Information processing device

Abstract

To clearly indicate whether a driver is likely to cause an accident.SOLUTION: An information processing apparatus includes a control unit which collects record data including videos from vehicles, for image analysis. The control unit uses a ratio of the total number of accidents with respect to the total number of accident predictor motions, based on results of the image analysis, regarding all drivers as a population, with respect to the number of accident predictor motions of each driver, to determine the possibility of an accident to be caused by each driver. When samples are insufficient, a fixed value is used for determination instead of the ratio.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The disclosed embodiments relate to an information processing device. Place Regarding. [Background technology]

[0002] Conventionally, a technology has been known in which vehicle data is collected from multiple vehicles, the dangerous state of each vehicle is determined in real time based on the analysis results of the vehicle data relating to past accidents or accident precursory behavior, and operational control is performed on vehicles determined to be dangerous (see, for example, Patent Document 1).

[0003] The above-mentioned accident precursors refer to so-called "near misses." Near misses can also be described as events that did not result in an accident, but have the potential to cause one.

[0004] Vehicle data collected from multiple vehicles can also be used to perform other tasks, such as evaluating the driving of each driver. Some fleet management servers for commercial vehicles constantly collect vehicle data from each vehicle under management and calculate an evaluation score for each driver based on the driving tendencies of each driver indicated by the collected vehicle data. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2020-166416 Summary of the Invention [Problem to be solved by the invention]

[0006] However, the above-mentioned prior art has room for further improvement in terms of enabling a clear indication of whether a driver is likely to cause an accident.

[0007] For example, with conventional technology, operations managers and drivers can understand each driver's driving tendencies and an evaluation score based on those tendencies, but they cannot specifically understand whether a driver is likely to cause an accident.

[0008] One aspect of the embodiment has been made in consideration of the above, and aims to provide an information processing device, an information processing system, and an information processing method that enable a clear indication of whether a driver is likely to cause an accident. [Means for solving the problem]

[0009] According to an embodiment, an information processing device includes a control unit that, based on image data collected from a plurality of vehicles, records a number of near misses, a number of accidents, and a ratio of the number of accidents to the number of near misses, for a population of drivers of the plurality of vehicles; If the number of the population is equal to or greater than a predetermined value, If the value obtained by multiplying the number of near misses of the target driver by the ratio exceeds a threshold, it is determined that there is a possibility that the target driver may have caused an accident. If the number of the population is less than the predetermined value, and if the value obtained by multiplying the number of near misses of the target driver by a fixed value instead of the ratio exceeds a threshold, it is determined that there is a possibility that an accident has occurred due to the target driver. . [Effects of the Invention]

[0010] According to one aspect of the embodiment, it is possible to clearly indicate whether a driver is likely to cause an accident. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram illustrating an outline of an information processing method according to an embodiment. [Figure 2] FIG. 2 is a block diagram showing an example of the configuration of the in-vehicle device according to the embodiment. [Figure 3] FIG. 3 is a block diagram illustrating an example of the configuration of a server device according to the embodiment. [Figure 4] FIG. 4 is an explanatory diagram of a method for calculating the number of possible accidents according to the embodiment. [Figure 5]FIG. 5 is a diagram showing an example of a screen display on a terminal device. [Figure 6] FIG. 6 is an explanatory diagram of the ratio according to the modified example. [Figure 7] FIG. 7 is an explanatory diagram of a case where a coefficient is assigned. [Figure 8] FIG. 8 is a flowchart showing a processing procedure executed by the server device according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, embodiments of an information processing device, an information processing system, and an information processing method disclosed in the present application will be described in detail with reference to the accompanying drawings. Note that the present invention is not limited to the following embodiments.

[0013] In the following, "possibility of accident occurrence" refers to the possibility that a driver will cause an accident. Therefore, "possibility of accident occurrence" refers to the possibility that a driver who is the subject of the judgment will cause an accident. Furthermore, "number of possible accident occurrences" refers to the number of cases in which a driver may cause an accident.

[0014] First, an overview of an information processing method according to an embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram illustrating an overview of an information processing method according to an embodiment.

[0015] As shown in FIG. 1, the information processing system 1 according to the embodiment includes: in-vehicle devices 10-1, 10-2, 10-3 . . . mounted on a plurality of vehicles V-1, V-2, V-3 .

[0016] In the following description, when it is not necessary to distinguish between the vehicles V-1, V-2, V-3, etc., they will be referred to as "vehicle V." Similarly, when it is not necessary to distinguish between the in-vehicle devices 10-1, 10-2, 10-3, etc., they will be referred to as "in-vehicle device 10."

[0017] The in-vehicle device 10 has a camera, and is a device that captures images of the inside and outside of the vehicle V with the camera and records the images as record data.

[0018] Furthermore, various sensors such as an acceleration sensor and a GPS (Global Positioning System) sensor are connected to the in-vehicle device 10 via an in-vehicle network such as a CAN (Controller Area Network). When the in-vehicle device 10 detects the occurrence of a predetermined event such as an accident based on various vehicle data acquired from these sensors, the in-vehicle device 10 protects the recorded data at the time of the accident from being overwritten.

[0019] The in-vehicle device 10 is configured to be able to upload recorded data to the server device 100 via a network N. That is, the in-vehicle device 10 is, for example, a communication-type drive recorder. The network N is the Internet, a C-V2X (Cellular Vehicle to Everything) communication network, or the like.

[0020] The server device 100 is a device that collects the recorded data recorded by the in-vehicle device 10. The server device 100 is configured, for example, as a cloud server that provides cloud services via the network N. If the vehicle V is a commercial vehicle, the server device 100 is, for example, a fleet management server operated and managed by a business operator.

[0021] The server device 100 is also configured to be able to perform image analysis of the recorded data of each vehicle V based on the recorded data collected from the in-vehicle device 10. The server device 100 is also configured to be able to determine whether or not the scene is a near miss or an accident scene based on the result of the image analysis.

[0022] In addition, the server device 100 is configured to be able to perform statistical processing of all drivers or each driver as a population based on the determined number of near misses and the number of accidents, calculation processing of the number of accidents likely to occur for each driver, and generation processing of evaluation information for each driver based on the calculation results.

[0023] The terminal device 200 is a computer used by the operation manager and each driver. The terminal device 200 is realized by, for example, a PC (Personal Computer) such as the terminal device 200-1, or a smartphone such as the terminal device 200-2.

[0024] The terminal device 200 is provided so as to be able to communicate with the server device 100 via the network N, and is provided so as to be able to view the evaluation information of each driver generated by the server device 100.

[0025] In the information processing system 1 configured as above, in the information processing method according to the embodiment, first, the in-vehicle device 10 of each vehicle V records record data and uploads the recorded record data to the server device 100 (step S1).

[0026] Meanwhile, the server device 100 collects recorded data uploaded from each vehicle V (step S2). Then, the server device 100 performs image analysis on the collected recorded data and determines whether it is a near miss or an accident (step S3).

[0027] Then, the server device 100 executes statistical processing based on the determination result of step S3, and calculates the ratio r of the total number of accident occurrences to the total number of near misses in the population of all drivers based on the results of the statistical processing.

[0028] However, if the number of samples is insufficient, the accuracy of the ratio r may not be guaranteed. Therefore, if the number of samples is small in step S4, the server device 100 sets a fixed value instead of the ratio r. The "number of samples" here refers to at least one of the number of samples of collected record data, the number of samples of near-miss scenes, and the number of samples of scenes where an accident occurs.

[0029] The fixed value can be a known statistical value such as a statistical value from statistical data published by the police or a statistical value known as "Heinrich's Law." A specific example of using a statistical value based on Heinrich's Law will be described later using Figure 4.

[0030] Then, the server device 100 calculates the number of accident occurrence possibilities for each driver by using the ratio r for the number of near misses for each driver based on the ratio r calculated in step S4 (step S5).

[0031] If the calculated number of accident occurrence possibilities exceeds a predetermined threshold, the server device 100 determines that the driver in question has a risk of causing an accident (step S6). Then, the server device 100 generates evaluation information for each driver including the determined accident possibility, and provides the information to the terminal device 200 (step S7). Note that the server device 100 may provide the evaluation information for each driver not only to the terminal device 200, but also to the in-vehicle device 10. That is, the in-vehicle device 10 also functions as the terminal device 200.

[0032] As described above, in the information processing method according to the embodiment, the server device 100 collects recorded data including video from each vehicle V, performs image analysis, and determines whether or not each driver is likely to have an accident by using the ratio r of the total number of accidents to the total number of near misses for the population of all drivers based on the results of the image analysis for the number of near misses for each driver. If the number of samples is insufficient, a fixed value is set for the ratio r.

[0033] Therefore, according to the information processing method of the embodiment, it is possible to clearly indicate whether or not a driver is likely to cause an accident. Hereinafter, a configuration example of the information processing system 1 to which the information processing method according to the embodiment described above is applied will be described in more detail.

[0034] Fig. 2 is a block diagram showing an example of the configuration of the in-vehicle device 10 according to the embodiment. Note that Fig. 2 and Fig. 3 shown later show only components necessary for explaining the features of the embodiment, and general components are omitted.

[0035] 2 and 3 are functional concepts and do not necessarily have to be physically configured as shown. For example, the specific form of distribution and integration of each block is not limited to that shown, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0036] In addition, in the description using FIGS. 2 and 3, the description of components that have already been described may be simplified or omitted.

[0037] 2, the in-vehicle device 10 according to the embodiment includes a communication unit 11, a storage unit 12, and a control unit 13. The in-vehicle device 10 is also connected to a sensor unit 3 via an in-vehicle network (not shown). The sensor unit 3 is a group of various sensors mounted on the vehicle V.

[0038] The sensor unit 3 includes a camera 3a, an acceleration sensor 3b, and a GPS (Global Positioning System) sensor 3c. The camera 3a captures images of the outside of the vehicle V. The acceleration sensor 3b measures the acceleration and speed of the vehicle V. The GPS sensor 3c measures the GPS position of the vehicle V.

[0039] At least some of the sensors included in the sensor unit 3 may be provided integrally with the in-vehicle device 10.

[0040] The communication unit 11 is realized by a network adapter etc. The communication unit 11 is wirelessly connected to the network N described above, and transmits and receives information to and from the server device 100.

[0041] The storage unit 12 is realized by a storage device such as a RAM (Random Access Memory) or a flash memory, or a disk device such as a hard disk device or an optical disk device. In the example of Fig. 2, the storage unit 12 stores record information 12a.

[0042] The recorded information 12a is information including a group of recorded data photographed by the camera 3a and recorded by the recording unit 13b, which will be described later.

[0043] The control unit 13 is a controller, and is realized by a CPU (Central Processing Unit), MPU (Micro Processing Unit), GPU (Graphics Processing Unit), etc., executing various programs (not shown) stored in the storage unit 12 using RAM as a work area. The control unit 13 can also be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array).

[0044] The control unit 13 has an acquisition unit 13a, a recording unit 13b, a detection unit 13c, a record protection unit 13d, and an upload unit 13e, and realizes or executes the functions and actions of information processing described below.

[0045] The acquiring unit 13a acquires sensor data output from the sensor unit 3. The acquiring unit 13a acquires video captured by the camera 3a as one of the sensor data. The acquiring unit 13a also acquires information transmitted from the server device 100 via the communication unit 11.

[0046] The recording unit 13b records the video captured by the camera 3a and the sensor data from the sensor unit 3 synchronized with the video as recorded data in the recording information 12a. The detection unit 13c detects the occurrence of an event such as an accident that requires protection of the recorded data based on the sensor data output from the sensor unit 3.

[0047] When the detection unit 13c detects the occurrence of the above-mentioned event, the record protection unit 13d protects the record data corresponding to the occurrence of the event.

[0048] The upload unit 13e uploads the data recorded by the recording unit 13b to the server device 100 via the communication unit 11 as appropriate.

[0049] Next, a configuration example of the server device 100 will be described with reference to Fig. 3. Fig. 3 is a block diagram showing a configuration example of the server device 100 according to the embodiment.

[0050] As shown in FIG. 3, the server device 100 according to the embodiment includes a communication unit 101, a storage unit 102, and a control unit 103.

[0051] The communication unit 101 is realized by a network adapter etc. The communication unit 101 is connected to the above-mentioned network N by wire or wirelessly, and transmits and receives information between the in-vehicle device 10 and the terminal device 200.

[0052] The storage unit 102 is realized by a storage device such as a RAM or a flash memory, or a disk device such as a hard disk device or an optical disk device. In the example of Fig. 3, the storage unit 102 stores a collected information DB (Database) 102a, a statistical information DB 102b, fixed value information 102c, and a driver-specific evaluation information DB 102d.

[0053] The collected information DB 102a is a database that stores recorded data collected from each in-vehicle device 10. The statistical information DB 102b is a database that stores the results of statistical processing executed by the statistical unit 103c, which will be described later.

[0054] The fixed value information 102c is information storing a fixed value of the ratio r used when the number of samples is small. The driver-specific evaluation information DB 102d is a database storing driver-specific evaluation information generated by a generating unit 103f (described later).

[0055] The control unit 103 is a controller, and is realized by a CPU, MPU, GPU, etc., executing various programs (not shown) stored in the storage unit 102 using RAM as a work area. The control unit 103 can also be realized by an integrated circuit such as an ASIC or FPGA.

[0056] The control unit 103 has a collection unit 103a, an analysis unit 103b, a statistics unit 103c, a calculation unit 103d, a determination unit 103e, a generation unit 103f, and a provision unit 103g, and realizes or executes the information processing functions and actions described below.

[0057] The collection unit 103a collects recorded data from each in-vehicle device 10 via the communication unit 101 and stores the data in the collected information DB 102a. The analysis unit 103b performs image analysis on the video images included in the recorded data stored in the collected information DB 102a. Based on the results of the image analysis, the analysis unit 103b determines whether each scene is a near miss scene or an accident scene.

[0058] The statistical unit 103c executes statistical processing based on the analysis results by the analysis unit 103b. The statistical unit 103c executes statistical processing to calculate at least the total number of near misses, the total number of accidents, and the number of near misses for each driver, with all drivers as the population.

[0059] The calculation unit 103d calculates the ratio r of the total number of accident occurrences to the total number of near misses for a population of all drivers based on the results of statistical processing by the statistics unit 103c. In addition, the calculation unit 103d calculates the number of accident occurrences possible for each driver by using the ratio r for the number of near misses for each driver.

[0060] Furthermore, when the number of samples of recorded data that constitute the population is small, the calculation unit 103d sets a fixed value to the ratio r based on the fixed value information 102c, and calculates the number of accident occurrence possibilities for each driver.

[0061] Here, for ease of understanding, the method for calculating the number of possible accident occurrences will be explained with reference to Fig. 4. Fig. 4 is an explanatory diagram of the method for calculating the number of possible accident occurrences according to the embodiment.

[0062] The calculation unit 103d calculates the ratio r of the total number of accidents to the total number of near misses based on statistical information for the population of all drivers, as shown in the left diagram of Fig. 4. In the example of the left diagram of Fig. 4, the ratio r is 0.07 (= 210 / 3000).

[0063] Then, the calculation unit 103d calculates the number of accident occurrence possibilities for each driver by using the ratio r to the number of near misses for each driver, as shown in the center diagram of Fig. 4. In the example of the center diagram of Fig. 4, the number of accident occurrence possibilities indicated by "?" is calculated to be 2.1 (=30×0.07).

[0064] When the number of samples of the recorded data that constitute the population is small, the calculation unit 103d uses a fixed value for the ratio r. As shown in the right diagram of Fig. 4, when applying the value of Heinrich's law to such a fixed value, the ratio r is set to 0.1 (= (1 + 29) / 300).

[0065] Returning to the explanation of Fig. 3, the determination unit 103e determines whether the number of accident occurrence possibilities for each driver calculated by the calculation unit 103d exceeds a predetermined threshold. If the number exceeds the threshold, the determination unit 103e determines that the driver in question has a possibility of causing an accident. If the number does not exceed the threshold, the determination unit 103e determines that the driver in question has no possibility of causing an accident.

[0066] The generating unit 103f generates evaluation information for each driver including the accident occurrence probability determined by the determining unit 103e, and stores the information in the driver-specific evaluation information DB 102d.

[0067] When the providing unit 103g receives a request from the terminal device 200 to view the evaluation information of an arbitrary driver, the providing unit 103g provides the terminal device 200 with the evaluation information of the driver.

[0068] Next, an example of a screen display of the driver evaluation information provided by the providing unit 103g on the terminal device 200 will be described with reference to Fig. 5. Fig. 5 is a diagram showing an example of a screen display on the terminal device 200.

[0069] As shown in Figure 5, the evaluation information for each driver is displayed on the terminal device 200 along with the driver's attributes such as the driver's name, age, gender, affiliation, etc., as well as a radar chart of the driving situation shown by sensor data synchronized with the video, score values, ranking values, etc.

[0070] Then, on the display screen, information about the possibility of the driver having an accident is also displayed, as shown in part M1 in the drawing. As shown in part M1, the information about the possibility of the driver having an accident includes the number of near misses, which is the cumulative number of near misses the driver has encountered up to now, whether or not there is a possibility of an accident having occurred, as determined by the determination unit 103e, and the number of possible accidents (2.1 in this case) calculated using the ratio r.

[0071] This makes it possible to clearly indicate whether the driver is likely to cause an accident, and enables users such as operation managers and drivers who view such information through the terminal device 200 to clearly understand information related to the driver's accident possibility. Note that the screen display example shown in Fig. 5 is merely an example, and does not limit the way in which information related to the driver's accident possibility shown in at least part M1 is presented.

[0072] Furthermore, up until now, the ratio r has been defined as the ratio of the total number of accidents to the total number of near misses in the population of all drivers, but this ratio r may also be calculated for each attribute of the population, each attribute of the driver, each type of accident, or each size of accident.

[0073] FIG. 6 is an explanatory diagram of the ratio r according to a modified example. As shown in FIG. 6, the ratio r may be calculated for each attribute of the population. The attribute of the population is, for example, industry. It may also be business type within the same industry. For example, trucks, buses, motorcycles (including bicycles), etc., all in the same transportation industry. That is, as shown in the same figure, the ratio r may be calculated as a ratio r1 for each truck, a ratio r2 for each bus, a ratio r3 for each motorcycle, and so on.

[0074] Similarly, the ratio r may be calculated for each driver attribute. As shown in the figure, the ratio r may be calculated as a ratio r4 by gender, a ratio r5 by age, a ratio r6 by years of driving experience, and so on.

[0075] Similarly, the ratio r may be calculated for each type of accident. As shown in the figure, the ratio r may be calculated as a ratio r7 for vehicle-related accidents, a ratio r8 for personal injury accidents, a ratio r9 for property damage accidents, and so on.

[0076] Similarly, the ratio r may be calculated for each accident severity. As shown in the figure, the ratio r is calculated based on the ratio r for serious accidents. 10 , the ratio r in minor accidents 11 , ... may be calculated as follows.

[0077] In this way, when the ratio r is calculated for each attribute of the population, each attribute of the driver, each type of accident, and each size of accident, the calculation unit 103d will calculate the number of accidents that may occur for each driver for each attribute of the population, each attribute of the driver, each type of accident, and each size of accident.

[0078] In addition, the generation unit 103f generates driver evaluation information including information regarding the likelihood of an accident occurring according to each attribute of the population, each attribute of the driver, each type of accident, and each scale of the accident, and the provision unit 103g provides the evaluation information to the terminal device 200.

[0079] Furthermore, the calculation unit 103d may assign a coefficient according to the attributes of the population and the attributes of the driver, and calculate the ratio r and the number of accident occurrence possibilities for the driver by taking into account the coefficient.

[0080] Fig. 7 is an explanatory diagram for assigning coefficients. As shown in Fig. 7, for explanatory variables corresponding to attributes that may be factors in causing accidents (here, average vehicle speed, age, etc.), it is advisable to assign coefficients so that the larger the value, the greater the weight.

[0081] For example, if the population is the transportation industry and the business type is motorcycle courier, the average vehicle speed is likely to be higher than other business types such as trucks and buses. Also, since it is a two-wheeled vehicle, there is a high possibility of injury accidents occurring in the first place. Therefore, if the population is motorcycles, a coefficient will be assigned so that it is weighted more heavily than if it is a truck or bus.

[0082] It is also known that the older a driver is, the narrower their field of vision becomes while driving. Therefore, it is advisable to assign a coefficient to the driver's age, among other attributes, so that the older the driver is, the greater the weighting.

[0083] In addition, the calculation unit 103d may calculate a score value indicating the likelihood of an accident occurring by a regression analysis method, instead of simply multiplying the number of near misses of each driver by the ratio r to calculate the number of accidents likely to occur for each driver.

[0084] In such a case, when an accident occurrence is considered a positive example, the calculation unit 103d assigns a coefficient to the explanatory variables that may be factors in the occurrence of the accident, for example, as shown in Figure 7, so that the weight increases as the explanatory variables have a feature vector of the accident occurrence.

[0085] Specifically, it is considered that the larger the explanatory variables such as average vehicle speed and age shown in FIG. 7 are, the more likely they are to have a characteristic vector of accident occurrence. Therefore, coefficients are assigned to these explanatory variables so that the larger the explanatory variables are, the greater the weighting.

[0086] The calculation unit 103d then solves the regression equation in the multiple regression analysis with the ratio r as one of the explanatory variables to calculate a score indicating the likelihood of an accident as the objective variable. If the score exceeds a predetermined threshold, the driver is determined to have a risk of an accident.

[0087] Other explanatory variables that may be factors in the occurrence of an accident include the number of violations and the number of sudden stops.

[0088] Next, a processing procedure executed by the server device 100 will be described with reference to Fig. 8. Fig. 8 is a flowchart showing a processing procedure executed by the server device 100 according to the embodiment.

[0089] 8, the collection unit 103a collects recorded data (step S101), and the analysis unit 103b performs image analysis on the recorded data to determine whether it is a near miss or an accident (step S102).

[0090] Then, the calculation unit 103d determines whether the number of samples exceeds a predetermined threshold (step S103). If the number of samples exceeds the threshold (step S103, Yes), the calculation unit 103d calculates the ratio r of the total number of accident occurrences to the total number of near misses based on the result of statistical processing by the statistics unit 103c (step S104).

[0091] If the number of samples does not exceed the threshold value (No at step S103), the calculation unit 103d sets a fixed value to the ratio r (step S105).

[0092] Then, the calculation unit 103d calculates the number of accident occurrence possibilities for each driver using the ratio r (step S106), and the determination unit 103e determines whether the number of accident occurrence possibilities exceeds a predetermined threshold (step S107).

[0093] Here, if the number of accident occurrence possibilities exceeds the threshold (step S107, Yes), the determination unit 103e determines that the driver in question has an accident occurrence possibility (step S108). On the other hand, if the number of accident occurrence possibilities does not exceed the threshold (step S107, No), the determination unit 103e determines that the driver in question has no accident occurrence possibility (step S109). Then, the process ends.

[0094] As described above, the server device 100 (corresponding to an example of an "information processing device") according to the embodiment includes a control unit 103 that collects recorded data including video from each vehicle V and performs image analysis. The control unit 103 also determines the likelihood of an accident occurring for each driver by using the ratio r of the total number of accidents to the total number of near misses (corresponding to an example of "accident predictive behavior") for all drivers as a population based on the results of the image analysis, for the number of near misses for each driver, and if the number of samples is insufficient, a fixed value is used in place of the ratio r for the determination.

[0095] Therefore, the server device 100 according to the embodiment can clearly indicate whether or not a driver is likely to cause an accident.

[0096] Furthermore, the control unit 103 uses a known statistical value related to an accident as the fixed value.

[0097] Therefore, according to the server device 100 according to the embodiment, even if the number of samples is insufficient, it is possible to clearly indicate whether a driver is likely to cause an accident based on known statistical values.

[0098] The above statistical values ​​are based on Heinrich's law.

[0099] Therefore, according to the server device 100 according to the embodiment, even if the number of samples is insufficient, it is possible to clearly indicate whether a driver is likely to cause an accident based on the statistical values ​​according to Heinrich's Law.

[0100] The number of samples is at least one of the number of samples of the collected record data, the number of samples of near miss scenes, and the number of samples of scenes where an accident occurs.

[0101] Therefore, according to the server device 100 of the embodiment, even if at least one of the number of samples of recorded data, the number of samples of near miss scenes, and the number of samples of accident scenes is insufficient, it is possible to clearly indicate whether a driver is likely to cause an accident.

[0102] In addition, the control unit 103 calculates the number of accident occurrence possibilities for each driver by multiplying the number of near misses for each driver by the ratio r, and if the number of accident occurrence possibilities exceeds a predetermined threshold, determines the accident occurrence possibility for the corresponding driver.

[0103] Therefore, according to the server device 100 of the embodiment, it is possible to clearly indicate whether a driver is likely to cause an accident based on the number of accident occurrences for each driver calculated from the number of near misses for each driver.

[0104] In addition, the control unit 103 calculates the ratio r for each attribute of the population, each attribute of the driver, each type of accident, and each size of accident, and uses each of these ratios r to calculate the number of possible accidents for each attribute of the population, each attribute of the driver, each type of accident, and each size of accident.

[0105] Therefore, according to the server device 100 of the embodiment, it is possible to clearly indicate whether a driver is likely to cause an accident depending on the attributes of the population, the attributes of the driver, the type of accident, and the magnitude of the accident.

[0106] Furthermore, the control unit 103 calculates the ratio r by applying a coefficient according to the attributes of the population and the attributes of the driver.

[0107] Therefore, according to the server device 100 according to the embodiment, it is possible to clearly indicate whether a driver is likely to cause an accident while taking into consideration weighting according to the attributes of the population and the attributes of the driver.

[0108] Furthermore, the control unit 103 generates driver evaluation information including information relating to the determined accident occurrence probability, and provides it to the terminal device 200 used by users including the driver.

[0109] Therefore, according to the server device 100 of the embodiment, information regarding the possibility of a driver causing an accident can be viewed on the terminal device 200 used by the user, allowing the user to clearly understand whether the driver is likely to cause an accident.

[0110] Moreover, the information processing system 1 according to the embodiment includes a plurality of in-vehicle devices 10 and a server device 100. The in-vehicle device 10 records recorded data including images of the interior and exterior of the vehicle V on which the in-vehicle device 10 is mounted, and transmits the recorded data to the server device 100. The server device 100 collects the recorded data from each vehicle V, performs image analysis, and determines the likelihood of an accident occurring for each driver by using a ratio r of the total number of accidents to the total number of near misses for a population of all drivers based on the results of the image analysis, for the number of near misses for each driver. If the number of samples is insufficient, a fixed value is used in place of the ratio r for the determination.

[0111] Therefore, the information processing system 1 according to the embodiment can clearly indicate whether or not there is a possibility that the driver will cause an accident.

[0112] Furthermore, the information processing method according to the embodiment is an information processing method executed by a server device 100 that collects recorded data including video from each vehicle V and performs image analysis, and includes determining the likelihood of an accident occurring for each driver by using a ratio r of the total number of accidents to the total number of near misses for a population of all drivers based on the results of the image analysis for the number of near misses for each driver, and using a fixed value instead of the ratio r if the number of samples is insufficient.

[0113] Therefore, according to the information processing method of the embodiment, it is possible to clearly indicate whether or not the driver is likely to cause an accident.

[0114] Further advantages and modifications will readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific details and representative embodiments shown and described above. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents. [Explanation of symbols]

[0115] 1. Information Processing Systems 3 Sensor section 3a Camera 3b Accelerometer 3c GPS sensor 10 Onboard equipment 11 Communications Department 12 Storage section 12a Record Information 13 Control Unit 13a Acquisition part 13b Recording section 13c Detector 13d Records Protection Department 13e Upload Section 100 Server device 101 Communications Department 102 Storage section 102a Collected Information DB 102b Statistics information DB 102c Fixed Value Information 102d Driver evaluation information DB 103 Control Unit 103a Collection Department 103b Analysis Department 103c Statistics Department 103d Calculation section 103e Judgment section 103f Generator 103g supply department 200 Terminal Device V vehicle r ratio

Claims

1. An information processing device including a control unit, The control unit Based on the imaging data collected from a plurality of vehicles, the number of near misses, the number of accidents, and the ratio of the number of accidents to the number of near misses are recorded for a population of drivers of the plurality of vehicles; If the number of the population is equal to or greater than a predetermined value, and if a value obtained by multiplying the number of near misses of the target driver by the ratio exceeds a threshold value, it is determined that there is a possibility that an accident will occur due to the target driver, If the number of the population is less than the predetermined value, and if a value obtained by multiplying the number of near misses of the target driver by a fixed value instead of the ratio exceeds a threshold value, it is determined that there is a possibility that an accident will occur due to the target driver. Information processing device.

2. The fixed value is a statistical value according to Heinrich's law. The information processing device according to claim 1 .

3. The control unit The population is made up of multiple drivers with the same industry, business type, gender, age group, or years of driving experience.

3. The information processing device according to claim 1.

4. The control unit Calculating the ratio by assigning a coefficient according to the business type or the age group of the population. The information processing device according to claim 3 .

5. The control unit notifying the target driver of the result of determining the possibility of an accident occurring due to the target driver; 5. The information processing device according to claim 1.

6. An information processing device having a control unit, The control unit Based on the imaging data collected from a plurality of vehicles, the number of near misses, the number of accidents, and the ratio of the number of accidents to the number of near misses are recorded for a population of drivers of the plurality of vehicles; calculate the number of accidents likely to occur for the target driver by multiplying the number of near misses for the target driver by the ratio, and if the calculated number of accidents likely to occur exceeds a threshold, determine that there is a possibility that an accident will occur due to the target driver; Information processing device.

Images