Fall monitoring device and method therefor

Abstract

A fall monitoring device is presented. The device comprises: an image sensor for acquiring an image of a person of interest; and a processor for processing the image of the person of interest from the image sensor, detecting, from the image, whether the person of interest falls, and determining the severity of the fall according to sequential changes in posture of the person of interest.

Description

Fall monitoring device and method therefor

[0001] The present invention relates to a fall monitoring device and a method therefor, and more particularly, to an image processing-based fall monitoring device and a method therefor.

[0002] Current fall detection methods used in hospitals involve nurses identifying high-risk factors for falls and assessing the patient's condition. To address falls, a common safety hazard in medical institutions, hospitals are currently undergoing smart transformation, and various related studies are underway.

[0003] Fall detection technology is being developed largely using sensor-based and lidar-based methods. Lidar technology can detect only certain types of falls by tracking the shape of the human skeleton, etc.

[0004] Sensor-based technology detects a set area based on a set pressure position, etc., and notifies the user when it deviates from this area through a danger detection system.

[0005] The present invention proposes a fall monitoring device and a method therefor.

[0006] More specifically, we propose a device, method, etc. for detecting fall of an object based on an image or image processing.

[0007] The problems to be solved by the present invention are not limited to the problems to be solved above, and other problems not mentioned can be clearly understood by a person having ordinary skill in the technical field to which the present invention belongs from the description below.

[0008] A fall monitoring device is proposed, the device including an image sensor for obtaining an image of a person of interest; and a processor for processing an image of the person of interest from the image sensor, detecting whether the person of interest has fallen from the image, and determining the severity of the fall based on sequential posture changes of the person of interest.

[0009] A fall monitoring method is proposed, the method including the steps of: acquiring an image of a person of interest; processing the acquired image and detecting from the image whether the person of interest has fallen; and determining the severity of the fall based on sequential posture changes of the person of interest.

[0010] The above problem solving methods are only some of the embodiments of the present invention, and various embodiments reflecting the technical features of the present invention can be derived and understood by a person having ordinary knowledge in the relevant technical field based on the detailed description of the present invention described below.

[0011] The present invention has the following effects.

[0012] According to the present invention, it is possible to efficiently detect whether a person has fallen.

[0013] According to the present invention, it is possible to track a fall victim and provide a corresponding image.

[0014] The effects according to the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art to which the present invention pertains from the detailed description of the invention below.

[0015] The accompanying drawings, which are included as part of the detailed description to aid in understanding the present invention, provide embodiments of the present invention and, together with the detailed description, explain the technical idea of ​​the present invention.

[0016] Figure 1 illustrates a block diagram of a fall monitoring device according to the present invention.

[0017] Figure 2 illustrates a perspective view of a fall monitoring device according to the present invention.

[0018] Figure 3 shows a flowchart of a fall monitoring method according to the present invention.

[0019] Figure 4 is a graph regarding the illumination control for a camera (image sensor) related to fall monitoring according to the present invention.

[0020] Figure 5 shows a flowchart of a fall monitoring method according to the present invention.

[0021] Figure 6 shows a flowchart of a fall monitoring method according to the present invention.

[0022] Figure 7 shows a flowchart of a fall monitoring method according to the present invention.

[0023] Figure 8 illustrates an example of setting a fall monitoring area according to the present invention.

[0024] Figure 9 shows a diagram of the importance or state of a fall according to the present invention.

[0025] Figure 10 shows a flowchart of a method for fall monitoring according to the present invention.

[0026] Figure 11 shows a flowchart of a method for fall monitoring according to the present invention.

[0027] Figure 12 shows a flowchart of a method for image-based heart rate measurement according to the present invention.

[0028] Figure 13 illustrates a structural diagram of a fall monitoring device according to the present invention that can be utilized with other devices, systems, etc.

[0029] Hereinafter, embodiments disclosed in this specification will be described in detail with reference to the attached drawings. Regardless of the drawing numbers, identical or similar components will be given the same reference numbers and redundant descriptions thereof will be omitted. The suffixes "module" and "part" used for components in the following description are assigned or used interchangeably only for the convenience of writing the specification, and do not in themselves have distinct meanings or roles. In addition, when describing the embodiments disclosed in this specification, if it is determined that a specific description of a related known technology may obscure the gist of the embodiments disclosed in this specification, a detailed description thereof will be omitted. In addition, the attached drawings are only intended to facilitate easy understanding of the embodiments disclosed in this specification, and the technical ideas disclosed in this specification are not limited by the attached drawings, and should be understood to include all modifications, equivalents, and substitutes included in the spirit and technical scope of the present invention.

[0030] Terms that include ordinal numbers, such as first, second, etc., may be used to describe various components, but the components are not limited by these terms. These terms are used solely to distinguish one component from another.

[0031] When a component is referred to as being "connected" or "connected" to another component, it should be understood that it may be directly connected or connected to that other component, but that there may be other components intervening. Conversely, when a component is referred to as being "directly connected" or "connected" to another component, it should be understood that there are no other components intervening.

[0032] Singular expressions include plural expressions unless the context clearly indicates otherwise.

[0033] In this application, terms such as “include” or “have” are intended to specify the presence of a feature, number, step, operation, component, part or combination thereof described in the specification, but should be understood not to exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts or combinations thereof.

[0034]

[0035] Figure 1 illustrates a block diagram of a fall monitoring device according to the present invention.

[0036] A fall monitoring device (100) includes a sensor (110) for acquiring an image or detecting ambient illuminance, a processor (120) for processing the acquired image to detect a fall of an object (person), and an output interface (130) for outputting visual or auditory information corresponding to the result of the fall detection.

[0037] The sensor (110) may include a plurality of sensors or multiple types of sensors.

[0038] A detailed description of the fall monitoring device (100) will be described with reference to other drawings.

[0039]

[0040] Figure 2 illustrates a perspective view of a fall monitoring device according to the present invention.

[0041] The fall monitoring device (100) may include a sensor (110) including a microphone sensor (111), an RGB image sensor (112), a night vision sensor (113), and an illuminance sensor (114). The type or number of sensors does not limit the scope of the present invention.

[0042] The fall monitoring device (100) may include a speaker (131) for sound output, a light-emitting diode (132) for indicating whether a fall has been detected or the condition of a faller, and a light-emitting diode (133) for indicating the condition of the device (100).

[0043] Additionally, the fall monitoring device (100) may include one or more infrared light emitting diodes (140) for a night vision sensor (113).

[0044] Although not shown, the fall monitoring device (100) includes a processor (120). The processor (120) processes images acquired through a sensor to extract fall-related information. The operation of the processor (120) will be described in detail with reference to FIGS. 3 to 12.

[0045]

[0046] Figure 3 illustrates a flowchart of a fall monitoring method according to the present invention. The fall monitoring method of Figure 3 may be performed by a fall monitoring device (100) or by a processor (120). For simplicity, the following description will be made as being performed by the processor (120). However, the subject of the performance does not limit the scope of the present invention.

[0047] The processor (120) receives an image input (S100). The image may include a still image (i.e., a photograph) or a video.

[0048] Image input is acquired through a sensor (110), and can be acquired through a general RGB image sensor (112) during the day and through an infrared image sensor (113) at night. In addition to image input, the processor (120) can also acquire voice input through a microphone sensor (111).

[0049]

[0050] Figure 4 is a graph illustrating illumination control for an image sensor related to fall monitoring according to the present invention. The processor (120) can implement an algorithm for the operation and switching of the image sensor in a daytime or nighttime environment. The daytime or nighttime environment is detected using illumination information acquired through the illumination sensor (114), and is distinguished as follows, for example, and the activation or operation of the image sensor is indicated accordingly.

[0051]

[0052] Ambient illumination: Day or night: Activated image sensor: 0 to 19 lux (lux) Night: Infrared image sensor: 19 to 21 lux Switching interval: RGB image sensor or infrared image sensor: 21 lux or more Day: RGB image sensor

[0053] As illustrated in Fig. 4, in a transition period where the illuminance has a value between 19 and 21 lux, it must be determined which image sensor to use. Line L1 represents the image sensor activation when the illuminance changes from a dark environment (night) to a bright environment (day), and line L2 represents the image sensor activation when the illuminance changes from a bright environment (day) to a dark environment (night). In Fig. 4, the transition of the image sensor is illustrated as being performed while maintaining the previously maintained state as much as possible. However, the illuminance value that serves as a reference for the transition may be changed depending on the design.

[0054] Additionally, a hunting phenomenon may occur in the illuminance values ​​acquired from the illuminance sensor near the illuminance value that serves as the reference for switching (hereinafter referred to as the "reference illuminance value"). Therefore, to reduce the hunting phenomenon, if the number of switching cycles within a preset time period exceeds the reference value, the average illuminance value over the preset time period may be compared with the reference illuminance value.

[0055] Additionally, in the case of night, an infrared image sensor (113) is used, and the processor (120) can adaptively control the brightness of the infrared light-emitting diode (140) for the image sensor (113) according to the ambient illuminance value.

[0056] The processor (120) processes the input image for fall detection (S200). Image processing is intended to detect objects (people) within the image and recognize the object's state. The object's state is information related to posture detection, such as whether the object is lying, standing, or sitting. A description of image processing will be provided below with reference to FIG. 5.

[0057] As a result of image processing, the processor (120) can determine whether the object has fallen.

[0058] The processor (120) can determine whether a fall has been detected based on the processed image (S300), and if it is determined to be a fall, it can perform post-processing such as outputting a warning alarm, transmitting information, etc. If it is determined to be not a fall based on the processed image, the processor (120) can receive image input again.

[0059]

[0060] Figure 5 illustrates a flowchart of a fall monitoring method according to the present invention. Figure 5 illustrates a specific procedure for image processing of Figure 4.

[0061] The processor (120) can configure presets before processing the image (S210). The presets can be input by the user through a human-machine interface (HMI), such as a touch display or keyboard. Alternatively, the presets can be selected and input from a plurality of presets based on the detection results of the surrounding environment, such as a sensor (110).

[0062] Regarding presetting, please refer to Fig. 6 for further details.

[0063] Referring to FIG. 6, single-person or multi-person monitoring settings (S211), monitoring area settings (S212), and facial recognition settings for registered individuals (S213) are provided. Although sequentially illustrated, the present invention is not limited to the order of FIG. 6, and preset settings may be configured or entered in a different order or simultaneously.

[0064] In the specification, it is expressed as configuring or entering a preset, which can be understood as setting the target of image processing before the processor (120) processes the image.

[0065] The processor (120) can configure a single or multiple person monitoring setting (S211). This is to set whether to detect a single person or multiple people in the input image through image processing.

[0066] The processor (120) can configure monitoring area settings (S212). The monitoring area corresponds to an area in the input image where fall detection is attempted, i.e., an area where fall monitoring is performed. That is, the input image can be configured to include a fall monitoring area and a safe area where fall monitoring is not performed. Fig. 8 illustrates a fall monitoring area (210) and a safe area (220). The area determined to be above the bed (200) is set as the safe area (220), and thus, object posture determination may not be performed in the safe area (220). Even if object posture determination is performed in the safe area (220), a determination as to whether or not a fall has occurred may not be made. At least some of the areas determined to be the floor within the hospital room are set as fall monitoring areas (210). In the fall monitoring area (210), object posture determination is performed and a determination as to whether or not a fall has occurred may be made.

[0067] The processor (120) can configure facial recognition settings for a registered person (S213). Before performing fall monitoring, the processor (120) can perform facial recognition and extraction on a person to be monitored for fall (hereinafter, referred to as “target person”). Then, the processor (120) can assign a unique value (e.g., ID) to the target person and store the facial recognition result (i.e., facial information) thereof in a storage such as a memory. There can be at least one target person, and the processor (120) can register the target person before performing fall monitoring. The registration of the target person can be triggered by a user input or by a pre-configured algorithm. By registering the target person, the processor (120) can manage the target person using an encrypted ID generated based on facial recognition. For example, it provides a function that can recognize the faces of multiple people, such as medical staff, guardians, and patients in a hospital room environment, and set and target a specific person for monitoring.

[0068] When registration of a target person is involved, the processor (120) attempts to detect a target person (i.e., a registered person) registered for fall monitoring from the input image. Even if multiple people are detected in the input image, the processor (120) can perform detection and fall monitoring for the registered person.

[0069] Returning to FIG. 5, the processor (120) can perform processing on the input image (S220). Processing on the input image includes determining the pose of an object or registered person in the input image.

[0070] Upon detection of a fall, the processor (120) may output a fall notification via a visual or auditory output interface (e.g., a light-emitting diode or speaker) (S230). For example, if a fall is detected based on the image processing result, the light-emitting diode (133) built into the fall monitoring device (100) may be set to emit a preset color. Additionally or alternatively, if a fall is detected based on the image processing result, the speaker (131) may be set to output a warning sound. Additionally or alternatively, if a fall is detected based on the image processing result, fall-related information may be transmitted to an external device, such as a display device or a remote server device, via a transceiver (not shown). The fall-related information may include a fall notification, unique information about the faller, location information about the faller, or coordinate information about the faller within the image. The fall-related information transmitted to the display device may include an input image or an image-processed input image, and may include a visual effect for notifying fall detection.

[0071]

[0072] Figure 7 shows a flowchart of a fall monitoring method according to the present invention.

[0073] If a fall is detected, the processor (120) can determine the severity of the fall (S410). The severity of the fall refers to the urgency or priority of the fall, and is based on the posture of the faller. Referring to FIG. 9, posture states P1 to P4 are illustrated, but the posture states are not limited thereto. If a fall is detected, the faller can be designated as state P2. P1 corresponds to a posture with no notable movement after the fall, P2 corresponds to a fallen posture, P3 corresponds to a standing posture, and P4 corresponds to a sitting posture.

[0074] However, if a state change occurs within a preset time period (P23, P24), a low level of fall importance can be assigned. Conversely, if another state change occurs within a preset time period (P21), a high level of fall importance can be assigned. The fall importance can be summarized as follows.

[0075] Importance Status Change Description Example Situation High Level P2->P1 (P21) Emergency Fall No movement detected after fall Medium Level P3->P2 (P32) Fall If you fall from a standing position and your face touches the floor P4->P2 (P42) Fall If you fall from a sitting position and your face touches the floor P1->P2 (P12) Fall If movement is detected again after a fall (after no movement was detected) Low Level P3->P4 (P34) or P4->P3 (P43) Not a fall If you change your posture

[0076] Meanwhile, information such as Table 2 is stored in a storage such as a memory (not shown) of a fall monitoring device (100), and the processor (120) may select importance according to changes in the condition of the fall victim.

[0077] The processor (120) can sequentially track objects or registered persons based on the fall severity (S420). This is activated when multiple objects or registered persons are detected in the input image. At this time, a tracking time can be set, and the higher the fall severity, the longer the tracking time can be set.

[0078] The processor (120) can perform PTZ (pan tilt zoom) control on a fall victim (S430). Even at this time, the aforementioned tracking time is applied, so that image tracking is performed through PTZ control for a longer period of time for objects or registered persons with a high degree of importance. For example, the higher the importance, the longer the tracking time is granted to the object or registered person, allowing a zoom image to be acquired, and the acquired image can be transmitted to a display device or a remote server device.

[0079] Time for tracking importance: High level T1, Medium level T2, Low level T3

[0080] Here, T1>=T2>=T3.

[0081] PTZ control is further explained with reference to Fig. 10.

[0082]

[0083] Figure 10 shows a flowchart of a method for fall monitoring according to the present invention.

[0084] The processor (120) receives detection area information related to an object or registered person, such as location coordinates and / or length and width (S431).

[0085] The processor (120) can check whether the PTZ (pan, tilt, zoom) setting value is being calculated using previously input information (S432).

[0086] The processor (120) can store the input detection area information (S433). By storing the input detection area information, it is assumed that the processor (120) is storing detection area information for a preset number of frames.

[0087] The processor (120) can calculate a zoom setting value using the stored detection area information (S434).

[0088] The processor (120) compares the calculated zoom setting value with the zoom setting value used in the previous frame, and if the difference between the two zoom setting values ​​exceeds a threshold, a zoom operation can be determined (S435).

[0089] The processor (120) compares the current detection area information with the detection area information of the previous frame, and if the difference between the two detection area information exceeds a threshold, the processor (120) can determine a pan and tilt operation (S436). At this time, the processor (120) can calculate a setting value for the pan and tilt operation.

[0090] The processor (120) can perform PTZ operation by determining zoom, pan, and tilt operations and using the setting values ​​therefor (S437).

[0091]

[0092] Figure 11 illustrates a flowchart of a fall monitoring method according to the present invention. Figure 11 illustrates heart rate measurement using face detection of a fall victim after a fall has been detected.

[0093] The processor (120) can check whether the face area of ​​the fall victim has been extracted from the input image (S510). If extraction of the face area of ​​the fall victim is not possible, the processor (120) can induce the fall victim to look at the image sensor (112) using the light-emitting diode (132) of the device (100).

[0094] If the extraction of the face area of ​​the faller is successful, the processor (120) can perform heart rate measurement based on the extracted face area (S520).

[0095] The processor (120) can store the measured heart rate or transmit it to a display device or the like to output it. In addition, the processor (120) can control the light emitting diode (132) to output a corresponding color as the measured heart rate falls within the normal range or exceeds the normal range. In addition, the processor (120) can control the speaker (131) to output a corresponding sound as the measured heart rate falls within the normal range or exceeds the normal range.

[0096] Video-based heart rate measurement is described in more detail with reference to Fig. 12.

[0097]

[0098] Figure 12 shows a flowchart of a method for image-based heart rate measurement according to the present invention.

[0099] Video-based, non-contact heart rate measurement relies on an algorithm capable of measuring heart rate and heart rate variability. This allows heart rate to be measured using a captured facial image.

[0100] The processor (120) performs initial setup for non-contact heart rate measurement (S521). In addition, the processor (120) performs image sensor or camera initialization for non-contact heart rate measurement (S522).

[0101] The processor (120) may acquire an input image or an image processed from the input image (S523). The acquired image records the appearance of the fall victim. The processor (120) may attempt to detect a face from the acquired image (S524). If face detection from the acquired image is successful, the processor (120) may extract a region of interest (S525). The region of interest includes at least the face of the fall victim.

[0102] The processor (120) may perform signal processing (e.g., noise removal, amplification, etc.) on the region of interest image to extract blood flow or pulse information (S526). Then, the processor (120) may extract a remote photoplethysmography (rPPG) signal from the signal processing result. The processor (120) may perform frequency analysis on the rPPG signal (S528). The processor (120) may calculate a heart rate (HR) using the frequency analysis result (S529).

[0103]

[0104] The fall monitoring device (100) described above can detect an object's fall from an input image, determine the severity of the fall, and control the PTZ of the image sensor for the faller on-device. In addition, the fall monitoring device (100) can measure heart rate by obtaining facial blood flow data from the image.

[0105]

[0106] Figure 13 illustrates a structure in which a fall monitoring device according to the present invention can be utilized with other devices, systems, etc.

[0107] The fall monitoring device (100) can communicate with a server (300), such as a cloud server, or a display device (400) via a wired or wireless network. Through this, the fall monitoring device (100) can externally transmit at least one of a fall notification and / or unique information of a fall victim, an input image or an image including a fall scene, or audiovisual effect information for a fall notification.

[0108] In particular, by transmitting to a patient monitoring system (500), it enables remotely located medical staff or administrators to monitor objects that may fall in real time.

[0109]

[0110] Referring back to Figure 1, the fall monitoring device (100) is summarized as follows.

[0111] The fall monitoring device (100) includes an image sensor (110) for acquiring an image of a person of interest. In addition, the fall monitoring device (100) includes a processor (120) for processing an image of the person of interest from the image sensor and detecting whether the person of interest has fallen from the image.

[0112] The processor (120) can detect the face of a registered person from an image acquired from an image sensor and designate the registered person whose face has been detected as a person of interest.

[0113] The processor (120) can set a certain area within an image acquired from an image sensor as a safe area and detect whether a person of interest has fallen in an area outside the safe area.

[0114] The processor (120) can classify or determine the severity of a fall based on the sequential changes in posture of the person of interest. If there are multiple persons of interest, the processor (120) can sequentially track multiple persons of interest based on the severity of the fall. In this case, the tracking time can be set based on the severity of the fall.

[0115] The processor (120) can control the image sensor to track a person of interest according to PTZ (Pan, Tilt, Zoom) control during the tracking time.

[0116] The processor (120) can detect the face of a person of interest and measure the heart rate of the person of interest from an image including the face.

[0117] The processor (120) can detect the face of a registered person from an image acquired from an image sensor and designate the registered person whose face has been detected as a person of interest.

[0118] Additionally, the fall monitoring device (100) may further include an output interface, such as a light emitting diode or speaker, for outputting a visual or auditory notification when it is detected that a person of interest has fallen.

[0119] When the processor (120) detects that a person of interest has fallen, it transmits fall-related information to the management server device, and the fall-related information may include at least one of a fall notification, unique information (identification information) of the person of interest (faller), location information of the person of interest, or coordinate information of the person of interest within the image.

[0120]

[0121] In addition, as another aspect of the present invention, the operation of the proposal or invention described above may be implemented, performed or executed by a “computer” (a comprehensive concept including a system on chip (SoC) or a (micro) processor, etc.), or may be provided as a code or a computer-readable storage medium storing or including the code or a computer program product, and the scope of the present invention may be extended to the code or the computer-readable storage medium storing or including the code or the computer program product.

[0122]

[0123] The detailed description of the preferred embodiments of the present invention disclosed above has been provided to enable those skilled in the art to implement and practice the present invention. While the above description has been made with reference to preferred embodiments of the present invention, those skilled in the art will appreciate that various modifications and variations of the present invention, as defined by the following claims, are possible. Accordingly, the present invention is not intended to be limited to the embodiments disclosed herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. As a fall monitoring device, An image sensor for acquiring images of a person of interest; and A processor comprising: a processor for processing an image of the person of interest from the image sensor, detecting whether the person of interest has fallen from the image, and determining the importance of the fall based on sequential posture changes of the person of interest. , fall monitoring device.

2. In paragraph 1, the processor: A portion of the image acquired by the image sensor is set as a safe area, and a fall of the person of interest is detected in an area outside the safe area. , fall monitoring device.

3. In the first paragraph, the processor If there are multiple people of interest, the multiple people of interest are sequentially tracked according to the importance of the fall. , fall monitoring device.

4. In the third paragraph, the time for the tracking is set according to the importance of the fall. , fall monitoring device.

5. In paragraph 4, The above processor: Controlling the image sensor to track the person of interest according to PTZ (Pan, Tilt, Zoom) control during the time for the above tracking. , fall monitoring device.

6. In the first paragraph, the processor Detecting the face of the person of interest and measuring the heart rate of the person of interest from an image including the face. , fall monitoring device.

7. In the first paragraph, the processor Detecting the face of a registered person from an image acquired from the image sensor, and designating the registered person whose face is detected as the person of interest. , fall monitoring device.

8. In the first paragraph, when it is detected that the person of interest has fallen, an output interface for outputting a visual or auditory notification is further included. , fall monitoring device.

9. In the first paragraph, the processor When it is detected that the person of interest has fallen, the fall-related information is transmitted to the management server device, The above fall-related information includes at least one of a fall accident occurrence notification, unique information of the person of interest, location information of the person of interest, or coordinate information of the person of interest within the video. , fall monitoring device.

10. As a fall monitoring method, Step of obtaining video of a person of interest; A step of processing the acquired image and detecting whether the person of interest has fallen from the image; and A step of determining the importance of the fall according to the sequential posture changes of the person of interest , a method for monitoring falls.

11. In paragraph 10, A step of setting a certain area within the acquired image as a safe area and detecting whether the person of interest has fallen in an area outside the safe area is included. , a method for monitoring falls.

12. In paragraph 10, In the case where there are multiple persons of interest, a step of sequentially tracking the multiple persons of interest according to the importance of the fall is included. , a method for monitoring falls.

13. In paragraph 12, The time for the above tracking is set according to the severity of the fall. , a method for monitoring falls.

14. In paragraph 13, A step of controlling the image sensor to track the person of interest according to PTZ (Pan, Tilt, Zoom) control during the time for the tracking. , a method for monitoring falls.

15. In paragraph 10, A step of detecting a face of the person of interest and measuring a heart rate of the person of interest from an image including the face. , a method for monitoring falls.

16. In paragraph 10, A step of detecting a face of a pre-registered person from an image acquired from the image sensor and designating the pre-registered person whose face is detected as the person of interest is included. , a method for monitoring falls.

17. In paragraph 10, A step of outputting a visual or auditory notification upon detection that the person of interest has fallen , a method for monitoring falls.

18. In paragraph 10, Including a step of transmitting fall-related information to a management server device when it is detected that the person of interest has fallen, The above fall-related information includes at least one of a fall accident occurrence notification, unique information of the person of interest, location information of the person of interest, or coordinate information of the person of interest within the video. , a method for monitoring falls.

19. A nonvolatile computer-readable medium storing a computer program configured to perform a method according to any one of claims 10 to 18.

Images