Disaster prevention support system

Abstract

To provide a disaster prevention support system that can identify the location of a person in a disaster monitoring area using image data captured by a mobile terminal.SOLUTION: A mobile terminal is implemented with an application that transmits image data captured by a camera to a cloud server. The cloud server includes: a machine learning unit that performs pre-learning to identify a location in a facility from the image data based on the image data at various locations in a disaster monitoring area received from the mobile terminal when the application is executed during normal times when no disaster is occurring; a location estimation unit that estimates location information regarding the location of the image data from the prior learning results by the machine learning unit, using the image data taken in the disaster monitoring area received from the mobile terminal by the execution of the application at the time of a disaster as input; and a notification processing unit that notifies external parties of the location information estimated by the location estimation unit.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present disclosure relates to a disaster prevention support system having a function of estimating the position information of a shooting location from image data captured within a disaster monitoring area.

Background Art

[0002] There is a cloud-based disaster prevention support system that supports emergency response during disasters such as fires and earthquakes (see, for example, Non-Patent Document 1). The disaster prevention support system according to Non-Patent Document 1 includes a smartphone installed with application software and a cloud server, and notifies necessary persons of task information necessary for actions during disasters, connects organizations with tasks, and enables close cooperation and prompt actions.

[0003] Specifically, the disaster prevention support system according to Non-Patent Document 1 enables remote centralized monitoring of various facilities installed across multiple buildings within a site. Furthermore, when an abnormality occurs, it is possible to provide prompt and accurate support by pushing notifications to grouped mobile terminals.

[0004] In addition, the disaster prevention support system according to Non-Patent Document 1 can transmit image data of the on-site situation at the time of a disaster captured by the camera of a smartphone to, for example, the smartphones of members of a fire department by a simple operation. As a result, the disaster situation can be easily shared.

Prior Art Documents

Non-Patent Documents

[0005]

Non-Patent Document 1

Summary of the Invention

[0006] In the event of a disaster such as a fire or earthquake, it is crucial to quickly locate victims and determine if anyone is trapped or unable to escape, as this is essential for effective relief efforts. However, while Non-Patent Document 1 provides information on the disaster situation, it does not provide information on the locations of people trapped within the disaster monitoring area.

[0007] Furthermore, Non-Patent Document 1 describes how, for example, members of a self-defense fire brigade can share disaster situations among themselves using image data they have captured. However, image data is not being effectively utilized for other purposes.

[0008] Furthermore, while automatic fire alarm systems can pinpoint the location of the fire source, they do not provide information on the location of people who were unable to escape within the disaster monitoring area.

[0009] In recent years, many people carry mobile devices such as smartphones. Therefore, if image data captured by these mobile devices can be used to identify the location of the image during a disaster and estimate the location of people who are unable to evacuate, it is expected that evacuation guidance and rescue operations can be carried out quickly and accurately.

[0010] This disclosure is made to solve the above-mentioned problems and aims to provide a disaster prevention support system that can identify the location of people within a disaster monitoring area using image data captured by a mobile device. [Means for solving the problem]

[0011] The disaster prevention support system relating to this disclosure is a disaster prevention support system comprising a cloud server and a mobile terminal, wherein the mobile terminal is equipped with photo reporting application software that captures images of desired locations within a disaster monitoring area using a camera and transmits the image data to the cloud server, and the cloud server comprises a machine learning unit that, in normal times when no disaster has occurred, performs pre-training to identify locations within a facility from image data based on image data from various locations within the disaster monitoring area received from the mobile terminal when the photo reporting application software is executed, and a location estimation unit that, in the event of a disaster, takes image data captured within the disaster monitoring area received from the mobile terminal when the photo reporting application software is executed as input and estimates location information related to the image capture location from the pre-training results of the machine learning unit, and a notification processing unit that notifies the location information estimated by the location estimation unit to an external party. [Effects of the Invention]

[0012] According to this disclosure, a disaster prevention support system can be obtained that can identify the location of a person within a disaster monitoring area using image data captured by a mobile device. [Brief explanation of the drawing]

[0013] [Figure 1] This is an explanatory diagram showing the basic configuration of the disaster prevention support system according to Embodiment 1 of this disclosure. [Figure 2] This is a functional block diagram of the disaster prevention support system according to Embodiment 1 of the present disclosure. [Figure 3] This is an explanatory diagram showing a specific example of capturing image data using photo reporting application software with a mobile terminal according to Embodiment 1 of this disclosure. [Figure 4] This is a layout diagram of an elderly care facility to which the disaster prevention support system according to Embodiment 2 of this disclosure is applied. [Figure 5] This is an explanatory diagram relating to the symbols shown in Figure 4 in Embodiment 2 of this disclosure.

Best Mode for Carrying Out the Invention

[0014] Hereinafter, a preferred embodiment of the disaster prevention support system of the present disclosure will be described with reference to the drawings. The disaster prevention support system according to the present disclosure performs pre-learning for specifying the position within a facility from image data based on various image data pre-captured at various locations in a disaster monitoring area, and uses the image data captured at the time of a disaster as an input, and has a technical feature in that a function of outputting an estimation result regarding the imaging location as position information from the pre-learning result is added to a conventional disaster prevention support system.

[0015] Embodiment 1. First, the basic configuration of the disaster prevention support system according to the present disclosure will be described. FIG. 1 is an explanatory diagram showing the basic configuration of the disaster prevention support system according to Embodiment 1 of the present disclosure. The disaster prevention support system according to Embodiment 1 has a basic configuration including a cloud server 200 and a mobile terminal 300.

[0016] In addition, the cloud server 200 can receive an emergency signal notifying that a fire has occurred from the automatic fire alarm equipment 100 installed in the disaster monitoring area. Although not shown in FIG. 1, the cloud server 200 can receive an emergency signal notifying that an emergency has occurred from, for example, an emergency earthquake early warning receiving device, a water level sensor, important equipment in a factory, etc., in addition to the automatic fire alarm equipment 100.

[0017] In addition, the cloud server 200 can communicate with, for example, a PC 400 owned by a disaster prevention manager and provide necessary information.

[0018] The configuration shown in FIG. 1 is the same as the configuration of the conventional disaster prevention support system disclosed in Non-Patent Document 1.

[0019] In the present disclosure, there is a technical feature in that a function of estimating an imaging location using a prior machine learning result is added to the image data captured by the mobile terminal 300 by the cloud server. Therefore, this additional function will be described in detail.

[0020] FIG. 2 is a functional block diagram of the disaster prevention support system according to Embodiment 1 of the present disclosure. The mobile terminals 300(1) to 300(M) correspond to the mobile terminal 300 in FIG. 1 and are standard-equipped with cameras. Each of the mobile terminals 300(1) to 300(M) is installed with photographic reporting application software that captures a desired location within the disaster monitoring area as image data using a camera and transmits the image data to the cloud server 200.

[0021] FIG. 3 is an explanatory diagram showing a specific example of capturing image data using the photographic reporting application software by the mobile terminal 300 according to Embodiment 1 of the present disclosure. The owner of the mobile terminal 300 can easily capture a desired location by activating the photographic reporting application software.

[0022] Furthermore, although not shown in the figure, the owner of the mobile terminal 300 can easily transmit the captured image data to the cloud server 200 by tapping the "Transmit" button displayed on the screen after imaging.

[0023] The cloud server 200 includes a machine learning unit 201, a position estimation unit 202, and a notification processing unit 203. The machine learning unit 201 acquires image data captured for each known location within the disaster monitoring area from the mobile terminal 300 that has executed the photographic reporting application software during normal times when no disaster has occurred.

[0024] Then, the machine learning unit 201 performs pre-learning processing for specifying the imaging position from the image data by performing machine learning on the image data captured at the known locations.

[0025] The position estimation unit 202 acquires image data captured within the disaster monitoring area from a mobile terminal 300 running photo reporting application software when a disaster occurs. Then, the position estimation unit 202 takes the image data captured at an arbitrary location at the time of the disaster as input and estimates the location information related to the capture location from the pre-training results of the machine learning unit 201.

[0026] In other words, by performing pre-training processing using various image data captured at known locations by the machine learning unit 201, the position estimation unit 202 can identify the location from a single image data captured at any location.

[0027] The notification processing unit 203 notifies an external party of the location information estimated by the location estimation unit 202. For example, the notification processing unit 203 can inform a higher-level device of the estimated location information via a network.

[0028] Furthermore, if location information is estimated by the location estimation unit 202 during a disaster, the notification processing unit 203 can also push notification of the location information to a pre-registered group of terminals 500.

[0029] In Figure 2, the terminal group 500 consists of the administrator PC 401 and the rescue personnel mobile terminals 301(1) to 301(N). Here, the administrator PC 401 corresponds to PC 400 in Figure 1 and is the PC owned by the administrator of the disaster monitoring area.

[0030] Furthermore, the rescue personnel handheld terminals 301(1) to 301(N) correspond to the handheld terminal 300 in Figure 1, and are, for example, handheld terminals carried by personnel who conduct rescue operations in disaster monitoring areas, such as personnel at disaster prevention centers of various companies and personnel of self-defense fire brigades.

[0031] Therefore, the notification processing unit 203 can quickly transmit location information of people within the disaster monitoring area to a group of pre-registered terminals 500 via push notifications in the event of a disaster.

[0032] As described above, according to Embodiment 1, a function can be added to the disaster prevention support system that estimates the location of image data captured at the time of a disaster based on pre-training results using various image data captured at known locations, and provides this as location information.

[0033] As a result, owners of mobile devices equipped with photo reporting application software that sends image data to a cloud server can easily communicate their location information in the event of a disaster by activating this application software and sending the image data.

[0034] Therefore, when a disaster strikes, various people, such as members of the self-defense fire brigade and disaster victims, can simply run the photo reporting application software, and the cloud server can quickly estimate the location of the person holding the mobile device that sent the image data and transmit that location information. As a result, disaster relief support, such as rescue operations for people who are unable to escape and evacuation guidance, can be carried out more quickly.

[0035] Furthermore, the disaster monitoring area can encompass a variety of uses, including factory facilities, elderly care facilities, hospitals, and hotels within the site. Therefore, by providing location information within the disaster monitoring area in the event of a disaster, it can contribute to disaster prevention support such as rescue operations for people who are unable to evacuate and evacuation guidance in various applications.

[0036] Embodiment 2. This second embodiment specifically describes how the disaster prevention support system described in the first embodiment is applied to a facility for the elderly, and how the location information estimation results are effectively utilized in the event of a disaster.

[0037] Figure 4 is a layout diagram of a facility for the elderly to which the disaster prevention support system according to Embodiment 2 of this disclosure is applied. In Figure 4, an example is shown in which, on the first floor of the elderly facility, smoke detectors 2, heat detectors 3, optical alarm devices 4, and emergency exit signs 5 are arranged within the disaster monitoring area as disaster prevention equipment, and an automatic fire alarm system 100 controlled centrally by a fire alarm receiver 1 is installed.

[0038] Furthermore, Figure 5 is an explanatory diagram relating to the symbols shown in Figure 4 in Embodiment 2 of this disclosure.

[0039] In an elderly care facility with a layout like that shown in Figure 4 and equipped with an automatic fire alarm system 100, the fire alarm receiver 1 can monitor fires occurring in the disaster monitoring area, which consists of resident rooms 1-6, the management room, and the corridor.

[0040] We will now specifically describe the case where the disaster prevention support system described in Figure 2 of Embodiment 1 is applied to an elderly care facility equipped with such a configuration.

[0041] During normal times, the administrator of an elderly care facility, or a resident of such a facility, can use a mobile terminal 300 to transmit image data captured at each known location within the disaster monitoring area to a cloud server 200.

[0042] The machine learning unit 201 within the cloud server 200 can perform pre-training processing to identify the imaging location from image data by applying machine learning to various image data captured at known locations during normal operation.

[0043] Furthermore, in the event of a disaster, the administrator or residents of the elderly care facility can easily send image data of their location to the cloud server 200 using a mobile terminal 300 equipped with photo reporting application software.

[0044] On the other hand, the position estimation unit 202, which acquires image data, can use the image data captured at any location during a disaster as input and estimate positional information related to the imaging location from the pre-training results of the machine learning unit 201.

[0045] Furthermore, if location information is estimated by the location estimation unit 202 during a disaster, the notification processing unit 203 can push notification of the location information to the pre-registered group of terminals 500.

[0046] Therefore, elderly people who are unable to escape in their rooms or other enclosed spaces can quickly notify a group of pre-registered terminals 500 of their location simply by using a mobile terminal 300 equipped with photo reporting application software to send image data.

[0047] Furthermore, managers or rescuers of elderly care facilities can quickly inform a group of pre-registered terminals 500 of their location and request assistance simply by sending image data using a mobile terminal 300 equipped with photo reporting application software at the location where rescue operations are taking place.

[0048] Furthermore, if a layout similar to the one shown in Figure 4 is used on other floors, it may not be possible to estimate a unique location, and multiple location information may be estimated, such as the emergency exit on the west side of the first to fifth floors.

[0049] However, even in such cases, if the location information is known, the manager of the elderly care facility or rescuers who have been notified of the location can prioritize rushing to the area closest to the fire source. In other words, even if the estimated location information is not just one location but multiple locations, the location information can be effectively utilized.

[0050] Furthermore, even if multiple floors have the same layout, it is possible to narrow down the floors by pre-training the model using image data that includes elements such as plants and views from windows, as shown in Figure 3 above.

[0051] Furthermore, even if elderly people are unable to escape in time during a disaster, if they experience an emergency such as falling or becoming ill in their home, they can use a mobile terminal 300 equipped with application software to send image data, allowing them to quickly inform others of their location and request assistance.

[0052] As described above, according to Embodiment 2, by applying the disaster prevention support system described herein to facilities for the elderly, etc., the location of the person carrying the mobile device within the disaster monitoring area can be easily identified, even during normal times when no fire is occurring.

[0053] As a result, in the event of a disaster such as a fire or earthquake, the location of people who are unable to escape or who are in need of help can be easily identified, and this can be effectively used for appropriate evacuation guidance and rapid rescue operations. Furthermore, even outside of disaster situations, location information of suspected unforeseen circumstances can be easily identified, allowing for rapid response and appropriate action. [Explanation of symbols]

[0054] 1 Fire alarm receiver, 2 Smoke detector, 3 Heat detector, 4 Light alarm device, 5 Emergency exit sign, 100 Automatic fire alarm system, 200 Cloud server, 201 Machine learning unit, 202 Position estimation unit, 203 Notification processing unit, 300, 300(1)~300(M) Mobile terminals, 301(1)~301(N) Rescue personnel mobile terminals (mobile terminals), 400 PC, 401 Administrator PC (PC), 500 Terminal group.

Claims

1. A disaster prevention support system comprising a cloud server and a mobile terminal, The aforementioned mobile terminal is equipped with photo reporting application software that captures images of desired locations within the disaster monitoring area using its camera and transmits the image data to the cloud server. The aforementioned cloud server is During normal times when no disaster has occurred, the photo reporting application software is executed, and based on image data from various locations within the disaster monitoring area received from the mobile terminal, a machine learning unit performs pre-training to identify locations within the facility from the image data. When a disaster occurs, the photo reporting application software is executed and receives image data captured within the disaster monitoring area from the mobile terminal as input. The position estimation unit estimates location information related to the image capture location based on the pre-training results of the machine learning unit. A notification processing unit that notifies the external location information estimated by the location estimation unit. A disaster prevention support system that includes [unclear / unclear].

2. When the location information is estimated by the location estimation unit during the occurrence of the disaster, the notification processing unit will push notification of the location information to the pre-registered group of terminals. The disaster prevention support system according to claim 1.

Images