Server, vehicle detection system, and vehicle detection method

Abstract

The present disclosure relates to a server, a vehicle detection system, and a vehicle detection method. The server (10) includes a control unit (14). The control unit (14) determines whether a vehicle is used as a shelter based on a first image obtained by photographing a parked vehicle and its surroundings, which is included in an image obtained by photographing with a camera.

Description

Technical Field

[0001] The present invention relates to a server, a vehicle detection system, and a vehicle detection method for detecting a vehicle used for evacuation in a vehicle during a disaster. Background Art

[0002] In recent years, natural disasters such as earthquakes and typhoons have occurred frequently. The government, local governments, and citizen groups that support disaster victims hope to provide support suitable for disaster victims of natural disasters. Therefore, a system has been proposed that obtains earthquake damage information during an earthquake and reports a suitable evacuation route map to evacuation shelters for each area (see, for example, Patent Application Laid-Open No. 2003-162784). Summary of the Invention

[0003] However, due to privacy issues, concerns about infection, and issues with evacuation shelter rules such as not being able to bring pets, many people use vehicles such as private cars as evacuation shelters instead of using public evacuation shelters. It is desired to provide relief supplies and suitable information to evacuees. However, it is difficult for the government, local governments, and citizen groups to grasp evacuees taking shelter in vehicles.

[0004] An object of the present invention achieved in view of the above circumstances is to provide a server, a vehicle detection system, and a vehicle detection method capable of detecting a vehicle used for evacuation during a disaster.

[0005] The server of the present disclosure includes a control unit. The control unit determines whether the vehicle is used as an evacuation shelter based on a first image obtained by photographing a parked vehicle and its surroundings, which is included in an image obtained by photographing by the camera.

[0006] The vehicle detection system of the present disclosure includes a moving body having a camera and a server capable of communicating with the moving body. The server includes a control unit that determines whether the vehicle is used as an evacuation shelter based on a first image obtained by photographing a parked vehicle and its surroundings, which is included in an image obtained by photographing by the camera.

[0007] The vehicle detection method of the present disclosure includes a control unit determining whether the vehicle is used as an evacuation shelter based on a first image obtained by photographing a parked vehicle and its surroundings, which is included in an image obtained by photographing by a camera.

[0008] According to the present disclosure, it is possible to provide a server, a vehicle detection system, and a vehicle detection method capable of detecting a vehicle used for evacuation during a disaster. Brief Description of the Drawings

[0009] Hereinafter, the features, advantages, and technical and industrial significance of exemplary embodiments of the present invention will be described with reference to the accompanying drawings, in which the same reference numerals denote the same elements, and wherein:

[0010] Figure 1 is a block diagram showing a schematic configuration of a vehicle detection system according to one embodiment of the present disclosure.

[0011] Figure 2 is a diagram showing Figure 1 the schematic configuration of the server.

[0012] Figure 3 is a diagram showing Figure 1 the schematic configuration of the mobile body.

[0013] Figure 4 is a diagram showing an example of an area for searching for a vehicle.

[0014] Figure 5 is a diagram illustrating an example of vehicle detection using a mobile body.

[0015] Figure 6 is a diagram illustrating a method for detecting a vehicle using an image of the vehicle and its surroundings.

[0016] Figure 7 is a flowchart illustrating a method for detecting a vehicle according to one embodiment.

[0017] Figure 8 is a diagram illustrating Figure 7 the first determination process.

[0018] Figure 9 is a flowchart illustrating a method for detecting a vehicle according to another embodiment.

[0019] Figure 10 is a diagram illustrating Figure 9 the second determination process. Detailed Embodiment

[0020] Hereinafter, one embodiment of the present disclosure will be described with reference to the accompanying drawings. In addition, the drawings used in the following description are schematic drawings. The dimensional ratios and the like in the drawings are not necessarily the same as those in reality.

[0021] (Configuration of Vehicle Detection System)

[0022] A vehicle detection system 1 according to one embodiment of the present disclosure is as Figure 1As shown, it includes a server 10 and one or more mobile bodies 20. The vehicle detection system 1 can be operated by a service provider for vehicles. The vehicle detection system 1 can be operated by a public organization such as a local government. The server 10 and the mobile body 20 are configured to communicate via a network 30. In the network 30, various communication networks including a public network and a private network are included.

[0023] The server 10 is capable of setting a movement path for the mobile body 20. The mobile body 20 includes an unmanned aerial vehicle, a vehicle, a robot, etc. that can autonomously move according to the set movement path. The unmanned aerial vehicle includes a so-called drone. In the following description, it is assumed that the mobile body 20 is an unmanned aerial vehicle for explanation. The mobile body 20 is equipped with a camera and can capture images of the vehicle and its surroundings.

[0024] The server 10 is capable of acquiring the images captured by the mobile body 20. The server 10 can determine the vehicle used as a shelter in the event of a disaster based on the acquired images. When the vehicle detection system 1 is operated by a service provider, the server 10 can provide information on the vehicle used as a shelter to an external system 40 owned by a local government or the like. Hereinafter, each component of the vehicle detection system 1 will be described in more detail.

[0025] (Structure of the server)

[0026] The server 10 according to one embodiment is as Figure 2 shown and has a communication unit 11, a storage unit 12, a map information database 13, and a control unit 14.

[0027] The communication unit 11 includes a communication interface connected to the network 30 by wire or wirelessly. The communication unit 11 performs protocol processing related to the transmission and reception of information, modulation of transmission signals, and demodulation of reception signals, etc. The communication unit 11 can transmit and receive information with the mobile body 20 via the network 30. The communication unit 11 can be rewritten as a communicator.

[0028] The storage unit 12 stores information and programs required for the processing executed by the control unit 14. The storage unit 12 may include one or more of a semiconductor storage device, a magnetic storage device, and an optical storage device. Among semiconductor storage devices, there are DRAM (Dynamic Random Access Memory), SRAM (Static Random Access Memory), and flash memory, etc. Among magnetic storage devices, there are magnetic tapes, floppy disks (registered trademark), hard disks, etc. Among optical storage devices, there are, for example, CDs (Compact Discs), DVDs (Digital Versatile Discs), and Blu-ray (registered trademark), etc. In the storage unit 12, information on the vehicle detected by the moving body 20 is stored together with the evaluation value. The information on the vehicle includes the position information of the vehicle. The information on the vehicle may further include the identification information of the vehicle. The evaluation value indicates the degree to which it is presumed that the vehicle is used for taking shelter in the vehicle.

[0029] The map information database 13 is a database that stores map information of the entire area where each moving body 20 flies. The map information database 13 includes three-dimensional terrain information. The map information database 13 includes the position information of roads, parking lots, schools, public facilities, and designated shelters in the event of disasters, etc. The map information database 13 may be included in the storage unit 12.

[0030] The control unit 14 is configured to include a single or multiple processors. Among processors, there are general-purpose processors that execute programmed functions by reading in specific programs, and dedicated processors dedicated to specific processing. As dedicated processors, DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), and FPGA (Field-Programmable Gate Array), etc. may be adopted. The control unit 14 may include a memory capable of storing programs executed by the processor and information during the operation of the processor, etc.

[0031] The control unit 14 controls each part of the server 10. The control unit 14 can transmit and receive information with each moving body 20 via the communication unit 11. The control unit 14 can write information to and read information from the storage unit 12. The control unit 14 can read out map information from the map information database 13.

[0032] The control unit 14 includes functional blocks such as a path setting unit 14a, an image recognition unit 14b, and a vehicle information management unit 14c. Each functional block can be installed as a hardware module or a software module. The respective structural parts of the path setting unit 14a, the image recognition unit 14b, and the vehicle information management unit 14c can be combined and recombined. The processing performed by the respective structural parts of the path setting unit 14a, the image recognition unit 14b, and the vehicle information management unit 14c can be rewritten as processing performed by the control unit 14.

[0033] The path setting unit 14a sets the paths for each moving body 20 to move. When there are multiple moving bodies 20, the path setting unit 14a sets the moving paths in such a way that the area of the object is covered by the paths of the multiple moving bodies 20. The path setting unit 14a sends the moving paths to each moving body 20 via the communication unit 11. The path setting unit 14a can obtain map information associated with the moving paths from the map information database 13 and send it to the moving body 20 together with the moving paths. The moving body 20 that has received the moving path starts detecting the vehicle according to the moving path. The path setting unit 14a can send the moving paths to the moving body 20 regularly or irregularly multiple times a day to instruct the detection of the vehicle.

[0034] The image recognition unit 14b obtains the images (first images) captured by the moving body 20 via the communication unit 11. The images captured by the moving body 20 include images of the vehicle and its surroundings. The image recognition unit 14b performs recognition processing on the images captured by the moving body 20. Based on the results of the recognition processing, the image recognition unit 14b determines whether the vehicle is used as a shelter. Determining whether the vehicle is used as a shelter includes performing an operation to evaluate the possibility that the vehicle is used as a shelter. The image recognition unit 14b can generate an evaluation value representing the possibility that the vehicle is used as a shelter.

[0035] The evaluation value can be expressed in various ways. For example, the evaluation value can also be an example of representing the possibility that the vehicle is used as a shelter by quantifying it with points. In this case, it is presumed that the higher the number of points, the higher the possibility of being used as a shelter. Additionally, the evaluation value can also be an example of representing the possibility that the vehicle is used as a shelter by a ratio. The ratio can be estimated by accumulating information on actual disaster cases. Or, regarding the evaluation value, it can be represented by two values such as "0" and "1" for the case where there is a possibility that the vehicle is used as a shelter and the case where there is no possibility that the vehicle is used as a shelter.

[0036] Furthermore, the image recognition unit 14b can determine the position of the vehicle based on the position information of the moving body 20 received from the moving body 20, the position of the vehicle in the image, and the map information stored in the map information database 13.

[0037] The vehicle information management unit 14c manages information on parked vehicles detected by the mobile body 20. The vehicle information management unit 14c can associate the detected position of the vehicle and an evaluation value indicating the possibility that the vehicle is used as a shelter, and store them in the storage unit 12. The vehicle information management unit 14c can further store the identification information of the vehicle in the storage unit 12. The vehicle information management unit 14c can, as needed, append, change, read out, etc. the information on parked vehicles.

[0038] (Structure of the mobile body)

[0039] The mobile body 20 according to one embodiment is as Figure 3 shown, and includes a communication unit 21, a camera 22, a position detection unit 23, an orientation detection unit 24, a drive mechanism 25, a control unit 26, and a storage unit 27. As an example, the mobile body 20 is an unmanned aerial vehicle that receives an instruction of a flight path from the server 10 and autonomously flies at least partially.

[0040] The communication unit 21 includes a communication interface for wirelessly connecting to the network 30. The communication unit 21 can handle the 4th generation mobile communication system (4G) such as LTE (Long Term Evolution), the 5th generation mobile communication system (5G), and wireless communications such as Wi-Fi (registered trademark). The communication unit 21 performs protocol processing related to the transmission and reception of information, modulation of transmission signals, demodulation of reception signals, etc. The communication unit 21 can transmit and receive information with the server 10 via the network 30. The communication unit 21 can be rewritten as a wireless communicator.

[0041] The camera 22 includes an optical system such as a lens, and a CCD image sensor (Charge-Coupled Device Image Sensor) or a CMOS image sensor (Complementary MOS Image Sensor) and other imaging elements. The camera 22 can further perform demosaicing processing, noise reduction processing, white balance adjustment, exposure adjustment, and distortion correction, etc. on the image signal output from the imaging element. The camera 22 can include an ISP (Image Signal Processor) for performing these signal processes.

[0042] The camera 22 mounted on the mobile body 20 can capture images of the surroundings of the mobile body 20. When the mobile body 20 is an unmanned aerial vehicle, the camera 22 can be arranged on the lower side of the mobile body 20 in order to capture images of the subject on the ground surface. The camera 22 can be set as a camera with a wide field of view. For example, the camera 22 can be set as a camera capable of performing 360-degree imaging of the lower hemisphere. The camera 22 can continuously capture images at a predetermined frame rate, such as 30 fps (frame per second). The camera 22 can output the signal of the captured image to the control unit 26.

[0043] The mobile body 20 can have a plurality of cameras 22. The ranges captured by the plurality of cameras 22 can be different from each other. One camera included in the plurality of cameras can be a camera capable of performing 360-degree imaging of the lower hemisphere. Other cameras included in the plurality of cameras can be cameras for photographing a specific direction. The plurality of cameras 22 can include an infrared camera.

[0044] The position detection unit 23 is a sensor that detects the current position of the mobile body 20. The position detection unit 23 can detect the absolute position represented by latitude, longitude, etc. The position detection unit 23 can include a receiver corresponding to the Global Navigation Satellite System (GNSS). In the receiver corresponding to GNSS, a Global Positioning System (GPS) receiver is included. The position detection unit 23 can be rewritten as a positioning sensor.

[0045] The azimuth detection unit 24 is a sensor that detects the azimuth. The azimuth detection unit 24 can, for example, detect the magnetic force of the geomagnetism to measure the azimuth. The azimuth detection unit 24 can be rewritten as an azimuth sensor.

[0046] The drive mechanism 25 is a mechanism for moving the mobile body 20. When the mobile body 20 is an unmanned aerial vehicle, the drive mechanism 25 includes a plurality of rotary wings and their drive devices. The number of rotary wings is, for example, 4, 6, or 8, but is not limited to these. As an example, the plurality of rotary wings are radially arranged in a horizontal plane from the center of the body of the mobile body 20. The drive mechanism 25 can make the mobile body 20 perform various actions such as stationary, ascending, descending, advancing, retreating, and turning around by adjusting the rotational speed of each rotary wing under the control of the flight control unit 26c of the control unit 26.

[0047] The control unit 26 is configured to include a single or multiple processors. Similar to the control unit 14 of the server 10, the control unit 26 is composed of various processors and memories. The control unit 26 controls each part and the whole of the mobile body 20. For example, the control unit 26 may include functional blocks such as a camera control unit 26a, a path control unit 26b, and a flight control unit 26c. Each structural part of the camera control unit 26a, the path control unit 26b, and the flight control unit 26c can be implemented as a hardware module or a software module. Each structural part of the camera control unit 26a, the path control unit 26b, and the flight control unit 26c can be combined and recombined. The processing performed by each structural part of the path control unit 26b and the flight control unit 26c can be rewritten as the processing performed by the control unit 26.

[0048] The camera control unit 26a controls the camera 22. The camera control unit 26a transmits the image obtained from the camera 22 to the server 10 via the communication unit 21. The camera control unit 26a can transmit the position information detected by the position detection unit 23 to the server 10 together with the image. The camera control unit 26a can receive an instruction from the server 10 via the communication unit 21 and photograph the identified vehicle from a predetermined distance and direction. Therefore, the camera control unit 26a can cooperate with the path control unit 26b and the flight control unit 26c.

[0049] The path control unit 26b obtains the movement path and the map information associated with the movement path from the server 10 via the communication unit 21. The path control unit 26b can store the movement path and the map information in the storage unit 27. The path control unit 26b moves the mobile body 20 according to the set movement path. When a vehicle is found around the movement path, the path control unit 26b can control the drive mechanism 25 to move the drive mechanism 25 relative to the vehicle to a predetermined distance and direction in order to determine whether the vehicle is used as a shelter.

[0050] The flight control unit 26c controls the drive mechanism 25 of the mobile body 20 and autonomously maintains the flight state. For example, the flight control unit 26c maintains the distance from the ground at a predetermined distance. When the position of the mobile body 20 deviates from the path due to external reasons such as wind, the flight control unit 26c controls the drive mechanism 25 to return to the path. Furthermore, when an unexpected obstacle is detected, the flight control unit 26c can control the drive mechanism 25 to avoid it. In order to be used for control by the flight control unit 26c, the mobile body 20 can have various sensors in addition to the position detection unit 23 and the azimuth detection unit 24. For example, the mobile body 20 can have an acceleration sensor, an angular velocity sensor, a ground height sensor, an obstacle sensor, etc.

[0051] The storage unit 27 stores data, programs, etc. required for the processing performed by the control unit 26. For example, the storage unit 27 stores the movement path and map information received from the server 10. Similar to the storage unit 12 of the server 10, the storage unit 27 may include one or more of a semiconductor storage device, a magnetic storage device, and an optical storage device.

[0052] (An example of an application scenario)

[0053] Next, with reference to Figures 4 to 6 , an operation example of the vehicle detection system 1 will be described. Figure 4 A simplified map 50 showing an example of an area for searching for vehicles used as evacuation shelters is shown. In the map 50, roads 51, parking lots 52a to 52c, and residential areas 53a and 53b of residents are shown. The parking lots 52a to 52c include parking lots of public facilities such as city government agencies, large facilities such as shopping centers, and parking lots of designated evacuation shelters. The parking lots 52a to 52c include open spaces that can be used as parking lots. The residential areas 53a and 53b include urban areas and residential areas where a large number of residents live.

[0054] In Figure 4 , the mobile body 20 is arranged at the mobile body base 55. The server 10 can be located at any place including the mobile body base 55. For each mobile body 20, the server 10 sets a movement path 56 that returns to the mobile body base 55 via any one or more of the parking lots 52a to 52c and the residential areas 53a and 53b. When the mobile body 20 is an unmanned aerial vehicle, the mobile body 20 flies in the air. The mobile body 20 can fly at a height of, for example, 3 m to 150 m from the ground surface, but is not limited thereto. In Figure 4 's example, a movement path passing through the residential area 53a including the parking lot 52a is assigned to one mobile body 20. In addition, the parking lots 52b, 52c, and the residential area 53b are covered by the movement paths of other mobile bodies 20.

[0055] As Figure 5 shown, the mobile body 20 can capture images of the vehicles 61 located on the ground surface through the camera 22 facing the ground surface side. The mobile body 20 can capture images of a plurality of vehicles 61 simultaneously. The images captured by the mobile body 20 are sent to the server 10 via the network 30. The control unit 14 of the server 10 detects the vehicles 61 from the received images. The detection of the vehicles 61 can be performed by image processing techniques such as template matching.

[0056] The control unit 14 of the server 10 as Figure 6As shown, recognition processing is performed on an image (first image) obtained by photographing the vehicle 61 and the surrounding area 62 of the vehicle 61 in the image photographed by the camera 22. The surrounding area 62 of the vehicle 61 can be set, for example, to a range within about 3 m from the vehicle 61, but is not limited thereto. The control unit 14 detects an object 63 located in the surrounding area 62 of the vehicle 61. The control unit 14 analyzes whether a predetermined object 63 is included in the surrounding area 62. The predetermined object 63 includes, for example, a tent, a chair, a table, clothing (clothing to be washed), and a floor mat. In Figure 6 the example of, the object 63 is a chair placed in the shape of a floor mat. When the vehicle 61 is used as a shelter, these objects 63 are often set around the vehicle 61. Therefore, when the control unit 14 determines that these objects 63 are included in the image, it can be evaluated that the possibility that the vehicle 61 is used as a shelter is higher than when these objects 63 are not included.

[0057] The control unit 14 of the server 10 can evaluate whether the vehicle 61 is used as a shelter based on an image of the vehicle 61 itself. For example, the control unit 14 can recognize the states of the doors and windows of the vehicle 61 from the image photographed by the camera 22. For example, when the control unit 14 recognizes that the window or door of the vehicle 61 is open, it can be evaluated that the possibility that the vehicle 61 is used as a shelter is higher than when it is recognized that the window or door of the vehicle 61 is not open.

[0058] When the control unit 14 can recognize that there is a person in the vehicle 61 in the image photographed by the camera 22, it can be evaluated that the possibility that the vehicle 61 is used as a shelter is higher than when no person is recognized. For example, the control unit 14 can detect the head of a person in the vehicle 61 by template matching. Alternatively, when there is a moving object in the vehicle 61, the control unit 14 can determine the object as a person.

[0059] When the control unit 14 can recognize in the image that a component that obstructs visual recognition of the internal configuration from the outside of the vehicle 61 is disposed on the window of the vehicle 61, it can be evaluated that the possibility that the vehicle 61 is used as a shelter is higher than when the component is not recognized. Components that obstruct visual recognition from the outside include, for example, a mask and curtains for light shielding. Such components are sometimes used for privacy protection when there is a person in the vehicle 61.

[0060] The control unit 14 of the server 10 can output an instruction to the mobile body 20 in such a way that, in order to obtain desired information from an image of the vehicle 61, the mobile body 20 moves relative to the vehicle 61 in a distance and / or direction within a predetermined range. The control unit 14 can determine the orientation of the vehicle 61 based on an image obtained by photographing the vehicle 61 from above. For example, the control unit 14 can instruct the mobile body 20 to photograph an image of the vehicle 61 from the front of the vehicle 61. In this case, the camera 22 can, for example, capture an image of the front of the vehicle 61 including the license plate. Thus, the server 10 can obtain the information of the license plate as the identification information of the vehicle 61. In addition, as the identification information, it is also possible to use information on the appearance such as the size and color of the vehicle 61 identified from the image.

[0061] In the case where the camera 22 includes a far-infrared camera, the control unit 14 of the server 10 can measure the temperature information of the object based on the image captured by the camera 22. For example, when a person uses cooling or heating in the vehicle interior of the vehicle 61 for shelter, the temperature of the window portion of the vehicle 61 may be different from the surroundings. In addition, when the engine of the parked vehicle 61 operates for cooling or heating, the temperature of a part of the engine close to the vehicle 61 is higher than the surroundings. The control unit 14 can detect the temperature distribution based on the image of the far-infrared camera, and when it is recognized that at least a part of the vehicle 61 has a different temperature from the surroundings, it is evaluated that the possibility of the vehicle 61 being used as a shelter is higher than in the same situation.

[0062] The control unit 14 of the server 10 determines whether the vehicle 61 is used as a shelter in consideration of the above various conditions. The control unit 14 can evaluate the possibility that the vehicle 61 is used as a shelter. The control unit 14 can quantify the possibility that the vehicle 61 is used as a shelter according to the above various conditions and output it as an evaluation value. When the evaluation value is higher than a predetermined value, the control unit 14 can determine that the possibility that the vehicle 61 is used as a shelter is high.

[0063] The control unit 14 can store the parking position of the vehicle 61, the vehicle identification information, and the evaluation value in the storage unit 12 together. The control unit 14 can, as needed, send the position information and identification information of the vehicle 61 determined to have a high possibility of being used as a shelter to the external system 40.

[0064] (Vehicle Detection Method 1)

[0065] Hereinafter, with reference to Figure 7 , an example of the vehicle detection method of the present disclosure executed by the control unit 14 of the server 10 will be described.

[0066] First, the control unit 14 sets the movement path of the mobile body 20 via the network 30 (step S101).

[0067] The moving body 20 with a set moving path moves in accordance with the moving path while sequentially transmitting the captured images to the server 10. The control unit 14 searches for the vehicle 61 from the received images (step S102).

[0068] When the control unit 14 identifies the vehicle 61 based on the images received from the moving body 20 (step S103: "Yes"), it proceeds to the next step S104. When the control unit 14 does not identify the vehicle 61 in the images received from the moving body 20 (step S103: "No"), it proceeds to step S107.

[0069] The control unit 14 determines the position of the identified vehicle 61 based on the images captured by the moving body 20, the position information of the moving body 20 received from the moving body 20, and the map information (step S104). The control unit 14 can determine the position of the vehicle 61 by correlating the roads 51 and parking lots 52, etc. on the image with these positions on the map information. When the moving body 20 is flying directly above the vehicle 61 or flying near the vehicle 61, the control unit 14 can use the position information of the moving body 20 as the position information of the vehicle 61.

[0070] The control unit 14 performs a first determination process of determining the possibility that the vehicle 61 is used as a shelter based on the images obtained from the moving body 20 (step S105). Hereinafter, according to Figure 8 an example of the first determination process will be described.

[0071] The control unit 14 identifies a predetermined item 63 located in the peripheral area 62 from the images obtained by capturing the vehicle 61 and the peripheral area 62 of the vehicle 61 (step S201). The predetermined item 63 is an item indicating the possibility that the vehicle 61 is used as a shelter as described above.

[0072] Next, the control unit 14 identifies the people inside the vehicle 61 based on the images obtained from the moving body 20 (step S202). When there are people inside the vehicle 61 despite it being parked, there is a possibility that the vehicle 61 is used as a shelter. In addition, when there are people in the peripheral area 62 of the vehicle 61 rather than inside the vehicle 61, the control unit 14 can also consider the possibility that the vehicle 61 is used as a shelter as positive information.

[0073] Furthermore, the control unit 14 identifies the state of the vehicle 61 based on the images obtained from the moving body 20 (step S203). The state of the vehicle 61 may include at least any one of the opening and closing states of the windows and doors, the configuration of components such as covers that obstruct visual recognition of the interior from the outside of the vehicle, and the temperature distribution of the vehicle 61.

[0074] To execute steps S201 to S203, the control unit 14 may instruct the mobile body 20 to perform imaging from a specific distance and direction. The control unit 14 may not execute all of steps S201 to S203. The control unit 14 may execute any one or more of steps S201 to S203.

[0075] After executing any one or more of steps S201 to S203, the control unit 14 calculates an evaluation value indicating the possibility that the vehicle 61 is used as a shelter (step S204). The control unit 14 may quantify the identification results of steps S201 to S203 and perform a predetermined operation to calculate the evaluation value. Alternatively, the control unit 14 may determine that the possibility that the vehicle 61 is used as a shelter is high or low based on the identification results of steps S201 to S203. In this case, for example, the evaluation value in the case of a high possibility may be set to "1", and the evaluation value in the case of a low possibility may be set to "0".

[0076] The control unit 14 may consider the position of the vehicle 61 in the calculation of the evaluation value. For example, when the vehicle 61 is located in the parking lot 52 of the designated shelter during a disaster or within a predetermined distance range from the designated shelter, the control unit 14 may evaluate the possibility that the vehicle 61 is used as a shelter to be higher. In the vicinity of the designated shelter, some disaster victims who do not stay at the shelter due to reasons such as shelter rules and privacy but hope to receive shelter support may park the vehicle 61 and take shelter in the vehicle.

[0077] After step S204, the control unit 14 proceeds to Figure 7 step S106. The control unit 14 stores the position information and the evaluation value of the identified vehicle 61 in the storage unit 12. The control unit 14 may also store the identification information.

[0078] The control unit 14 repeats the processes of steps S102 to S106 until the mobile body 20 moves on all the movement paths 56 and the control unit 14 of the server 10 searches for the vehicles 61 on all the movement paths 56 (step S107: "No"). After the mobile body 20 moves on all the movement paths 56 and the control unit 14 finishes searching for the vehicles 61 on all the movement paths 56 (step S107: "Yes"), the detection process of the vehicle 61 ends.

[0079] The control unit 14 of the server 10 can, at an appropriate timing, extract vehicles 61 with evaluation values higher than a predetermined value from the vehicle information stored in the storage unit 12, and send the identification information of these vehicles together with the position information of the vehicle 61 to the external system 40. Alternatively, the server 10 can output this information to an output device such as a display or a printer. An autonomous body or a support group that receives the information can support these vehicles 61 on the assumption that they may be used as shelters by disaster victims.

[0080] (Vehicle detection method 2)

[0081] The server 10 can periodically or irregularly cause the mobile body 20 to repeatedly detect the vehicle 61 used as a shelter. In this case, the server 10 can obtain an image of the same location as the location where the vehicle 61 was detected in the past from the mobile body 20. Depending on whether the same vehicle 61 continues to park at the location where the vehicle 61 parked in the past, the determination of whether the vehicle 61 is used as a shelter may change.

[0082] Figure 9 It is a flowchart showing an example of the processing after the second time in the case of repeatedly performing the detection process of the vehicle 61. Figure 9 Perform determination on the vehicle 61 detected in the past. During the flight of the mobile body 20 on the movement path, it can be Figure 7 executed in parallel with the Figure 9 processing flow.

[0083] First, the control unit 14 of the server 10 sets the movement path of the mobile body 20 (step S301).

[0084] The control unit 14 reads out the vehicle information of the vehicle 61 identified in the past near the movement path from the storage unit 12 (step S302). The vehicle information includes the position information, identification information, and evaluation value of the vehicle 61.

[0085] The control unit 14 causes the mobile body 20 to capture an image (second image) of the position according to the position information of the vehicle 61 read out from the storage unit 12. Thus, the control unit 14 obtains an image of the position where the vehicle 61 that parked in the past was detected from the mobile body 20 at a time different from the time when the vehicle 61 was captured in the past (step S303).

[0086] The control unit 14 performs a second determination process based on the image obtained in step S303 (step S304). Hereinafter, according to Figure 10 , the second determination process will be described.

[0087] First, the control unit 14 determines whether there is a vehicle 61 at the position where the vehicle 61 was recognized as parked in the past based on the image acquired from the moving body 20 (step S401).

[0088] When there is a vehicle 61 at the position where the vehicle 61 was recognized in the past (step S401: "Yes"), the control unit 14 uses the recognition information to determine whether the currently photographed vehicle 61 is the same vehicle as the vehicle 61 stored in the storage unit 12 (step S402). The control unit 14 can also perform the following processing: pre-store the image of the vehicle 61 detected in the past in the storage unit 12, and determine the identity of the vehicle by comparing the image of the vehicle 61 obtained by image photography and the image of the vehicle 61 photographed in the past. The control unit 14 can perform the Figure 8 first determination process and newly calculate the evaluation value (step S403).

[0089] When there is no vehicle 61 at the position where the vehicle 61 was recognized in the past (step S401: "No"), the control unit 14 determines whether there is a setting object at the position where the vehicle 61 parked in the past (step S404). The setting object is, for example, a road sign, a pet water bottle filled with water, or an item used in the place such as a floor mat. When there is such a setting object, the vehicle 61 that parked in the past returns to the same place more often. For example, the victims of a disaster sometimes use the vehicle 61 as a shelter only at night and park at the same place.

[0090] The control unit 14 calculates the evaluation value based on the results of steps S401 to S404 (step S405).

[0091] For example, when the control unit 14 determines that there is the same vehicle 61 at the position where the vehicle 61 was recognized in the past in step S402, it can use the higher evaluation value among the evaluation value calculated in the past and the evaluation value calculated in step S403 as the new evaluation value. Furthermore, the control unit 14 can determine that the possibility of the vehicle 61 being used as a shelter is higher and add the evaluation value. For example, when the control unit 14 determines that there is no same vehicle 61 at the position where the vehicle 61 was recognized in the past and a different vehicle 61 is parked, it can delete the vehicle information of the original vehicle 61. In this case, for the newly recognized vehicle 61, the evaluation value calculated in step S403 is assigned.

[0092] For example, when there is no fixture at the parking position in step S404, the control unit 14 may maintain or decrease the evaluation value of the vehicle 61. In this case, since the vehicle 61 may be temporarily used for movement, the vehicle information of the vehicle 61 may not be deleted. In addition, when there is a fixture at the parking position in step S404, the vehicle 61 that parked at this location in the past is highly likely to return to the same location. Therefore, compared to the case where there is no fixture, the control unit 14 may evaluate the possibility that the vehicle 61 is used as a shelter to be higher.

[0093] After step S404, the control unit 14 proceeds to Figure 9 step S305. The control unit 14 updates the vehicle information stored in the storage unit 12 regarding the identified vehicle 61.

[0094] The control unit 14 repeats the processes of steps S303 to S305 until the mobile body 20 moves on all the movement paths and searches for all the vehicles 61 read from the vehicle information (step S306: "No"). After the mobile body 20 moves on all the movement paths and the control unit 14 finishes searching for all the vehicles 61 (step S306: "Yes"), the vehicle detection process ends.

[0095] Thus, the control unit 14 can evaluate that the vehicle 61 that has parked at the same parking position for a long time and the vehicle 61 that has temporarily left the parking position and returned to the same position again are more likely to be used as shelters.

[0096] As described above, according to the present embodiment, the server 10 can determine the vehicle 61 that is highly likely to be used as a shelter based on the image captured by the mobile body 20. The method of the present disclosure can easily determine the vehicle 61 used as a shelter without going through manual work by using image processing. Thus, the server 10 can provide the position and identification information of the vehicle 61 that is highly likely to be used as a shelter to the government, local autonomous bodies, groups that support disaster victims, etc. Therefore, the government, local autonomous bodies, groups that support disaster victims, etc. can appropriately provide support materials and information for the vehicle 61.

[0097] In addition, the present invention is not limited to the above-described embodiment, and a large number of modifications or changes can be implemented. For example, the functions included in each unit, each step, etc. can be reconfigured logically without contradiction, and multiple units or steps, etc. can be combined into one or divided.

[0098] For example, in the above-described embodiment, the mobile body 20 transmits the image obtained by imaging with the camera 22 to the server 10, and the server 10 performs image recognition processing. However, functions can be shared in various ways between the control unit 26 of the mobile body 20 and the control unit 14 of the server 10. For example, the control unit 26 of the mobile body 20 may recognize the vehicle 61 from the image obtained by imaging with the camera 22 and sequentially transmit only the recognized vehicle 61 and the image around it to the server 10.

[0099] In addition, in the above-described embodiment, the map information database 13 is included in the server 10, but the map information database 13 may be provided independently of the server 10. The mobile body 20 may be configured to obtain map information from the map information database 13 provided independently of the server 10 when receiving path information from the server 10. In addition, the mobile body 20 itself may previously have necessary map information.

[0100] In addition, in the above-described embodiment, the mobile body 20 is described on the premise of being an unmanned aerial vehicle. However, the mobile body 20 can also use an autonomous vehicle. In this case, the mobile body 20 does not move in the air, but searches for the vehicle 61 used as a shelter while moving on the road 51 and in the parking lot 52. Furthermore, in the above-described embodiment, the camera 22 is mounted on the mobile body 20. However, the camera of the present disclosure is not only the camera mounted on the mobile body, but can also include surveillance cameras installed in the street.

[0101] The method for detecting the vehicle 61 disclosed in this specification can be executed by a processor included in the server 10 and the mobile body 20 in accordance with a program. Such a program can be stored in a non-temporary computer-readable medium. Examples of the non-temporary computer-readable medium include a hard disk, RAM, ROM, flash memory, CD-ROM, optical storage device, magnetic storage device, etc., but are not limited to these.

Claims

1. A server having a control unit, The control unit calculates an evaluation value of the vehicle being used as a shelter based on one or more judgment conditions in a first image obtained by a camera photographing a parked vehicle and its surroundings. Among them, In the first image, there is an image of a far-infrared camera capable of measuring the temperature of an object. When the control unit identifies that at least a part of the vehicle has a different temperature from its surroundings in the image of the far-infrared camera, it evaluates that the possibility of the vehicle being used as a shelter is higher than when it does not identify that at least a part of the vehicle has a different temperature from its surroundings. The control unit also obtains a second image, which is obtained by the camera photographing at a position where the parked vehicle is photographed at a time different from the time when the first image of the parked vehicle is photographed. The control unit determines whether the second image contains the vehicle. When it does not contain the vehicle, the control unit determines whether the second image contains a fixture. When it does not contain the fixture, the control unit maintains or reduces the evaluation value. When it contains the fixture, it evaluates that the possibility of the vehicle being used as a shelter is higher than when the second image does not include the vehicle and does not include the fixture. The control unit evaluates the possibility of the vehicle being used as the shelter based on the evaluation value obtained from the second image.

2. The server according to claim 1, wherein One or more judgment conditions in the first image include whether a predetermined item is included in the first image. When the control unit includes an image of a predetermined item in the first image, it evaluates that the possibility of the vehicle being used as a shelter is higher than when the item is not included.

3. The server according to claim 2, wherein The predetermined item includes at least any one of a tent, a chair, a table, clothing, and a floor mat.

4. The server according to claim 1, wherein One or more judgment conditions in the first image include whether a person inside the vehicle is identified in the first image. When the control unit identifies a person inside the vehicle in the first image, it evaluates that the possibility of the vehicle being used as a shelter is higher than when no person is identified.

5. The server according to claim 1, wherein One or more judgment conditions in the first image include whether the window or door of the vehicle is identified as being open in the first image. When the control unit identifies that the window or door of the vehicle is open in the first image, it evaluates that the possibility of the vehicle being used as a shelter is higher than when it does not identify that the window or door of the vehicle is open.

6. The server according to claim 1, wherein One or more determination conditions in the first image include whether a component that obstructs visual recognition of the interior from the outside of the vehicle is disposed on a window of the vehicle. When the control unit recognizes in the first image that a component that obstructs visual recognition of the interior from the outside of the vehicle is disposed on the window of the vehicle, it evaluates that the possibility of the vehicle being used as a shelter is higher than in the case where the component is not recognized.

7. The server according to claim 1, wherein The camera is mounted on a moving body. Before capturing the first image, the camera captures an image based on the movement of the moving body. When the control unit recognizes in the image that the vehicle includes the parked vehicle, the control unit moves the moving body to capture the first image at least either at a distance within a predetermined range or in a direction within a predetermined range based on the orientation of the vehicle.

8. The server according to claim 1, wherein The camera is mounted on a moving body. The control unit is configured to: be able to recognize the identification information of the vehicle from the first image, obtain the position information of the vehicle from the moving body, and when it determines that the evaluation value indicating the possibility of the vehicle represented by the first image being used as a shelter is higher than a predetermined value, send the identification information and the position information of the vehicle to the outside.

9. A vehicle detection system, comprising: A moving body having a camera; and A server capable of communicating with the moving body, The server includes a control unit that calculates an evaluation value of the vehicle being used as a shelter based on one or more determination conditions in a first image obtained by the camera capturing the parked vehicle and its surroundings. Among them, In the first image, an image of a far-infrared camera capable of measuring the temperature of an object is included. When the control unit recognizes from the image of the far-infrared camera that at least a part of the vehicle has a different temperature from its surroundings, it evaluates that the possibility of the vehicle being used as a shelter is higher than in the case where it is not recognized that at least a part of the vehicle has a different temperature from its surroundings. The moving body captures a second image at a position where the parked vehicle is imaged at a time different from the time when the first image of the parked vehicle is captured. The control unit of the server obtains the second image. The control unit determines whether the second image includes the vehicle. When the vehicle is not included, the control unit determines whether the second image includes a structure. When the structure is not included, the control unit maintains or reduces the evaluation value. When the structure is included, it evaluates that the possibility of the vehicle being used as a shelter is higher than in the case where the vehicle and the structure are not included in the second image. The control unit evaluates the possibility of the vehicle being used as the shelter based on the evaluation value obtained from the second image.

10. The vehicle detection system according to claim 9, wherein One or more determination conditions in the first image include whether an image of a predetermined article is included in the first image. When the image of the predetermined article is included in the first image, the control unit of the server evaluates that the possibility of the vehicle being used as a shelter is higher than the case where the article is not included.

11. The vehicle detection system according to claim 10, wherein, The predetermined article includes at least any one of a tent, a chair, a table, clothing, and a floor mat.

12. The vehicle detection system according to claim 9, wherein, The moving body transmits the position information of the vehicle to the server, The control unit of the server is configured to: be able to identify the identification information of the vehicle from the first image, obtain the position information of the vehicle from the moving body, and when it is determined that the evaluation value indicating the possibility of the vehicle represented by the first image being used as a shelter is higher than a predetermined value, transmit the identification information and the position information of the vehicle to the outside.

13. A vehicle detection method, wherein, The control unit calculates an evaluation value of the vehicle being used as a shelter based on one or more determination conditions in the first image obtained by photographing a parked vehicle and its surroundings with a camera. Wherein, in the first image, an image of a far-infrared camera capable of measuring the temperature of a subject is included. When the control unit identifies that at least a part of the vehicle has a different temperature from its surroundings from the image of the far-infrared camera, the control unit evaluates that the possibility of the vehicle being used as a shelter is higher than the case where it is not identified that at least a part of the vehicle has a different temperature from its surroundings. The control unit also obtains a second image, which is obtained by the camera photographing at a position where the parked vehicle is photographed at a time different from the time when the first image of the parked vehicle is photographed. The control unit determines whether the vehicle is included in the second image. When the vehicle is not included, the control unit determines whether a fixture is included in the second image. When the fixture is not included, the control unit maintains or reduces the evaluation value. When the fixture is included, the control unit evaluates that the possibility of the vehicle being used as a shelter is higher than the case where the vehicle and the fixture are not included in the second image. The control unit evaluates the possibility of the vehicle being used as the shelter based on the evaluation value obtained from the second image.

14. The vehicle detection method according to claim 13, wherein, One or more determination conditions in the first image include whether an image of a predetermined article is included in the first image. When the image of the predetermined article is included in the first image, the control unit evaluates that the possibility of the vehicle being used as a shelter is higher than the case where the article is not included.

15. The vehicle detection method according to claim 14, wherein, The predetermined article includes at least any one of a tent, a chair, a table, clothing, and a floor mat.

16. The vehicle detection method according to claim 13, wherein, the control unit identifies the identification information of the vehicle from the first image, the control unit obtains the position information of the vehicle from the moving body equipped with the camera, when the control unit determines that the evaluation value indicating the possibility that the vehicle in the first image is used as a shelter is higher than a predetermined value, the control unit transmits the identification information and the position information of the vehicle to the outside.

Images