Accident analysis device, accident analysis method, and program

Abstract

To provide an accident analysis device and the like that can favorably handle multiple case masters.SOLUTION: An accident analysis device 10 comprises: a storage unit 100 that stores an accident case DB, a first case master that associates a first tag with an accident case indicated by the accident case DB, and a second case master that associates a second tag with an accident case indicated by the accident case DB; an acquisition unit to acquire vehicle data measured by a sensor included in an accident vehicle and video data captured by a camera included in the accident vehicle; an analysis unit 102 which analyzes a situation of an accident caused by the accident vehicle on the basis of the acquired vehicle data and video data; a tag estimation unit 104 which estimates the first and second tags corresponding to the accident situation on the basis of, the accident situation analyzed by the analysis unit 102; and a search unit 106 which searches the accident case DB on the basis of, the estimated first and second tags by referring to the first and second case masters.SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present invention relates to an accident analysis device, an accident analysis method, and a program.

Background Art

[0002] When a traffic accident occurs, an insurance company calculates a loss ratio according to the accident situation. As an accident analysis device for calculating such a loss ratio and analyzing an accident, for example, Patent Document 1 discloses an accident analysis device that quickly matches an accident situation with past accident cases. An operator of an insurance company calculates a loss ratio based on the content of the matched past accident cases (such as the accident situation and loss ratio of the past case).

[0003] The accident analysis device described in Patent Document 1 matches an accident situation with past accident cases as follows. The accident analysis device described in Patent Document 1 includes an accident case DB (database) in which items (tags) indicating the situation of each case are set for each past accident case. The accident analysis device described in Patent Document 1 analyzes an accident situation to identify a tag indicating the accident situation, and matches the identified tag with the tags set for each accident case, thereby matching the accident situation with past accident cases.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In an accident analysis device such as Patent Document 1, as a method of setting tags for each accident case, for example, it is conceivable to prepare a case master in which an ID (identifier) indicating an accident case is associated with a tag separately from the accident case DB.

[0006] There are several possible methods for assigning tags to each accident case in the case master. For example, by using coarser tags, it becomes more difficult to search for accident cases in detail, but the burden of assigning tags to each accident case is reduced. Conversely, by using finer tags, it becomes possible to search for accident cases in detail, but the burden of assigning tags increases.

[0007] Therefore, it would be convenient for users if accident analysis equipment could appropriately handle multiple case masters tagged using different methods.

[0008] In view of the above circumstances, the present invention aims to provide an accident analysis device, an accident analysis method, and a program that can suitably handle multiple case masters. [Means for solving the problem]

[0009] To achieve the above objective, the accident analysis apparatus according to the first aspect of the present invention is: A storage means for storing accident case data showing accident cases, a first case master which associates the accident cases shown in the accident case data with a first tag, and a second case master which associates the accident cases shown in the accident case data with a second tag. Acquisition means for acquiring vehicle data measured by sensors installed in the accident vehicle and video data captured by cameras installed in the accident vehicle, An analysis means for analyzing the circumstances of the accident caused by the accident vehicle based on the acquired vehicle data and video data, A tag estimation means that estimates the first tag and the second tag corresponding to the circumstances of the accident based on the circumstances of the accident analyzed by the analysis means, A search means for searching the accident case data based on the estimated first tag and second tag by referring to the first case master and the second case master, It is equipped with.

[0010] The system further comprises tag conversion means for converting the estimated first tag to the second tag, The tag estimation means estimates the second tag by converting the estimated first tag into the second tag using the tag conversion means. You may do so.

[0011] The search means ranks each accident case shown in the first case master based on the estimated first tag, and ranks each accident case shown in the second case master based on the estimated second tag. You may do so.

[0012] The search means obtains search results by integrating each accident case ranked in the first case master and each accident case ranked in the second case master. You may do so.

[0013] To achieve the above objective, the accident analysis method according to the second aspect of the present invention is: An acquisition step of acquiring vehicle data measured by sensors installed on the accident vehicle and video data captured by a camera installed on the accident vehicle, An analysis step to analyze the circumstances of the accident caused by the accident vehicle based on the acquired vehicle data and video data, A tag estimation step in which, based on the analyzed circumstances of the accident, a first tag and a second tag corresponding to the circumstances of the accident are estimated, A search step for searching accident case data that shows accident cases, comprising: a search step for searching accident case data based on the estimated first tag and second tag by referring to a first case master which associates the accident cases shown in the accident case data with the first tag and a second case master which associates the accident cases shown in the accident case data with the second tag; It is equipped with.

[0014] To achieve the above objective, the program according to the third aspect of the present invention is: On the computer, An acquisition step of acquiring vehicle data measured by sensors provided in the accident vehicle and video data captured by a camera provided in the accident vehicle; An analysis step of analyzing the situation of the accident that occurred to the accident vehicle based on the acquired vehicle data and the video data; A tag estimation step of estimating a first tag and a second tag corresponding to the situation of the accident based on the analyzed situation of the accident; A search step of searching for accident case data indicating an accident case, referring to a first case master associating the accident case indicated by the accident case data with the first tag and a second case master associating the accident case indicated by the accident case data with the second tag, and searching for the accident case data based on the estimated first tag and second tag; To cause the above to be executed.

Advantages of the Invention

[0015] According to the present invention, it is possible to provide an accident analysis apparatus, an accident analysis method, and a program that can suitably handle a plurality of case masters.

Brief Description of the Drawings

[0016] [Figure 1] It is a diagram showing an example of an accident analysis system according to the present embodiment. <​​​​​​​​​​​​​​​​​​ [Figure 8] This diagram shows the relationship between the accident case database, the first case master, and the second case master in the accident analysis device according to this embodiment. [Figure 9] This table shows an example of a first case master in the accident analysis device according to this embodiment. [Figure 10] This table shows an example of a second case master in the accident analysis device according to this embodiment. [Figure 11] This figure shows an example of the conversion from a first tag to a second tag by the tag conversion unit of the accident analysis device according to this embodiment. [Figure 12] This figure shows an example of how accident cases are ranked by the search unit of the accident analysis device according to this embodiment. [Figure 13] This figure shows an example of a screen displayed on the terminal according to this embodiment. [Figure 14] This flowchart outlines the processing procedure when the accident analysis device according to this embodiment analyzes the circumstances of an accident and searches for accident cases. [Figure 15] This flowchart outlines the processing procedure for tag modification and re-searching after tag modification in the accident analysis device according to this embodiment, when a tag is specified by the operator. [Modes for carrying out the invention]

[0017] Embodiments of the present invention will be described below with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals.

[0018] Figure 1 shows an example of an accident analysis system according to this embodiment. In the accident analysis system 1 according to this embodiment, a drive recorder 1100 (see Figure 2) is installed in the insurance policyholder's vehicle 1000, and the drive recorder 1100 connects to a cloud storage server (not shown) via wireless communication over a network (including a public network). In this embodiment, an example is shown in which the drive recorder 1100 connects to a cloud storage server (not shown) via wireless communication over a network (including a public network), but for example, a mirror-type dedicated terminal device with similar functionality may be prepared and installed instead of the mirror 1300.

[0019] A data storage area for this system is reserved on the cloud storage server. More specifically, a data storage area accessible by the accident analysis device 10 is reserved for each drive recorder 1100 installed in the vehicle 1000.

[0020] The accident analysis device 10 has the function of acquiring vehicle data measured by sensors installed in the accident vehicle 1000 (which may be referred to as "the vehicle" for convenience in the following description) and video data captured by the camera installed in the drive recorder 1100 of the vehicle 1000 from a cloud storage server via the network, and analyzing the circumstances of the accident caused by the vehicle 1000 based on the acquired vehicle data and video data. In addition, the accident analysis device 10 has the function of matching the circumstances of the accident caused by the vehicle 1000 with accident cases by comparing the circumstances of the accident obtained through the analysis with an accident case database that associates the circumstances of past accidents with information on past accident cases (hereinafter referred to as "accident cases").

[0021] In the following explanation, "sensors equipped in vehicle 1000" do not refer to sensors directly built into vehicle 1000, but rather to sensors equipped in the drive recorder 1100 mounted on vehicle 1000. As will be described later, the drive recorder 1100 is equipped with sensors such as a positioning sensor to acquire the absolute position of vehicle 1000 and an acceleration sensor to measure the acceleration of vehicle 1000. Since the drive recorder 1100 is mounted on vehicle 1000, the various sensors equipped in the drive recorder 1100 are also considered "sensors equipped in vehicle 1000."

[0022] Terminal 20 is a terminal operated by, for example, an insurance company operator, and displays the accident situation analyzed by the accident analysis device 10, accident examples corresponding to the accident situation, and images generated by the accident analysis device 10. Terminal 20 can be any information processing device equipped with a display, such as a PC, notebook PC, tablet terminal, or smartphone. Terminal 20 is connected to the accident analysis device via a network so as to be able to communicate. When an insurance company operator receives an accident report from, for example, an insurance policyholder driving vehicle 1000, they operate Terminal 20 to grasp the accident situation, accident examples, etc., and calculate the percentage of fault based on the results of matching the accident situation with the accident examples.

[0023] Figure 2 shows the view from inside the vehicle 1000, looking forward of the vehicle 1000. The drive recorder 1100 includes a camera and, as shown in the figure, is mounted on the front of the vehicle 1000 or on the windshield so as to be able to capture images of at least the direction in which the vehicle 1000 is traveling. The camera may also be capable of capturing images of the side and rear of the vehicle 1000. As shown in Figure 2, a well-known automatic diagnostic system 1400 (for example, OBD (On-Board Diagnostic system -II)) is installed inside the vehicle 1000, and this automatic diagnostic system 1400 is connected to the drive recorder 1100.

[0024] Figure 3 shows a functional block diagram of the drive recorder 1100 according to this embodiment. The drive recorder 1100 mounted on the vehicle 1000 includes a communication unit 1120, a positioning unit 1130, a recording unit 1140, an audio recording unit 1150, an acceleration measurement unit 1160, an automatic diagnostic data acquisition unit 1170, a control unit 1180, and a data storage unit 1200.

[0025] The control unit 1180 consists of processors such as a CPU (Central Processing Unit) and a GPU (Graphical Processing Unit), and causes each component to execute processing related to this customer service.

[0026] The communication unit 1120 has the function of transmitting data to a cloud storage server via packet communication using wireless communication, for example, a public telephone network.

[0027] When instructed by the control unit 1180, the positioning unit 1130 acquires the absolute position (e.g., latitude and longitude) of the vehicle 1000 using, for example, GPS (Global Positioning System), and stores it in the data storage unit 1200.

[0028] The recording unit 1140 stores video data (image data) captured by a camera, for example, mounted on a drive recorder 1100, in the data storage unit 1200. The sound recording unit 1150 stores sound data input from a microphone in the data storage unit 1200. The sound recording unit 1150 may be integrated with the recording unit 1140. The recording unit 1140 will also continuously record and store the data in the data storage unit 1200 when instructed to operate by, for example, the control unit 1180. The control unit 1180 will be able to extract video data from a certain period before a specific time and video data from a certain period after that specific time. The same applies to the sound recording unit 1150. The recording unit 1140 is capable of capturing video images from outside the vehicle 1000 and video images from inside the vehicle 1000.

[0029] The acceleration measurement unit 1160 measures the value of acceleration, for example, using an acceleration sensor, and outputs it to the control unit 1180. The automatic diagnostic data acquisition unit 1170 acquires automatic diagnostic data from the automatic diagnostic system 1400 installed inside the vehicle 1000 when instructed by the control unit 1180.

[0030] The data storage unit 1200 consists of a memory, an HDD (Hard Disk Drive), and / or an SSD (Solid State Drive) or other storage device, and pre-stores device data such as company name, organization name, vehicle registration number, driver identifier (ID), and telephone number, and also stores data acquired by each component in response to instructions from the control unit 1180.

[0031] Next, using Figure 4, we will explain the processing steps in the accident analysis system shown in Figure 1, from the time an accident occurs until the accident-related data is sent to the cloud storage server.

[0032] The control unit 1180 of the drive recorder 1100 mounted on the vehicle 1000 continuously causes the acceleration measuring unit 1160 to measure acceleration and determines whether the measured acceleration is above a predetermined threshold. Here, the control unit 1180 assumes that an acceleration above the threshold has been detected (Figure 4: Step S100).

[0033] When acceleration exceeding a threshold is detected, the control unit 1180 reads equipment data from the data storage unit 1200 and extracts vehicle data including impact event data, video data, etc. (step S101). The impact event data includes the date and time obtained from the clock, position data obtained by the positioning unit 1130, the detected acceleration, and automatic diagnostic data obtained by the automatic diagnostic data acquisition unit 1170. The video data includes video data captured by the recording unit 1140 and sound data recorded by the sound recording unit 1150. As mentioned above, the video data, etc. includes video data, etc. for a certain period before the point in time when acceleration exceeding the threshold is detected, and video data, etc. for a certain period after that point in time. The automatic diagnostic data includes data indicating whether or not there is damage to the engine, battery, fuel system, etc. Note that equipment data and impact event data are collectively referred to as vehicle data.

[0034] Then, the control unit 1180 sends the extracted data (vehicle data (equipment data, impact event data) and video data, etc.) to the cloud storage server to the communication unit 1120 (step S102). After that, the control unit 1180 terminates the subsequent processing (step S103).

[0035] When the cloud storage server receives extracted data from the drive recorder 1100, it stores it in a data storage area identified by, for example, the vehicle registration number or telephone number included in the device data (step S103).

[0036] Next, using Figures 5 to 14, we will explain the analysis of accident situations by the accident analysis device 10 in the accident analysis system shown in Figure 1, and how to match accident situations with accident case examples.

[0037] As described above, the accident analysis device 10 acquires vehicle data and video data from a cloud storage server via the network, analyzes the circumstances of the accident caused by vehicle 1000 based on the acquired vehicle data and video data, and has the function of matching the circumstances of the accident caused by vehicle 1000 with accident examples. Furthermore, the accident analysis device 10 has the function of generating an image showing the circumstances of the accident caused by vehicle 1000 by mapping the positions of vehicle 1000 and other vehicles obtained by analyzing the circumstances of the accident onto map data. The image showing the circumstances of the accident caused by vehicle 1000 can be of any kind, for example, an overhead view or a video.

[0038] The accident analysis device 10 may consist of one or more physical information processing devices such as a mainframe, workstation, or personal computer (PC), or it may be configured using a virtual information processing device that operates on a hypervisor, or it may be configured using a cloud server.

[0039] Figure 5 shows an example of the hardware configuration of the accident analysis device 10. The accident analysis device 10 has a processor 11 such as a CPU (Central Processing Unit) and a GPU (Graphical Processing Unit), a storage device 12 such as memory, an HDD (Hard Disk Drive) and / or an SSD (Solid State Drive), a communication interface 13 for wired or wireless communication, an input device 14 for receiving input operations, and an output device 15 for outputting information. The input device 14 is, for example, a keyboard, a touch panel, a mouse and / or a microphone. The output device 15 is, for example, a display and / or a speaker.

[0040] Figure 6 is a functional block diagram of the accident analysis device 10 according to this embodiment. The accident analysis device 10 includes a storage unit 100, an acquisition unit 101, an analysis unit 102, a generation unit 103, a tag estimation unit 104, a tag conversion unit 105, a search unit 106, and an output unit 107. The storage unit 100 can be realized using a storage device 12 provided by the accident analysis device 10. The acquisition unit 101, the analysis unit 102, the generation unit 103, the tag estimation unit 104, the tag conversion unit 105, the search unit 106, and the output unit 107 can be realized by the processor 11 of the accident analysis device 10 executing a program stored in the storage device 12. The program can be stored in a storage medium. The storage medium storing the program may be a non-transitory computer-readable medium. The non-transitory storage medium is not particularly limited, but may be a storage medium such as a USB memory or CD-ROM.

[0041] The memory unit 100 stores the accident case database, the first case master, the second case master, and the map data database. The memory unit 100 corresponds to the storage means.

[0042] The accident case database is a database that shows past accident cases. For example, the accident case database is a database of court cases related to automobile accidents. As shown in Figure 7, for example, the accident case database includes a case ID that identifies each accident case, the content of the accident case, the percentage of fault in the accident case, and modifying factors. Here, the percentage of fault is the basic percentage of fault and can be modified according to the modifying factors. Modifying factors are factors that modify the percentage of fault according to the circumstances of the accident. For example, the percentage of fault may be modified if the accident occurred at night, if the condition of the vehicle involved in the accident was poor, or if the pedestrian was elderly. The accident case database corresponds to accident case data.

[0043] As detailed below, the relationship between the accident case database and the first and second case masters is shown in Figure 8.

[0044] The first case master is a master that associates accident cases shown in the accident case database with first tags corresponding to the content of each accident case. First tags are more granular than second tags, which will be described later. Therefore, the burden of setting tags for each accident case is greater than that of second tags.

[0045] The second case master is a master that associates accident cases shown in the accident case database with second tags corresponding to the content of each accident case. The second tags are coarser in granularity than the first tags. Therefore, the burden of setting tags for each accident case is less than that of the first tags.

[0046] The first case master is shown in Figure 9, for example, and the second case master is shown in Figure 10, for example. In the examples shown in Figures 9 and 10, both the first and second case masters refer to accident cases included in the accident case database, using the case ID from the accident case database shown in Figure 7. The first case master, shown in Figure 9, has finer granularity in road-related tags than the second case master, shown in Figure 10. Also, the second case master has more accident cases with tags set than the first case master, shown in Figure 9. This is because the first case master, with its finer granularity of tags, places a greater burden on tag setting, and it is assumed that the progress of tag setting will be slower compared to the second case master.

[0047] In addition to what is shown in Figure 9, examples of first tags included in the first case master include tags indicating priority relationships between roads, whether a right-turning vehicle can move to the center, whether a left-turning vehicle can move to the far left, the color of the traffic light when the vehicle and the target vehicle enter, the speed when the vehicle and the target vehicle enter, whether or not there was an accident caused by another vehicle, and whether or not there were injuries. On the other hand, in addition to what is shown in Figure 10, examples of second tags included in the second case master include the movement of the vehicle, the vehicle speed at the time of collision, and the color of the traffic light when the vehicle and the target vehicle entered.

[0048] The accident case database, the first case master, and the second case master are stored in the storage unit 100, for example, by the administrator of the accident analysis device 10 manually entering the data. Alternatively, accident case data distributed by a third-party company handling accident cases may be stored in the storage unit 100 as the accident case database, and only the first case master and the second case master may be manually entered by the administrator. Alternatively, the third-party company may distribute the accident case database and the second case master, and only the first case master may be manually entered by the administrator.

[0049] Because the granularity and content of the tags set in the first case study master and the second case study master differ, tags will be set from different perspectives in the first case study master and the second case study master.

[0050] Refer to Figure 6 again. The map data database includes various data such as road data, road width, direction of travel, road type, traffic signs (stop, no entry, etc.), speed limits, traffic light locations, and the number of intersecting roads at intersections.

[0051] As mentioned above, in Figure 6, the storage unit 100 is implemented using the memory device 12 provided by the accident analysis device 10. However, the storage unit 100 may also be implemented using an external server capable of communicating with the accident analysis device 10.

[0052] The acquisition unit 101 has the function of acquiring vehicle data and video data measured and photographed by the vehicle 1000 (accident vehicle) from a cloud storage server. The acquisition unit 101 also has the function of acquiring information related to the operation of the terminal 20 by the operator (hereinafter referred to as "operation information"). The acquisition unit 101 corresponds to the acquisition means.

[0053] The analysis unit 102 has the function of analyzing the circumstances of an accident caused by vehicle 1000 based on the vehicle data and video data acquired by the acquisition unit 101. The analysis unit 102 corresponds to the analysis means. The circumstances of the accident analyzed by the analysis unit 102 may include at least the road conditions of the accident caused by vehicle 1000 (whether the road is a public road or an expressway, whether the road is a straight road or an intersection, etc.), the color of the traffic signals when vehicle 1000 passes through an intersection, the priority relationship when passing through the intersection, and whether or not the speed of vehicle 1000 exceeds the speed limit.

[0054] Furthermore, the vehicle data of vehicle 1000 includes information indicating the absolute position of vehicle 1000 measured by the GPS of the positioning unit 1130. The analysis unit 102 may estimate the absolute position of other vehicles based on the information indicating the absolute position of vehicle 1000 (position data acquired by the positioning unit 1130) and information indicating the relative positional relationship between vehicle 1000 and other vehicles, obtained by analyzing the images of other vehicles captured in the video data. The analysis unit 102 may also estimate the absolute positions of vehicle 1000 and other vehicles in a time series. The analysis unit 102 may also estimate the absolute position of its own vehicle more precisely using SfM (Structure from Motion) technology, based on the information indicating the absolute position measured by GPS and the data of each frame of the video data.

[0055] Furthermore, the analysis unit 102 may analyze the circumstances of the accident by comparing the image size of the portion of the video data in which other vehicles are visible with data showing the correspondence between image size and distance, and estimating the distance between vehicle 1000 and the other vehicle, which is one of the pieces of information showing the relative positional relationship between vehicle 1000 and the other vehicle.

[0056] Furthermore, the analysis unit 102 may analyze the accident situation by comparing the difference between the coordinates of the parts of the video data in which other vehicles are visible and the center coordinates of the video data with data showing the correspondence between coordinates and angles, and estimating the angle difference between the direction of travel of vehicle 1000 and the direction in which the other vehicles are located, which is one of the pieces of information showing the relative positional relationship between vehicle 1000 and other vehicles.

[0057] The generation unit 103 has the function of generating an image showing the circumstances of an accident caused by vehicle 1000 by referring to the map data DB of the storage unit 100 and mapping the absolute position of vehicle 1000 and the absolute positions of other vehicles onto the map data based on the analysis results by the analysis unit 102. The image may include an overhead view or a video. For example, the generation unit 103 may generate a video in which images showing the circumstances of the accident are arranged in chronological order.

[0058] The tag estimation unit 104 estimates a first tag and a second tag representing the accident situation analyzed by the analysis unit 102. For example, the tag estimation unit 104 estimates the tags related to road conditions from the first tag and the second tag based on the road conditions analyzed by the analysis unit 102. As will be described later, some or all of the second tags may be estimated by converting the estimated first tags by the tag conversion unit 105. The tag estimation unit 104 corresponds to the tag estimation means.

[0059] The tag conversion unit 105 converts the first tag estimated by the tag estimation unit 104 into a second tag. For example, as shown in Figure 11, when the first tags estimated by the tag estimation unit 104 are "Road type: General road" and "Road shape: Intersection", the tag conversion unit 105 converts these two first tags into a single second tag, "Road form: Intersection". The tag conversion unit 105 corresponds to the tag conversion means.

[0060] In addition to the above, for example, if the estimated first tag is "Road type: Intersection" and "Relative position: Left side of vehicle, right side of target", the tag conversion unit 105 may convert these two first tags into a single second tag, "Accident type: Head-on collision, side collision, contact". In this case, the first tag "Road type: Intersection" is used to convert to the second tag "Road type: Intersection", and also to convert to the second tag "Accident type: Head-on collision, side collision, contact". In other words, one first tag can be used to convert to two or more second tags.

[0061] The search unit 106 refers to the first case master and the second case master in the storage unit 100 and searches for accident cases based on the first tag and the second tag estimated by the tag estimation unit 104. Since the tags estimated by the tag estimation unit 104 are tags that represent accident situations, the search unit 106 can match accident situations with accident cases. The search unit 106 corresponds to the search means.

[0062] Furthermore, the search unit 106 may not only search for accident cases, but also rank each accident case in the case master (first case master or second case master) according to the degree of match between the estimated tag (first tag or second tag) and the tag set for each accident case in the case master, and use the results obtained from this ranking as the search results. Alternatively, the search results may be an integrated version of the results obtained by ranking each individual accident case. The following explanation will refer to Figure 12.

[0063] First, the search unit 106 ranks each accident case in the first and second case masters, respectively, based on the degree of match between the estimated tags and the tags set for each accident case in the case master, as shown in the upper part of Figure 12. Here, the case IDs in the second case master shown in Figure 12 that are shaded indicate accident cases that also exist in the first case master. Accident cases with a match degree of zero with the estimated tags are excluded from ranking.

[0064] For the "matching score," you may use a simple matching coefficient, or other metrics. For example, you could pre-determine different scores for each tag and use the total score of matching tags as the "matching score."

[0065] Next, the search unit 106 integrates the ranking results for the first case master and the second case master, respectively. Specifically, for example, based on the ranking results of the second case master, for accident cases in the second case master that also exist in the first case master, the ranking in the first case master is applied to integrate the ranking results. For example, in the case shown in Figure 12, in the first case master, the rankings of case IDs 002, 017, 008, and 010 are ranked from 1st to 4th, while in the second case master, the rankings of case IDs 002, 017, 008, and 010 are ranked 2nd, 1st, 3rd, and 5th, respectively. In this case, the relative rankings in the first case master take precedence during integration, and the rankings of case IDs 002, 017, 008, and 010 at the time of integration become 1st, 2nd, 3rd, and 5th, respectively.

[0066] The reason for applying the ranking from the first case study master during the integration is that the first case study master has a finer granularity of tags, making the ranking more appropriate.

[0067] Furthermore, as will be explained in more detail later, the search unit 106 also has a function to perform a re-search based on the operation information acquired by the acquisition unit 101 when the operator modifies the tag by operating the terminal 20.

[0068] The output unit 107 has the function of outputting to the terminal 20 information showing the accident situation analyzed by the analysis unit 102, information showing the tags estimated by the tag estimation unit 104, information showing the search results from the search unit 106, and an image showing the accident situation generated by the generation unit 103. Since the search results obtained by the search unit 106 only include the case ID and rank, the output unit 107 refers to the accident case database and obtains detailed information of the accident case (such as "content," "percentage of fault," and "modifying factors" shown in Figure 7) based on the case ID, and includes this detailed information in the information showing the search results.

[0069] Based on the output of the output unit 107, a screen like the one shown in Figure 13 is displayed on the terminal 20. The screen shown in Figure 13 includes at least images and information showing the accident situation analyzed by the analysis unit 102 (images are generated by the generation unit 103), a first tag and a second tag estimated by the tag estimation unit 104, and search results from the search unit 106. In addition, the output unit 107 can display the screens described below based on the operation information acquired by the acquisition unit 101.

[0070] The operator operating terminal 20 can view the video of the accident recorded by the vehicle 1000's drive recorder 1100 by selecting the "View video of the accident" button on the screen. This is achieved when the output unit 107 displays the video based on the operation information indicating that "View video of the accident" has been selected.

[0071] Furthermore, the operator operating terminal 20 can select "View Details" in the search results to check the details of each accident case listed in the search results, factors for modifying the percentage of fault, etc. The operator compares the accident situation with the details of the accident case and calculates the percentage of fault based on the accident case. This is achieved when the output unit 107 displays the details of the accident case, etc., based on the operation information indicating that "View Details" has been selected.

[0072] Furthermore, the operator operating terminal 20 can modify tags by selecting the "Modify Tags" button on the screen. "Modifying tags" includes correcting estimated tags, deleting some estimated tags, and adding new tags. When tags are modified, terminal 20 displays the search results using the modified tags. For example, if the tag estimation unit 104 fails to estimate a tag, or if the tags estimated by the tag estimation unit 104 are inappropriate, the operator can review the video, correct the tags to be appropriate, and perform the search again. This is achieved by the search unit 106 performing a search based on the modified tags based on the operation information indicating that "Modify Tags" was selected and the tags were subsequently modified, and the output unit 107 displaying the search results after the tag modification.

[0073] Furthermore, since the functions of the accident analysis device 10 may be realized in conjunction with the terminal 20, there may also be functions provided on the terminal 20 side. Alternatively, the accident analysis device 10 and the terminal 20 may be integrated into a single information processing device.

[0074] Next, the processing procedure for analyzing accident situations and searching for accident cases by the accident analysis device 10 will be explained with reference to Figure 14. Figure 14 is a flowchart outlining the processing procedure when the accident analysis device 10 analyzes accident situations and searches for accident cases. In the following explanation, it is assumed that vehicle data and video data each contain time information or synchronization information. That is, in this embodiment, when analyzing video data at a certain point in time, it is possible to perform the analysis using vehicle data corresponding to that point in time, and conversely, when analyzing vehicle data at a certain point in time, it is possible to perform the analysis using video data corresponding to that point in time.

[0075] First, the acquisition unit 101 of the accident analysis device 10 acquires vehicle data and video data of the vehicle 1000 involved in the accident (steps S10 and S11). Specifically, the accident analysis device 10 acquires the vehicle data and video data from a cloud storage server via a network. Alternatively, the data may be recorded on a non-temporary storage medium such as an SD card or USB memory, and the vehicle data and video data may be imported into the accident analysis device 10 by connecting the recording medium to the accident analysis device 10. The processing in steps S10 and S11 corresponds to the acquisition step.

[0076] Next, the analysis unit 102 of the accident analysis device 10 analyzes the circumstances of the accident caused by vehicle 1000 based on the vehicle data acquired in step S10 and the video data acquired in S11 (step S12). The processing in step S12 corresponds to the analysis step.

[0077] Next, the tag estimation unit 104 of the accident analysis device 10 estimates a first tag and a second tag indicating the accident situation based on the analysis results in step S12 (step S13). As described above, in estimating some or all of the second tags, the estimated first tags may be converted into second tags by the tag conversion unit 105 to estimate the second tags. The processing in step S13 corresponds to the tag estimation step.

[0078] Next, the search unit 106 of the accident analysis device 10 refers to the first case master and ranks the accident cases based on the first tag estimated in step S13 (step S14). Similarly, the search unit 106 refers to the second case master and ranks the accident cases based on the second tag estimated in step S13 (step S15). The search unit 106 integrates the ranking results from steps S14 and S15 to obtain the search results (step S16). The processing from steps S14 to S16 corresponds to the search step.

[0079] Next, the generation unit 103 of the accident analysis device 10 generates an image showing the accident situation based on the analysis results obtained in step S12 (step S17). Note that the process in step S17 may be performed between the processes in steps S12 to S16, or it may be performed in parallel with these processes.

[0080] The output unit 107 of the accident analysis device 10 then outputs to the terminal 20 information indicating the analysis results (accident situation) obtained in step S12, information indicating the tags estimated in step S13, information indicating the search results obtained in steps S14 to S16, and an image indicating the accident situation generated in step S17 (step S18), and the accident analysis device 10 finishes the analysis and search process.

[0081] These processes may be performed in any order, as long as no inconsistencies arise in the process.

[0082] Next, the procedure for re-searching by the accident analysis device 10 when an operator modifies a tag will be explained with reference to Figure 15. Since the procedure shown in Figure 15 is a re-search, it is executed when the operator modifies the tag by operating the terminal 20 after the procedure shown in Figure 14 has been performed and the search results have been displayed.

[0083] First, the acquisition unit 101 of the accident analysis device 10 acquires operation information indicating that the operator has modified the tag by operating the terminal 20 (step S20).

[0084] Next, the search unit 106 of the accident analysis device 10 identifies the modified tag based on the operation information (step S21).

[0085] Next, the search unit 106 searches again for accident cases containing the tag identified in step S21 from the search results, while maintaining the ranking order (step S22). For example, consider the case where the initial search results were the merged results shown in Figure 12. In this case, suppose there are three accident cases containing the corrected tag, with case IDs 017, 032, and 010. The ranks of these three accident cases before tag correction were 2nd, 4th, and 5th, respectively. In this case, the search unit 106 assigns the accident cases with case IDs 017, 032, and 010 to ranks 1st, 2nd, and 3rd, respectively. Since the ranks before tag correction were 2nd, 4th, and 5th, respectively, the ranks after tag correction become 1st, 2nd, and 3rd, respectively.

[0086] Then, the output unit 107 of the accident analysis device 10 outputs information indicating the re-search results after tag correction obtained in step S22 to the terminal 20 (step S23), and the accident analysis device 10 terminates the re-search process.

[0087] According to the embodiment described above, the accident analysis device 10 estimates two types of tags, a first tag and a second tag, from the accident situation, and searches for past accident cases from two case masters, a first case master and a second case master, based on these tags. Therefore, the accident analysis device 10 can suitably handle multiple case masters.

[0088] (modified version) Furthermore, this invention is not limited to the above embodiments, and various modifications and applications are possible. For example, the vehicle 1000, drive recorder 1100, cloud storage server, accident analysis device 10, and terminal 20 do not have to have all the technical features shown in the above embodiments, and may have some of the configurations described in the above embodiments so as to solve at least one problem in the prior art. In addition, at least some of each of the following modifications may be combined.

[0089] In the above embodiment, the search unit 106 obtained search results by integrating the ranking results based on the first tag and the ranking results based on the second tag. However, the results of each ranking can also be used as search results as they are. In this case, the search results shown in Figure 13 will display two types of search results, one based on the first tag and the other on the second tag.

[0090] In the above embodiment, the first case master was set to have fine-grained tags, and the second case master was set to have coarse-grained tags. Alternatively, for example, the first case master could be set to have tags specialized for a specific type of accident case, and the second case master could be set to have tags related to general accident cases. For example, by setting the first case master to have tags specialized for four-wheeled vehicle-to-four-wheeled vehicle accidents, it becomes possible to search for four-wheeled vehicle-to-four-wheeled vehicle accident cases with high accuracy, while also being able to search for general accident cases.

[0091] In the above embodiment, we handled two case masters, the first case master and the second case master, but we can handle three or more case masters in the same way.

[0092] In the above embodiment, an example was shown in which the drive recorder 1100 is installed in the insurance policyholder's vehicle 1000, but this is just one example. Instead of the drive recorder 1100, the insurance policyholder's smartphone may be used. External and internal video data can be captured using cameras mounted on the back and front of the smartphone.

[0093] In the above embodiment, the drive recorder 1100 is equipped with various sensors, but the vehicle 1000 may have the various sensors built in instead of the drive recorder 1100. In this case, the vehicle 1000 will have the same configuration as the communication unit 1120, positioning unit 1130, automatic diagnostic data acquisition unit 1170, control unit 1180, acceleration measurement unit 1160, and data storage unit 1200 in the above embodiment. The accident analysis device 10 will then acquire video data from the drive recorder 1100 and vehicle data from the vehicle 1000.

[0094] The accident analysis system 1 can be implemented using a regular computer without requiring specialized equipment. For example, the accident analysis system 1 that performs the above-mentioned processes may be configured by installing a program for executing one of the above-mentioned tasks from a recording medium into the computer. Alternatively, multiple computers may work together to form a single accident analysis system 1.

[0095] Furthermore, the method for supplying programs to computers is arbitrary. For example, they may be supplied via communication lines, communication networks, communication systems, etc.

[0096] Furthermore, if the OS (Operating System) provides some of the above-mentioned functions, then the parts not provided by the OS should be provided by the program.

[0097] The embodiments described above are provided to facilitate understanding of the present invention and are not intended to limit its interpretation. The flowcharts, sequences, elements, and their arrangement, materials, conditions, shapes, and sizes described in the embodiments are not limited to those exemplified and can be modified as appropriate. Furthermore, configurations shown in different embodiments can be partially substituted or combined. [Explanation of Symbols]

[0098] 1... Accident analysis system, 10... Accident analysis device, 11... Processor, 12... Memory device, 13... Communication interface, 14... Input device, 15... Output device, 20... Terminal, 100... Storage unit, 101... Acquisition unit, 102... Analysis unit, 103... Generation unit, 104... Tag estimation unit, 105... Tag conversion unit, 106... Search unit, 107... Output unit, 1000... Vehicle, 1100... Drive recorder, 1120... Communication unit, 1130... Positioning unit, 1140... Recording unit, 1150... Audio recording unit, 1160... Acceleration measurement unit, 1170... Automatic diagnostic data acquisition unit, 1180... Control unit, 1200... Data storage unit, 1300... Mirror, 1400... Automatic diagnostic system.

Claims

1. A storage means for storing accident case data showing accident cases, a first case master which associates the accident cases shown in the accident case data with a first tag, and a second case master which associates the accident cases shown in the accident case data with a second tag. Acquisition means for acquiring vehicle data measured by sensors installed in the accident vehicle and video data captured by cameras installed in the accident vehicle, An analysis means for analyzing the circumstances of the accident caused by the accident vehicle based on the acquired vehicle data and video data, A tag estimation means that estimates the first tag and the second tag corresponding to the circumstances of the accident based on the circumstances of the accident analyzed by the analysis means, A search means for searching the accident case data based on the estimated first tag and second tag by referring to the first case master and the second case master, Equipped with, The first tag is a tag with finer granularity than the second tag. The number of accident cases associated with the second tag using the second case master is greater than the number of accident cases associated with the first tag using the first case master. An accident analysis device characterized by the following features.

2. The system further comprises tag conversion means for converting the estimated first tag to a second tag, The tag estimation means estimates the second tag by converting the estimated first tag into the second tag using the tag conversion means. The accident analysis apparatus according to feature 1.

3. The search means ranks each accident case shown in the first case master based on the estimated first tag, and ranks each accident case shown in the second case master based on the estimated second tag. The accident analysis apparatus according to claim 1 or 2.

4. The search means obtains search results by integrating each accident case ranked in the first case master and each accident case ranked in the second case master. The accident analysis apparatus according to feature 3.

5. A computer-based accident analysis method, An acquisition step of acquiring vehicle data measured by sensors installed on the accident vehicle and video data captured by a camera installed on the accident vehicle, An analysis step to analyze the circumstances of the accident caused by the accident vehicle based on the acquired vehicle data and video data, A tag estimation step in which, based on the analyzed circumstances of the accident, a first tag and a second tag corresponding to the circumstances of the accident are estimated, A search step for searching accident case data that shows accident cases, the search step of searching the accident case data based on the estimated first tag and second tag by referring to a first case master which associates the accident cases shown in the accident case data with the first tag and a second case master which associates the accident cases shown in the accident case data with the second tag, Equipped with, The first tag is a tag with finer granularity than the second tag. The number of accident cases associated with the second tag using the second case master is greater than the number of accident cases associated with the first tag using the first case master. An accident analysis method characterized by the following.

6. On the computer, An acquisition step of acquiring vehicle data measured by sensors installed on the accident vehicle and video data captured by a camera installed on the accident vehicle, An analysis step to analyze the circumstances of the accident caused by the accident vehicle based on the acquired vehicle data and video data, A tag estimation step in which, based on the analyzed circumstances of the accident, a first tag and a second tag corresponding to the circumstances of the accident are estimated, A search step for searching accident case data that shows accident cases, the search step of searching the accident case data based on the estimated first tag and second tag by referring to a first case master which associates the accident cases shown in the accident case data with the first tag and a second case master which associates the accident cases shown in the accident case data with the second tag, Make it run, The first tag is a tag with finer granularity than the second tag. The number of accident cases associated with the second tag using the second case master is greater than the number of accident cases associated with the first tag using the first case master. A program characterized by the following features.

Images