JP7743328B2 - Vehicle search device, vehicle search method, and vehicle search program - Google Patents

Description

The present invention relates to a vehicle search device, a vehicle search method, and a vehicle search program.

Devices that extract vehicle information from captured images are known (see, for example, Patent Document 1). Patent Document 1 discloses that vehicle characteristic information such as the license plate, body color, and model are detected from images captured by an on-board camera, and this information is then sent to a center server.

Japanese Patent Application Laid-Open No. 2019-197496

There is a need for a device that can accurately search for vehicles in motion even when sufficient information cannot be read from captured images, such as when license plates cannot be read.

The present invention was made in consideration of these circumstances, and aims to provide a vehicle search device, vehicle search method, and vehicle search program that can accurately extract a desired vehicle even when vehicle information is difficult to read from captured images.

In order to solve the above problem, the vehicle search device of the present invention includes, for example, a data acquisition unit that acquires image data, image capture date and time information that is information about the time the image data was acquired, and location information that is information about the location where the image data was acquired; a reading unit that reads vehicle information from the image data, which is information about the vehicle, including the vehicle's color and direction of travel; a storage unit that stores search data that associates the vehicle information with the location information; an input unit that accepts search conditions that include information about the characteristics, date and time, and location of the vehicle to be searched for; and a search unit that extracts vehicles that are estimated to be the vehicle to be searched for from the search data as candidate vehicles based on the search conditions. and a map generation unit that arranges information indicating the candidate vehicles on a map based on the position information, wherein the extraction unit extracts a first vehicle that matches the characteristics of the searched vehicle from a first search target area that includes a position included in the search criteria during a first time range that is a predetermined time range that includes a first point in time included in the search criteria, and extracts a second vehicle that matches the characteristics of the searched vehicle from a second search target area in which the first vehicle is estimated to be located during a second time range that is a predetermined time after the first time range or a predetermined time before the first time range, and extracts the first vehicle and the second vehicle as the candidate vehicles.

A vehicle search method according to another aspect of the present invention includes, for example, an acquisition step of acquiring image data, image capture date and time information that is information about the time the image data was acquired, and location information that is information about the location where the image data was acquired; a reading step of reading vehicle information from the image data, which is information about the vehicle, including the vehicle's color and direction of travel; a storage step of storing search data that associates the vehicle information with the location information; an input step of accepting search conditions that include information about the characteristics, date and time, and location of the vehicle to be searched for; and a step of extracting vehicles that are estimated to be the vehicle to be searched for from the search data as candidate vehicles based on the search conditions. and a generating step of arranging information indicating the candidate vehicles on a map based on the position information, wherein the extracting step extracts a first vehicle that matches the characteristics of the searched vehicle from a first search target area that includes a position included in the search criteria in a first time range, which is a predetermined time range that includes a first point in time included in the search criteria, and extracts a second vehicle that matches the characteristics of the searched vehicle from a second search target area in which the first vehicle is estimated to be located in a second time range that is a predetermined time after the first time range or a predetermined time before the first time range, and extracts the first vehicle and the second vehicle as the candidate vehicles.

A vehicle search program according to another aspect of the present invention includes, for example, a computer, a data acquisition unit that acquires image data, image capture date and time information that is information about the time the image data was acquired, and location information that is information about the location where the image data was acquired, a reading unit that reads vehicle information from the image data, which is information about the vehicle, including the color and direction of travel of the vehicle, a storage unit that stores search data that associates the vehicle information with the location information, an input unit that accepts search conditions that include information about the characteristics, date and time, and location of a vehicle to be searched for, and a search program that extracts vehicles that are estimated to be the vehicle to be searched for from the search data as candidate vehicles based on the search conditions. and a map generation unit that arranges information indicating the candidate vehicles on a map based on the position information, wherein the extraction unit extracts a first vehicle that matches the characteristics of the searched vehicle from a first search target area that includes a position included in the search criteria in a first time range that is a predetermined time range that includes a first point in time included in the search criteria, extracts a second vehicle that matches the characteristics of the searched vehicle from a second search target area in which the first vehicle is estimated to be located in a second time range that is a predetermined time after the first time range or a predetermined time before the first time range, and extracts the first vehicle and the second vehicle as the candidate vehicles.
The computer program can be provided by downloading it via a network such as the Internet, or by recording it on various computer-readable recording media such as a CD-ROM.

In any of the above aspects of the present invention, a vehicle estimated to be the vehicle to be searched is extracted as a candidate vehicle from search data that associates vehicle information with location information. Specifically, a first vehicle that matches the characteristics of the vehicle to be searched is extracted from a first search target area that includes a location included in the search criteria within a first time range (a predetermined time range that includes a first point in time included in the input search criteria). A second vehicle that matches the characteristics of the vehicle to be searched is extracted from a second search target area in which the first vehicle is estimated to be located within a second time range (a predetermined time after the first time range or a predetermined time before the one-hour range). The first and second vehicles estimated to be the vehicle to be searched are extracted as candidate vehicles. This allows for accurate extraction of the desired vehicle even when vehicle information is difficult to read from captured images.

The map generation unit may display a virtual route connecting the first vehicle and the second vehicle as the estimated travel route of the candidate vehicle. This allows the travel route of the searched vehicle to be clearly understood on the map.

The extraction unit may estimate the travel distance of the searched vehicle during the specified time period and use the estimated travel distance as the radius of the first search target area and the second search target area. This allows the search target area to be adjusted in consideration of the vehicle's travel distance. This enables highly accurate searches.

The extraction unit may exclude from the candidate vehicles, based on the traveling direction, any second vehicle that is located in a position that the first vehicle cannot reach after the predetermined time has elapsed. This allows for more accurate extraction of candidate vehicles after the predetermined time has elapsed, taking into account the actual vehicle movements that may occur.

The extraction unit may acquire information about passable routes, and based on the acquired route information, exclude from the candidate vehicles any second vehicles that are located in a position that the first vehicle cannot reach after the predetermined time has elapsed. This allows candidate vehicles that were imaged at different times to be accurately matched in accordance with actual road traffic conditions.

The vehicle information includes first information, which is the color and the direction of travel, and second information other than the first information. The extraction unit may extract the first vehicle and the second vehicle based on the first information, and may exclude the second vehicle from the candidate vehicles if the second information of the second vehicle does not match the second information of the first vehicle. This can increase the accuracy of extracting candidate vehicles.

According to the present invention, it is possible to accurately extract the desired vehicle even when vehicle information is difficult to read from the captured image.

1 is a diagram showing an outline of electrical functional blocks of a vehicle search device 1 according to a first embodiment; FIG. 2 is a diagram schematically illustrating the data structure of a vehicle information table T1, which is an example of search data. 10 is a process flow for the vehicle search device 1 to search for a searched vehicle. 10 is a schematic diagram showing the process by which the vehicle search device 1 extracts searched vehicles, in which (a) shows the arrangement of vehicles located on a map M1 of a predetermined range including point F, (b) is a schematic diagram showing a state in which only black vehicles that match the search conditions have been extracted, (c) is a schematic diagram showing a state in which, of the vehicles extracted in (b), only vehicles included in the specified area M11 have been extracted, and (d) is a schematic diagram showing a state in which the vehicles excluded in step SP14 have been deleted from (c). 1 is a schematic diagram showing the process by which the vehicle search device 1 extracts searched vehicles, where (a) is a schematic diagram showing the arrangement of vehicles on map M1 at a second point in time after a predetermined time (e.g., 5 minutes) has elapsed since date and time E (first point in time), (b) is a diagram showing the range in which a first vehicle that was present in the first search target area at the first point in time may be present at the second point in time, and (c) is a schematic diagram showing the linking of the first vehicle and the second vehicle. These are examples of images displayed on the display unit, where (a) is a diagram showing an example of a map M1 drawn by connecting candidate vehicles, and (b) is a diagram showing the result of displaying the estimated movement route of vehicle C16.

Embodiments of a vehicle search device according to the present invention will be described in detail below with reference to the drawings. The vehicle search device extracts information about vehicles captured in captured images and searches for vehicles to be searched based on input search criteria. A vehicle to be searched may be, for example, a vehicle involved in a crime and fleeing the scene.

Figure 1 is a diagram showing an outline of the electrical functional blocks of a vehicle search device 1 according to an embodiment.

The vehicle search device 1 is connected to the imaging unit 100 via a network NW. The imaging unit 100 is, for example, a device installed in a traveling vehicle that captures images of the vehicle's surroundings, such as a drive recorder. The imaging unit 100 may be a camera installed near, on, or in the air above the road, such as a security camera, as long as its imaging range includes the road and traveling vehicles. The imaging unit 100 transmits imaging data to the network NW. The imaging data includes, for example, captured image data, identification information for the imaging unit 100, information on the time the image data was acquired (the image was captured) (hereinafter referred to as imaging date and time information), and information on the location where the image was captured (hereinafter referred to as location information). Note that the information included in the imaging data is merely an example, and the identification information for the imaging unit 100 is not required.

The images captured by the imaging unit 100 may be either moving images or still images. While one imaging unit 100 is shown in Figure 1, multiple imaging units 100 are typically connected to the network NW.

The vehicle search device 1 mainly comprises a data acquisition unit 10, an input unit 11, a display unit 12, and a control unit 20.

The data acquisition unit 10 is a functional unit that acquires imaging data from the imaging unit 100 via the network NW. The data acquisition unit 10 may also acquire map information via the network NW.

The input unit 11 is a functional unit that accepts search conditions for vehicles to be searched for in the vehicle search device 1. The input unit 11 may have a keyboard, mouse, or touch panel as hardware components. The input unit 11 may also be connected to an appropriate device via the network NW and receive information input to the appropriate device. The appropriate device may be, for example, a smartphone, tablet terminal, or personal computer.

The search criteria entered into the input unit 11 include information about the characteristics, date and time, and location of the vehicle to be searched. The location information may be, for example, the name of an intersection or an address. The location information may also be specified by inputting a point or area on a map displayed on an appropriate device connected via the network NW using the input means of the appropriate device. The date and time information may be, for example, the time of sighting or the time period. The search criteria may also include information such as the color and driving direction of the vehicle. The input unit 11 may also accept vehicle speed as a search criterion.

The control unit 20 includes a computing device such as a CPU (Central Processing Unit) for executing information processing, and storage devices such as RAM (Random Access Memory) and ROM (Read Only Memory), and primarily includes a reading unit 21, a memory unit 22, an extraction unit 23, and a map generation unit 24 as software resources.

The reading unit 21 is a functional unit that reads information about the vehicle (hereinafter referred to as vehicle information) from the image data acquired by the data acquisition unit 10. The vehicle information includes vehicle characteristics and information about the vehicle's traveling status at the time the image data was acquired (the image was captured). The information about the vehicle's traveling status also includes information about the vehicle's speed and direction of travel.

Vehicle information includes identification information, such as color and direction of travel, which is essential for vehicle search. Hereinafter, this will be referred to as first information.

Vehicle information also includes information not included in the identification information (information that is not essential for a search but improves search accuracy), such as the license plate, vehicle model, vehicle type and manufacturer (emblem), vehicle speed, steering wheel position, whether or not there are stickers, whether or not there are smoke windows, and information about the vehicle's usage status at the time of image capture (for example, information about the number of passengers). Hereinafter, this will be referred to as second information.

The vehicle information extracted by the reading unit 21 is stored in the memory unit 22 together with image capture date and time information and location information. The memory unit 22 may also store map information and various other information in addition to search data that associates vehicle information, image capture date and time information, and location information.

Figure 2 shows a schematic diagram of the data structure of vehicle information table T1, an example of search data. The data in vehicle information table T1 is broadly divided into required data and sub-data. Required data includes a data ID, which is a numeric value unique to each image data, and data required for vehicle searches. Data required for vehicle searches includes, for example, the date and time when the image was captured, location information indicating the location where the image was captured, and vehicle information (here, the direction of travel and color, which are the first information). Sub-data is data (second information) used to improve search accuracy, and includes, for example, vehicle type, manufacturer, vehicle name, vehicle speed, number of occupants, steering wheel position, and the presence or absence of smoke.

Note that information that must be obtained from image data may be considered essential data, and information that is stored only when it can be obtained based on the results of capturing the image data may be considered sub-data. The items of essential data and sub-data are merely examples.

The vehicle information table T1 also includes an associated data ID field that stores information associated through processing by the extraction unit 23, which will be described later. The information stored in the associated data ID field will be described in detail later. Note that the associated data ID field is not required.

Returning to the explanation of Figure 1, the extraction unit 23 is a functional unit that extracts vehicles that are estimated to be searched vehicles from the vehicle information table T1 stored in the memory unit 22 as candidate vehicles based on search conditions input from the input unit 11.

The extraction unit 23 extracts a vehicle (hereinafter referred to as the first vehicle) that matches the characteristics of the searched vehicle from a first search target area that includes a location included in the search criteria at the time of imaging included in the input search criteria (hereinafter referred to as the first time point). The extraction unit 23 may also set a point on a map received by the input unit 11 as the first search target area.

The extraction unit 23 also extracts a vehicle (hereinafter referred to as the second vehicle) that matches the characteristics of the searched vehicle from a second search target area in which the first vehicle is estimated to be present at a time point a predetermined time after the first time point or a time point a predetermined time before the first time point (hereinafter referred to as the second time point). The second search target area is an area different from the first search target area.

For example, the extraction unit 23 may set a predetermined range centered on a position included in the search criteria as the first search target area. The predetermined range may be stored in advance in the storage unit 22. Furthermore, for example, the extraction unit 23 may estimate the travel distance of the vehicle to be searched over a predetermined period of time, and set the estimated travel distance as the radius of the first search target area and the second search target area. In this way, by adjusting the search target area in consideration of the travel distance of the vehicle, accurate searches are possible.

The following methods can be considered for estimating the travel distance. For example, the extraction unit 23 may estimate the travel distance of the searched vehicle based on the vehicle speed read by the reading unit 21. Furthermore, for example, if the vehicle speed of the searched vehicle is input as a search condition from the input unit 11, the extraction unit 23 may estimate the travel distance using that vehicle speed. Furthermore, for example, the extraction unit 23 may estimate the travel distance based on the speed limit of the road at the location input as a search condition from the input unit 11.

The extraction unit 23 extracts the first and second vehicles based on the first information among the characteristics of the searched vehicle included in the search criteria input to the input unit 11. Therefore, even if only minimal information can be read from the image, it is possible to estimate the searched vehicle.

Next, we will explain how to set the second search target area. For example, the extraction unit 23 sets the second search target area to an area that is moved a predetermined distance from the first search target area. This predetermined distance can be stored in advance in the storage unit 22. This allows extraction of the second vehicle to be performed with a small computational load.

Also, for example, the extraction unit 23 estimates the travel distance of the searched vehicle over a predetermined time period, and moves the first search target area by the estimated travel distance to set the second search target area as the area obtained. This allows the second search target area to be set in an appropriate position, thereby improving the accuracy of extracting the second vehicle.

The extraction unit 23 associates a first vehicle extracted in the first search target area at a first time point with a second vehicle extracted in the second search target area at a second time point. The vehicle extracted in the second search target area is estimated to be the appearance of the vehicle extracted in the first search target area after a predetermined time has passed. Therefore, by associating the same vehicle captured in image data taken at different times, it is possible to organize information about candidate vehicles.

The extraction unit 23 associates the data ID corresponding to the first vehicle with the data ID corresponding to the second vehicle in the vehicle information table T1. Specifically, for example, the extraction unit 23 stores the data ID corresponding to the second vehicle (hereinafter referred to as the second data ID) in the associated data ID column of the vehicle information table T1 linked to the data ID corresponding to the first vehicle (hereinafter referred to as the first data ID) in the vehicle information table T1, and stores the first data ID in the associated data ID column of the vehicle information table T1 linked to the second data ID. The extraction unit 23 then estimates the associated first and second vehicles to be searched vehicles and extracts them as candidate vehicles.

Note that the extraction unit 23 may exclude the second vehicle from the candidate vehicles if the second information of the second vehicle does not match the second information of the first vehicle. For example, if multiple candidates for the second vehicle are extracted, the second vehicle is extracted based on the second information. This increases the accuracy of extracting the second vehicle. Note that, of course, if license plate information from the second information has been obtained, the extraction unit 23 may extract vehicle information with the same license plate as the same vehicle.

The extraction unit 23 may also acquire information about passable routes and, based on the acquired route information, exclude from candidate vehicles any second vehicles extracted from the second search target area that are located in a position that the first vehicle cannot reach after a predetermined time has elapsed. Information about passable routes is included in the map information.

For example, a vehicle on a ground road may be extracted in the first search target area, and a vehicle matching the search criteria may be extracted on an elevated road such as a highway in the second search target area. If the entrance to this elevated road is not in the first search target area or the second search target area, the vehicle extracted on the ground road at the first point in time will not be able to reach the elevated road at the second point in time. Therefore, the extraction unit 23 excludes vehicles on the elevated road from the candidate vehicles.

Furthermore, for example, if there is a river in the first or second search target area and there is no bridge across the river in the first or second search target area, it will be impossible to reach the opposite bank at the second time point. In other words, the extraction unit 23 excludes from candidate vehicles any vehicles extracted in the second search target area that are located on the opposite bank from the vehicles extracted in the first search target area. This configuration makes it possible to accurately extract candidate vehicles in accordance with actual road traffic conditions.

The extraction unit 23 may refer to information on passable routes, such as information on passability arising from traffic regulations, including not only rivers and highways, but also large buildings, private property, and one-way streets, and exclude from the candidate vehicles any second vehicle that is located in a position that the first vehicle cannot reach at the second time point. In addition to a configuration that refers to information on passable routes, the extraction unit 23 may also be configured to refer to impassable areas and exclude certain vehicles from the candidate vehicles.

The extraction unit 23 may also attempt to automatically generate a travel route from a first vehicle in the first search target area to a second vehicle in the second search target area, traveling through a passable area, and determine whether a route can be generated that allows the first vehicle to reach the position of the second vehicle within a predetermined time. If the route can be generated, the extraction unit 23 presumes that the vehicles are the same vehicle, associates them, and sets the second vehicle as a candidate vehicle.

The extraction unit 23 may also set the second search target area based on the direction of travel included in the vehicle information. This allows candidate vehicles after a predetermined time has elapsed to be extracted with high accuracy, taking into account the actual vehicle movement.

In addition, taking into account the possibility that the vehicle may make a U-turn, the second search target area may be determined by taking into account the time required to make a U-turn, rather than excluding vehicles based on their direction of travel.

The map generation unit 24 is a functional unit that arranges the first vehicle, second vehicle, and candidate vehicles extracted by the extraction unit 23 on a map based on their location information.

The map generation unit 24 outputs a map with vehicle icons arranged on the display unit 12. The display unit 12 may be provided on a smartphone, tablet terminal, or personal computer that inputs information to the input unit 11. The map generation unit 24 may also generate a list of candidate vehicles with icons displayed and display it on the display unit.

Note that in FIG. 1, the vehicle search device 1 has a display unit 12, but the form of the display unit is not limited to this. For example, the vehicle search device 1 may not have a display unit, and the display unit may be connected via the network NW.

The map generation unit 24 may display a virtual route connecting the candidate vehicles (first vehicle and second vehicle) extracted by the extraction unit 23 as the estimated travel route of the candidate vehicles. This configuration allows the travel route of the searched vehicle to be clearly grasped on the map. This processing will be described in detail later.

Here, the flow of processing by the vehicle search device 1 to search for a searched vehicle will be described with reference to FIGS.
Fig. 3 shows a processing flow in which the vehicle search device 1 searches for a searched vehicle. The processing shown in Fig. 3 is mainly performed by the control unit 20. Figs. 4 and 5 are schematic diagrams showing the process in which the vehicle search device 1 extracts a searched vehicle (the diagrams in Figs. 4 and 5 do not actually need to be displayed on the display unit 12). Fig. 6 is an example of an image displayed on the display unit. Each of the diagrams included in Figs. 4 to 6 will be explained later.

In the following explanation, we will use as an example a case where a searched vehicle is extracted when the search criteria "a black wagon traveling at a specific point F on date and time E (including information on the year, month, day, and time)" is acquired via the input unit 11.

(Step SP11)
The extraction unit 23 receives the search conditions for the searched vehicle via the input unit 11 and obtains the search results, and then the process proceeds to step SP12.

(Step SP12)
The extraction unit 23 extracts vehicles that match the search criteria for the searched vehicle by referring to the storage unit 22. For example, if the color of the vehicle included in the search criteria is "black," the extraction unit 23 extracts a black vehicle C10.

Figure 4(a) is a diagram that shows a schematic diagram of vehicles positioned on a map M1 of a predetermined range including point F. Map M1 includes point M10, which indicates point F. Figure 4(a) shows the vehicles included in vehicle information table T1 that are positioned on map M1 within a predetermined time range (hereinafter referred to as time range H) that includes a first time point (date and time E) included in the search criteria.

Map M1 shows a general road R11 and an elevated road R12, with multiple vehicles C10, C30, etc. shown as icons on the road. Vehicles C10 and C30 are included in vehicle information table T1 and can be located based on the location information in vehicle information table T1. For purposes of illustration, the figure shows multiple black vehicles C10 and multiple white vehicles C30, with no vehicles of other colors present.

Figure 4(b) is a diagram that shows a state in which only black vehicles that match the search criteria have been extracted. In Figure 4(b), the icon for white vehicle C30 has been deleted, and only black vehicle C10 has been extracted. Figure 4(b) also shows point M10, which indicates point F, and a line indicating the range of designated area M11, which is a predetermined range from point M10.

Of the black vehicles C10 that are candidates for the first vehicle, the extraction unit 23 extracts black vehicles C10 that are within a specified area M11, which is a predetermined range from a point M10 included in the search criteria, as vehicles that match the search criteria.

Figure 4(c) is a diagram that shows a state in which, of the vehicles extracted in Figure 4(b), only the vehicles included in the designated area M11 have been extracted. In Figure 4(c), only the black vehicles C11, C12, C13, C14, C15, and C16 within the designated area M11 are placed on the map M1. As a result, the black vehicle C10 is extracted from the designated area M11. Processing then proceeds to step SP13.

(Step SP13)
The map generation unit 24 arranges the candidates extracted in step SP12, i.e., vehicles that match the search conditions, on a map. That is, the map generation unit 24 generates the image shown in FIG. 4(c). Note that displaying the point M10 and the designated area M11 is not essential. Then, the processing proceeds to step SP14. Note that the map generation unit 24 may output the generated image to the display unit 12.

(Step SP14)
The extraction unit 23 determines whether or not any of the vehicles found in the search satisfy the exclusion criteria based on the map information. For example, in FIG. 4(c), it is estimated that vehicles C14 and C15 are traveling on elevated road R12 based on the location information, but point M10 is not an elevated road. Therefore, the extraction unit 23 determines that the vehicles traveling on elevated road R12 are not candidate vehicles, i.e., that there is a vehicle that satisfies the exclusion criteria.

If there is a vehicle that meets the exclusion conditions (Y in step SP14), proceed to step SP15. If there is no vehicle that meets the exclusion conditions (N in step SP14), proceed to step SP16.

(Step SP15)
The extraction unit 23 excludes vehicles that meet the exclusion conditions from the candidate vehicles, and the map generation unit 24 deletes the excluded vehicles from the map.

Figure 4(d) is a schematic diagram showing the state of Figure 4(c) after the vehicles excluded in step SP14 have been removed. In Figure 4(d), of the black vehicles C11 to C16 within the designated area M11, vehicles C14 and C15 have been removed.

The map generation unit 24 generates an image (i.e., the image shown in FIG. 4(d)) from the image generated in step SP13, with the vehicles excluded in step SP14 deleted. Note that displaying point M10 and designated area M11 is not required. The map generation unit 24 may also output the generated image to the display unit 12. Processing then proceeds to step SP16.

(Step SP16)
The extraction unit 23 determines whether or not image data prior to the time range (here, time range H) that includes the extraction time of the candidate vehicle has been processed. If the previous image data has been processed (Y in step SP16), the process proceeds to step SP17. If the previous image data has not been processed (N in step SP16), the process proceeds to step SP18.

(Step SP17)
If the previous image data has already been processed (N in step SP16), this means that this is not the first time that the processing of steps SP12 to SP15 has been performed. In this case, the extraction unit 23 links the vehicle that was not deleted in step SP15 with the vehicle that was not deleted in the previous step SP15. Step SP17 will be described in detail later.

(Step SP18)
If the previous image data has not been processed (N in step SP16), this means that the processes of steps SP12 to SP15 are being performed for the first time. In this case, the extraction unit 23 links the vehicles that were not deleted in step SP15 (vehicles C11, C12, C13, and C16 in the example shown in FIG. 4) with the center of the concentric circle indicating the designated area (point M10 in the example shown in FIG. 4). The center of the concentric circle represents the starting point of the vehicle's travel route extracted in steps SP13 to SP15. The extraction unit 23 then stores the linked information in the vehicle information table T1.

For example, if the result of performing steps SP12 to SP15 for the first time is as shown in Figure 4(c), the extraction unit 23 links vehicles C11 to C14 with the center of the designated area M11, i.e., point M10. The extraction unit 23 then stores the location information of point M10 in the associated data ID of the data having the data ID linked to vehicles C11 to C13 and C16.

(Step SP19)
The extraction unit 23 determines whether to continue the search based on appropriate information (such as search conditions input from the input unit 11). For example, the extraction unit 23 may receive input of a time period during which the search will be performed from the input unit 11, and determine to continue the search as long as the search does not deviate from that time period. Furthermore, for example, the extraction unit 23 may receive the number of times to perform the search from the input unit 11, and determine to continue the search until the number of times has elapsed.

If it is determined that the search should continue (Y in step SP18), proceed to step SP20. If the conditions for continuing the search are not met (N in step SP19), end the process.

(Step SP20)
The extraction unit 23 advances the time range of the imaging data to be searched by a predetermined time, and returns the process to step SP 12. The flow of the process of steps SP12 to SP17 that are performed after step SP20 will now be described.

Figure 5(a) is a schematic diagram showing vehicles arranged on map M1 in time range I (corresponding to the second time range of the present invention) a predetermined time (e.g., 5 minutes) after time range H (corresponding to the first time range of the present invention) including date and time E in Figure 4. In Figure 5(a), in step SP12, only black vehicles (candidate second vehicles) that meet the search criteria for the searched vehicle in time range I are extracted from the vehicles included in vehicle information table T1.

In Figure 5(a), for reference, point M10 and designated area M11 (corresponding to the first search target area) and vehicles C11, C12, C13, and C16 (first vehicle) extracted from designated area M11 at the first time point are shown by dashed lines (in reality, vehicles C11, C12, C13, and C16 do not exist at these positions at the second time point).

Furthermore, in step SP12, the extraction unit 23 extracts the second vehicle from among the black vehicles that are candidates for the second vehicle, from a second search target area in which the first vehicle is estimated to be located at the second time point. By setting the second search target area based on the direction of travel included in the vehicle information (see vehicle information table T1), candidate vehicles after a predetermined time has elapsed can be extracted with high accuracy, taking into account the actual vehicle movement.

Figure 5(b) is a diagram showing the range (second search target area) in which a first vehicle that was present in the first search target area (designated area M11) at the first time point may be present at the second time point. The designated areas N11, N12, N13, and N16 are shown, which are defined by estimating the distance traveled by each of the first vehicles C11, C12, C13, and C16 from the first time point to the second time point, and then moving the designated area M11 by the estimated distance traveled in the direction of each of the first vehicles C11, C12, C13, and C16 as viewed from point M10.

At the second point in time, there is a high probability that vehicles C11, C12, C13, and C16 have moved into the designated areas N11, N12, N13, and N16. Therefore, in step SP12, which is executed at the second point in time, the extraction unit 23 leaves only vehicles included in the designated areas N11, N12, N13, and N16 as the second vehicle.

In step SP13, the map generation unit 24 places the second vehicle extracted in step SP12 on the map. That is, the map generation unit 24 generates the image shown in FIG. 5(b). Note that it is not necessary to display point M10 and designated areas N11, N12, N13, and N16.

In step SP14, the extraction unit 23 determines whether any of the vehicles found in the search satisfy the exclusion conditions based on the map information. For example, in Figure 5(b), vehicles C31, C32, and C33 are estimated to be traveling on elevated road R12 based on their location information, but first vehicles C11, C12, C13, and C16 are not on the elevated road. Therefore, the extraction unit 23 determines that vehicles C31, C32, and C33 traveling on elevated road R12 are not candidate vehicles, i.e., they satisfy the exclusion conditions.

In step SP15, the extraction unit 23 excludes vehicles that meet the exclusion conditions (vehicles C31, C32, and C33) from the candidate vehicles, and the map generation unit 24 deletes the excluded vehicles from the map.

In step SP17, the extraction unit 23 links the vehicle (second vehicle) that was not deleted in step SP15 with the vehicle (first vehicle) that was not deleted in the previous step SP15. Step SP17 is described in detail below.

Figure 5(c) is a diagram that shows a schematic diagram of linking a first vehicle and a second vehicle. In Figure 5(c), vehicles C31, C32, and C33 have been deleted, and vehicles C21, C22, C23, and C26 are shown as second vehicles.

Because vehicles C22 and C23 are located in the designated area N11 for vehicle C11, the extraction unit 23 links vehicle C11 with vehicles C22 and C23. Because vehicle C21 is located in the designated area N12 for vehicle C12, the extraction unit 23 links vehicle C12 with vehicle C21. Because vehicle C21 is located in the designated area N13 for vehicle C13, the extraction unit 23 links vehicle C13 with vehicle C21. Because vehicles C22 and C26 are located in the designated area N16 for vehicle C16, the extraction unit 23 links vehicle C16 with vehicles C22 and C26.

The extraction unit 23 may also acquire information about passable routes and, based on the acquired route information, exclude from candidate vehicles any second vehicles that are located in a position that the first vehicle cannot reach after a predetermined time has elapsed. For example, vehicle C11 is linked to vehicles C22 and C23, but vehicle C11 and vehicle C22 are on a different road, so vehicle C22 is excluded from candidate vehicles and the link between vehicle C11 and vehicle C22 is released.

The extraction unit 23 may also exclude from candidate vehicles, based on the direction of travel, any second vehicle that is located in a position that the first vehicle cannot reach after a predetermined time has elapsed. For example, vehicle C11 and vehicle C23 are linked, but because vehicle C11 and vehicle C23 have different (opposite) directions of travel, vehicle C23 is excluded from candidate vehicles and the link between vehicle C11 and vehicle C23 is removed. Similarly, because vehicle C13 and vehicle C21 have different directions of travel, vehicle C21 is excluded from candidate vehicles and the link between vehicle C13 and vehicle C21 is removed.

Although vehicles C12 and C21 are traveling in different directions, by referencing the route information, it can be determined that vehicle C12 can reach the position of vehicle C21 if it turns right. Therefore, vehicle C21 is not excluded from the candidate vehicles.

As a result of the above, the extraction unit 23 links vehicle C16 with vehicles C22 and C26 and extracts them as candidate vehicles. Then, the extraction unit 23 associates vehicle C16 with vehicles C22 and C26 and stores them in the vehicle information table T1. Here, the extraction unit 23 stores the data IDs linked to vehicles C22 and C26 in the associated data ID of the data having the data ID linked to vehicle C16, and stores the data ID linked to vehicle C16 in the associated data ID of the data having the data ID linked to vehicles C22 and C26.

The candidate vehicles extracted in step SP18 are placed on a map by the map generation unit 24. The map generation unit 24 may also connect the candidate vehicles to each other and draw them on the map M1. Figure 6(a) is a diagram showing an example of a map M1 on which lines connecting the extracted candidate vehicles (first vehicle and second vehicle) are drawn. The map generation unit 24 outputs the map on which the candidate vehicles are placed to the display unit 12.

By repeatedly performing the above process, i.e., steps SP12 to S19, the movement route of the candidate vehicle can be estimated along the time axis.

The extraction unit 23 may further extract the positions of vehicles extracted at the first and second time points and whose vehicle information is associated with each other at a third time point, a predetermined time after the second time point. Using a similar method, the extraction unit 23 determines a third search target area based on the second search target area and extracts vehicles (third vehicles) that match the search criteria in the third search target area. The extraction unit 23 repeatedly performs this process to associate vehicles captured at different capture times.

Figure 6(b) is a diagram showing the result of displaying the estimated travel route of vehicle C16 based on the results of repeatedly executing steps SP12 to S19. By repeating steps SP12 to S19, the extraction unit 23 links vehicle C16 to vehicle C26, links vehicle C26 to vehicles C41, C43, and C46, links vehicle C41 to vehicles C42 and C45, and links vehicle C43 to vehicle C44. The map generation unit 24 displays a virtual route R21 connecting the linked vehicles C16, C26, C41, and C42 on map M1. Similarly, the map generation unit 24 displays on the map M1 a virtual route R22 connecting the linked vehicles C16, C26, C41, and C45, displays on the map M1 a virtual route R23 connecting the linked vehicles C16, C26, and C46, and displays on the map M1 a virtual route R24 connecting the linked vehicles C16, C26, C43, and C44. The virtual routes R21, R22, R23, and R24 are estimated travel routes along which vehicle C16 can be estimated to have traveled. This makes it possible to present the user with estimated travel routes of candidate vehicles, making it easier for the user to actually search the site.

In addition, when multiple estimated travel routes are estimated, the extraction unit 23 can narrow down the estimated travel routes based on the sub-data in the vehicle information table T1.

Alternatively, the user may input an instruction via the input unit 11 to perform processing again within the range of an extension of one of the virtual routes R21 to R24, causing the vehicle search device 1 to continue the search.

According to this embodiment, the desired vehicle can be extracted with high accuracy even when vehicle information is difficult to read from the captured image.

In this embodiment, in step SP19, the time range (second time range) including the time point at which the search target was captured was set a predetermined time later than the time range (first time range) of the immediately preceding processing, but the second time range may also be set a predetermined time earlier than the first time range.

In addition, in this embodiment, in step SP12, the first search target area is set by moving the first search target area by the estimated movement distance in the direction of the first vehicle as viewed from the center of the first search target area, but the method of setting the second search target area is not limited to this. For example, the extraction unit 23 may set the second search target area by moving the first search target area by the movement distance in the direction of travel of the first vehicle. Also, for example, the extraction unit 23 may set the second search target area by referring to map information and moving the first search target area along the road.

In addition, in this embodiment, the search information did not include the direction of travel. However, if the search information does include the direction of travel, some of the first and second vehicles may be excluded from the candidate vehicles based on the direction of travel. For example, in the example shown in FIG. 4, if the search information input is "a black wagon traveling uphill (upward in the figure) from a certain point F on date and time E (including information on the year, month, day, and time)," then in step SP15, vehicles C11, C12, and C13, which are not traveling uphill (upward in the figure), may be excluded from the candidate vehicles.

The above describes an embodiment of the present invention in detail with reference to the drawings, but the specific configuration is not limited to this embodiment and includes design modifications and the like that do not deviate from the gist of the present invention.

1: Vehicle search device 10: Data acquisition unit 11: Input unit 12: Display unit 20: Control unit 21: Reading unit 22: Storage unit 23: Extraction unit 24: Map generation unit 100: Imaging unit

Claims (8)

a data acquisition unit that acquires image data, imaging date and time information that is information about the time when the image data was acquired, and location information that is information about the location where the image data was acquired;
a reading unit that reads vehicle information from the image data, the vehicle information including the color and traveling direction of the vehicle;
a storage unit that stores search data that associates the vehicle information with the location information;
an input unit for receiving search conditions including information on the characteristics, date and time, and location of the vehicle to be searched;
an extraction unit that extracts, as candidate vehicles, vehicles that are estimated to be the searched vehicle from the search data based on the search conditions;
a map generation unit that arranges information indicating the candidate vehicles on a map based on the position information;
Equipped with
the extraction unit extracts a first vehicle that matches the characteristics of the searched vehicle from a first search target area that includes a position included in the search criteria, in a first time range that is a predetermined time range that includes a first time point included in the search criteria, and extracts a second vehicle that matches the characteristics of the searched vehicle from a second search target area in which the first vehicle is estimated to be present, in a second time range that is a predetermined time after the first time range or a predetermined time before the first time range, and extracts the first vehicle and the second vehicle as the candidate vehicles;
The map generation unit places a virtual route connecting the first vehicle and the second vehicle on the map.
A vehicle search device characterized by: The vehicle search device according to claim 1 , wherein the map generation unit displays the virtual route as an estimated travel route of the candidate vehicle. 3. The vehicle search device according to claim 1, wherein the extraction unit estimates a travel distance of the searched vehicle during the predetermined time period and sets the estimated travel distance as a radius of the first search target area and the second search target area. The vehicle search device according to any one of claims 1 to 3, characterized in that the extraction unit excludes from the candidate vehicles, among the second vehicles, vehicles that are located in a position that the first vehicle cannot reach after the predetermined time has elapsed, based on the traveling direction. The vehicle search device according to any one of claims 1 to 4, characterized in that the extraction unit acquires information on passable routes, and based on the acquired route information, excludes from the candidate vehicles any second vehicle that is located in a position that the first vehicle cannot reach after the predetermined time has elapsed. the vehicle information includes first information that is the color and the traveling direction, and second information other than the first information that includes a license plate, a vehicle model, a vehicle type, and a vehicle speed ;
6. The vehicle search device according to claim 1, wherein the extraction unit extracts the first vehicle and the second vehicle based on the first information, and excludes the second vehicle from the candidate vehicles if the second information of the second vehicle does not match the second information of the first vehicle. an acquisition step of acquiring image data, imaging date and time information which is information on the time when the image data was acquired, and location information which is information on the location where the image data was acquired;
a reading step of reading vehicle information, which is information about a vehicle, from the image data, including a color and a traveling direction of the vehicle;
a storage step of storing search data in which the vehicle information and the location information are associated with each other;
an input step of accepting search conditions including information on the characteristics, date and time, and location of the vehicle to be searched;
an extraction step of extracting, as a candidate vehicle, a vehicle that is estimated to be the searched vehicle from the search data based on the search conditions;
a generating step of arranging information indicating the candidate vehicles on a map based on the position information;
Equipped with
The extraction step extracts a first vehicle that matches the characteristics of the searched vehicle from a first search target area that includes a position included in the search criteria in a first time range that is a predetermined time range that includes a first time point included in the search criteria, extracts a second vehicle that matches the characteristics of the searched vehicle from a second search target area in which the first vehicle is estimated to be present in a second time range that is a predetermined time after the first time range or a predetermined time before the first time range, and extracts the first vehicle and the second vehicle as the candidate vehicles .
The generating step arranges a virtual route connecting the first vehicle and the second vehicle on the map.
A vehicle search method comprising: Computer,
a data acquisition unit that acquires image data, imaging date and time information that is information about the time when the image data was acquired, and location information that is information about the location where the image data was acquired;
a reading unit that reads vehicle information from the image data, the vehicle information including the color and traveling direction of the vehicle;
a storage unit that stores search data that associates the vehicle information with the location information;
an input unit for receiving search conditions including information on the characteristics, date and time, and location of the vehicle to be searched;
an extraction unit that extracts, as candidate vehicles, vehicles that are estimated to be the searched vehicle from the search data based on the search conditions;
a map generation unit that arranges information indicating the candidate vehicles on a map based on the position information;
It functions as
the extraction unit extracts a first vehicle that matches the characteristics of the searched vehicle from a first search target area that includes a position included in the search criteria, in a first time range that is a predetermined time range that includes a first time point included in the search criteria, and extracts a second vehicle that matches the characteristics of the searched vehicle from a second search target area in which the first vehicle is estimated to be present, in a second time range that is a predetermined time after the first time range or a predetermined time before the first time range, and extracts the first vehicle and the second vehicle as the candidate vehicles;
The map generation unit places a virtual route connecting the first vehicle and the second vehicle on the map.
A vehicle search program characterized by: