calculation unit 213 calculates the current accident occurrence degree for each of the plurality of sections based on accident occurrence degree information (for example, a self-organizing map) and information on the current traffic situation.

[0055] Next, in step S3, the statistical processing unit 214 statistically processes the calculation results of the accident occurrence degrees in a time series to obtain the tendency of the accident occurrence degree for each time zone for each of the plurality of sections.

[0056] Next, in step S4, the statistical accident prediction information creation unit 215 creates statistical accident prediction information (such as a hazard map) indicating the statistical accident occurrence degree for each time zone for each of the plurality of sections based on the processing result in step S3.

[0057] Next, in step S5, the information distribution unit 311 of the information distribution device 30 that has received a distribution instruction such as statistical accident prediction information from the accident prediction device 20 distributes the statistical accident prediction information and the like to the information board DT1, the highway radio DT2, the information terminal DT3, the mobile information terminal DT4, the in-vehicle device DT5, and the like.

[0058] FIG. 5 is a diagram showing an example of a hazard map in the first embodiment. This hazard map is an example of the statistical accident prediction information created by the statistical accident prediction information creation unit 215. In the hazard map of FIG. 5, for a section where the statistical accident occurrence degree exceeds a predetermined threshold value in units of sections between interchanges IC on the highway, any one of accident occurrence degree levels L1, L2, L3 (L1 < L2 < L3) is represented.

[0059] In this case, the statistical accident prediction information creation unit 215, for example, sets the section where the accident occurrence rate is equal to or greater than the first threshold and less than the second threshold as accident occurrence rate level L1, sets the section where the accident occurrence rate is equal to or greater than the second threshold and less than the third threshold as accident occurrence rate level L2, and sets the section where the accident occurrence rate is equal to or greater than the third threshold and less than the fourth threshold as accident occurrence rate level L3.

[0060] In this way, according to the traffic control system 10 of the first embodiment, when predicting accidents on roads, the above-mentioned statistical accident prediction information (hazard map, etc.) is created, so that highly accurate and stable accident prediction results that do not change frequently can be obtained. The traffic control system 10 can alert traveling vehicles and the like to accidents by transmitting such statistical accident prediction information to the information board DT1, highway radio DT2, information terminal DT3, mobile information terminal DT4, in-vehicle device DT5, etc.

[0061] For example, a driver of a traveling vehicle can know the accident occurrence probability for each section by looking at a hazard map such as that shown in Figure 5 on an information board DT1 or the like, and can drive with appropriate attention depending on the accident occurrence probability for the current traveling section.

[0062] The accident prediction device 20 may update such statistical accident prediction information (hazard map, etc.) at intervals of, for example, 30 minutes or 1 hour. In this way, the accuracy of the statistical accident prediction information can be maintained high, and the possibility of confusing the drivers of traveling vehicles can be reduced because the update frequency is not high.

[0063] Second embodiment Next, a second embodiment will be described. Descriptions of matters similar to those of the first embodiment will be omitted as appropriate. The second embodiment differs from the first embodiment mainly in the processing of the calculation unit 213 of the accident prediction device 20.

[0064] The calculation unit 213 calculates the current accident occurrence rate for each of a plurality of sections for each predetermined time unit (e.g., 5 minutes) based on the accident occurrence rate information and information on the current traffic conditions, and further extracts the maximum accident occurrence rate from the most recent predetermined number of accident occurrence rates (e.g., 3 accidents).

[0065] For example, in the example of Figure 3, if the calculation unit 213 calculates the accident occurrence rates for the most recent three accident occurrences in section A1 as "3," "5," and "4," it extracts the maximum of these, "5."

[0066] Then, the statistical processing unit 214 statistically processes a plurality of extraction results in chronological order by the calculation unit 213, thereby determining the tendency of the accident occurrence rate for each time period for each of the plurality of sections.

[0067] However, the above-mentioned multiple times is not limited to three times, but may be two times or four or more times.

[0068] 6 is a flowchart showing the process in the traffic control system 10 of the second embodiment. Steps S1 and S2 are the same as those in FIG.

[0069] After step S2, in step S21, the calculation unit 213 extracts the maximum accident occurrence degree from among the accident occurrence degrees of a predetermined number of most recent occurrences (for example, three occurrences).

[0070] Next, in step S3, the statistical processing unit 214 obtains a trend of the accident occurrence rate for each time period for each of the multiple sections by statistically processing the extraction results of the multiple accident occurrence rates in time series by the calculation unit 213. Steps S4 and S5 are the same as those in FIG.

[0071] Thus, according to the traffic control system 10 of the second embodiment, in addition to the same effects as in the first embodiment, it is possible to create more accurate statistical accident prediction information (hazard maps, etc.) that takes into account not only the current traffic conditions but also the traffic conditions from a short time ago, such as 5 minutes ago, 10 minutes ago, etc.

[0072] Third embodiment Next, a third embodiment will be described. Descriptions of matters similar to those of the first embodiment will be omitted as appropriate. In the third embodiment, the process of the statistical accident prediction information creation unit 215 of the accident prediction device 20 is mainly different from that of the first embodiment.

[0073] When creating statistical accident prediction information indicating the statistical accident occurrence rate for each time period for each of a plurality of sections based on the processing results by the statistical processing unit 214, the statistical accident prediction information creation unit 215 creates at least one of statistical accident prediction information for each time period within a day, statistical accident prediction information for each day of the week, statistical accident prediction information for each weather condition, and statistical accident prediction information for each traffic condition.

[0074] In other words, the statistical accident prediction information creation unit 215 does not create a single statistical accident prediction information (hazard map, etc.) for each section, but creates separate statistical accident prediction information for each time of day (daytime / nighttime), each day of the week, each weather condition (weather (sunny, rainy, foggy, snow, etc.), temperature, humidity, etc.), and each traffic condition (normal, crowded, congested, etc.).

[0075] Thus, according to the traffic control system 10 of the third embodiment, in addition to the same effects as in the first embodiment, it is possible to create more accurate statistical accident prediction information for each section, time, weather conditions, and traffic conditions.

[0076] Although the embodiment of the present invention has been described above, the above embodiment is merely an example and is not intended to limit the scope of the invention. The above embodiment can be implemented in various forms, and various omissions, substitutions, and modifications can be made without departing from the gist of the invention. The above embodiment is included in the scope and gist of the invention, and is included in the scope of the invention and its equivalents described in the claims.

[0077] For example, the target roads are not limited to expressways, but may be toll roads other than expressways, general roads, and the like.

[0078] Furthermore, the time unit for calculating the accident occurrence rate is not limited to 5 minutes, but may be another time unit such as 1 minute. [Explanation of symbols]

[0079] 2...vehicle detector, 3...surveillance camera, 4...emergency phone, 5...mobile information terminal, 6...general vehicle, 10...traffic control system, 20...accident prediction device, 21...processing unit, 22...storage unit, 23...input unit, 24...display unit, 30...information distribution device, 31...processing unit, 32...storage unit, 33...input unit, 34...display unit, 35...communication unit, 211...accident occurrence information creation unit, 212...acquisition unit, 213...calculation unit, 214...statistical processing unit, 215...statistical accident prediction information creation unit, 216...display control unit, 221...road information, 222...acquired information, 223...traffic condition information, 224...accident occurrence information, 225...statistical accident prediction information, 311...information distribution unit, 312...display control unit, 321...distribution information

Claims

1. an accident occurrence information creation unit that creates accident occurrence information for each of a plurality of sections of a road by using at least information on past traffic conditions and past accident data, in which an accident occurrence degree that indicates the likelihood of an accident occurring is associated with each of a plurality of road conditions; An acquisition unit that acquires information on current traffic conditions for each of the plurality of sections; a calculation unit that calculates a current accident occurrence degree for each of the plurality of sections for each predetermined time unit based on the accident occurrence degree information and the information on the current traffic situation, and extracts the calculated accident occurrence degree as an accident occurrence degree; a statistical processing unit that statistically processes the plurality of accident occurrence rates in the time series extracted by the calculation unit in the time direction to obtain a tendency of the accident occurrence rate for each of the plurality of sections for each time period having a length multiple times the time unit; a statistical accident prediction information creation unit that creates statistical accident prediction information indicating a statistical accident occurrence rate for each time period for each of the plurality of sections based on a processing result by the statistical processing unit; An accident prediction device comprising:

2. the calculation unit calculates a current accident occurrence degree for each of the plurality of sections for each of the predetermined time units based on the accident occurrence degree information and the information on the current traffic conditions, and extracts a maximum accident occurrence degree from among the accident occurrence degrees for a predetermined number of most recent accident occurrences. The accident prediction device according to claim 1.

3. The accident prediction device described in claim 1, wherein when creating statistical accident prediction information indicating the statistical accident occurrence rate for each time period for each of the multiple sections based on the processing results by the statistical processing unit, the statistical accident prediction information creation unit creates at least one of statistical accident prediction information for each time period within a day, statistical accident prediction information for each day of the week, statistical accident prediction information for each weather condition, and statistical accident prediction information for each traffic condition.

4. an accident occurrence information creation step in which an accident occurrence information creation unit creates accident occurrence information in which an accident occurrence degree indicating the likelihood of an accident occurring is associated with each of a plurality of road conditions, for each of a plurality of sections on the road, by using at least information on past traffic conditions and past accident data; an acquisition step in which an acquisition unit acquires information on a current traffic situation for each of the plurality of sections; an extraction step in which a calculation unit calculates a current accident occurrence degree for each of the plurality of sections for each predetermined time unit based on the accident occurrence degree information and the information on the current traffic situation, and extracts the current accident occurrence degree as an accident occurrence degree; a statistical processing step in which a statistical processing unit statistically processes the plurality of accident occurrence rates in the time series extracted by the extraction step in the time direction to obtain a tendency of the accident occurrence rate for each of the plurality of sections for each time period having a length multiple times the time unit; a statistical accident prediction information creating step in which a statistical accident prediction information creating unit creates statistical accident prediction information indicating a statistical accident occurrence rate for each time period for each of the plurality of sections based on a processing result of the statistical processing step; An accident prediction method comprising:

Images