Fall sensing device, fall sensing system, and fall sensing method

Abstract

The present invention comprises: a fall determination unit (120) that detects the fall of an object on the basis of a moving body component that is a signal component which indicates the moving body and is extracted from a sensor signal; an immediate post state detection unit (130) that, when the fall determination unit has detected the fall of the object, detects the motion state of the object after the fall; and a comprehensive determination unit (140) that determines whether the fall is the fall of a living body or the fall of a non-living body on the basis of the motion state detected by the immediate post state detection unit.

Description

Fall detection device, fall detection system, and fall detection method

[0001] The disclosed technology relates to a fall detection technology for detecting the fall of an object.

[0002] Among fall detection technologies, for example, there is a technology for detecting the fall of a "person". The "abnormality determination system" described in Patent Document 1 infers that an occupant has fallen when it is determined that the transition speed of the occupant from the posture immediately before the fall to the fallen posture (lying position) is outside a predetermined range (first range) based on information from a room sensor provided in the room.

[0003] Japanese Unexamined Patent Application Publication No. 2023-050019 (Daiwa House Industry Co., Ltd.)

[0004] However, when there is an object that falls within the detectable range of the sensor other than a person, the technology described in Patent Document 1 has a problem that it may erroneously determine that a person has fallen even when an object has fallen.

[0005] The present disclosure aims to solve the above problems and enable more accurate determination of the fall of a detection target compared to the prior art.

[0006] The fall detection device of the present disclosure includes: a fall determination unit that detects the fall of an object based on a moving body component, which is a signal component indicating a moving body extracted from a sensor signal; an immediately after state detection unit that detects the motion state of the object after the fall when the fall of the object is detected by the fall determination unit; and a comprehensive determination unit that determines whether the fall is a fall of a living body or a non-living body based on the motion state detected by the immediately after state detection unit.

[0007] According to the present disclosure, it is possible to more accurately determine the fall of a detection target compared to the prior art.

[0008] Figure 1 is a diagram showing an example of the configuration of a fall detection device according to Embodiment 1 of this disclosure. Figure 2 is a flowchart showing an example of the processing of a fall detection device according to Embodiment 1 of this disclosure. Figure 3 is a diagram showing an example of the configuration of a fall detection device according to Embodiment 2 of this disclosure. Figure 4 is a diagram showing a first example of criteria for determining the type of object in the fall detection device of this disclosure. Figure 5 is a diagram showing a second example of criteria for determining the type of object in the fall detection device of this disclosure. Figure 6 is a flowchart showing an example of the processing of a fall detection device according to Embodiment 2 of this disclosure. Figure 7 is an image diagram showing an example of a sensing area set in a fall detection device according to Embodiment 3 of this disclosure. Figure 8 is a flowchart showing an example of the processing of a fall detection device according to Embodiment 3 of this disclosure. Figure 9 is a diagram showing an example of the configuration of a fall detection device according to Embodiment 5 of this disclosure. Figure 10 is a flowchart showing an example of the processing of a fall detection device according to Embodiment 5 of this disclosure. Figure 11 is a diagram showing an example of the configuration of a fall detection device according to Embodiment 6 of this disclosure. Figure 12 is a flowchart showing an example of the processing of a fall detection device according to Embodiment 6 of this disclosure. Figure 13 shows a first example of a hardware configuration for realizing the functions according to the configuration of this disclosure. Figure 14 shows a second example of a hardware configuration for realizing the functions according to the configuration of this disclosure.

[0009] To further illustrate this disclosure, embodiments of this disclosure will be described below with reference to the accompanying drawings.

[0010] Embodiment 1. Embodiment 1 describes an example of the basic configuration of a fall detection device.

[0011] An example of the configuration of a fall detection device according to Embodiment 1 of this disclosure will be described. Figure 1 is a diagram showing an example of the configuration of a fall detection device according to Embodiment 1 of this disclosure. With respect to the fall detection device 100 in Figure 1, in the description, in order to distinguish it from the configuration of other embodiments, the fall detection device 100 according to Embodiment 1 will be referred to as "fall detection device 100 (100A)".

[0012] The fall detection device 100 (100A) detects when an object that is the target of detection falls over. The fall detection device 100 (100A) detects when an object that may fall over falls over falls over within the detection target area, which is the space in which such an object exists or is likely to exist. The targets of detection by the fall detection device 100 (100A) can be, for example, the fall of a living being, the fall of a person, the fall of a non-living being, the fall of an object, or a combination thereof. The fall detection device 100 (100A) may be configured to include some or all of the functions described in the following embodiments. In this description, the fall detection device 100 is described as an integrated device having the functions of each component, but it is not limited to this, and some of the components or functions may be configured to be provided on a cloud server, server device, or other device. The fall detection device 100 (100A) shown in Figure 1 is configured to include a fall determination unit 120, an immediate post-fall state detection unit 130, and an overall determination unit 140.

[0013] The fall detection unit 120 of the fall detection device 100 (100A) detects the fall of an object in the detection target area. The fall detection unit 120 receives a motion component, which is a signal component indicating a moving object extracted from the sensor signal, and detects the fall of the object based on this motion component. The fall detection unit 120 receives the motion component from, for example, a motion reflection component extraction unit (see the embodiment described later) provided outside or inside the device.

[0014] The immediate post-tip state detection unit 130 of the fall detection device 100 (100A) detects the motion state of an object after it has fallen. The immediate post-tip state detection unit 130 detects the motion state of an object after it has fallen if the fall determination unit 120 has detected that the object has fallen. The immediate post-tip state detection unit 130 detects the motion state of an object after it has fallen in a time series from immediately after the fall is detected. The detected motion state is, for example, a motion state that differs between living and non-living objects, and is a motion state obtained based on, for example, the time series change in the velocity of the object from immediately after the fall is detected, or the time series change in the velocity of the object from immediately after the fall is detected. The detected motion state is also characteristic of the human body, such as vital information and limb movements.

[0015] The comprehensive determination unit 140 of the fall detection device 100 (100A) determines whether the detected object has fallen. Based on the motion state detected by the immediate post-state detection unit 130, the comprehensive determination unit 140 determines, for example, whether it is a fall of a living being or a fall of a non-living being. If the detected object is a living being, the comprehensive determination unit 140 determines that it is a fall of a living being and outputs the determination result. If the detected object is a non-living being, the comprehensive determination unit 140 determines that it is a fall of a non-living being and outputs the determination result. Alternatively, the comprehensive determination unit 140 may be configured to output the determination result, along with the type of object that fell, based on the determination result of whether it is a fall of a living being or a fall of a non-living being. The determination result is, for example, fall information that includes information identifying the object that fell. The fall information may also include information indicating the state of the object that fell. Alternatively, the comprehensive determination unit 140 may be configured to output fall information as a determination result, including the probability of falling for each type of object that likely fell. For example, the system may be configured to output fall information that includes the probability of whether the fall is caused by a living organism or a non-living object. In this case, the comprehensive judgment unit 140 outputs fall information that includes, for example, the percentage of possibility that the fall is caused by a living organism, the percentage of possibility that the fall is caused by a non-living object, or both of the percentages of possibility that the fall is caused by a living organism and the percentage of possibility that the fall is caused by a non-living object. Specifically, the fall information is information that includes a combination of information that identifies the type of object and a value (probability value) that represents the probability of the fall, and can be expressed in the form of, for example, "X% probability of a living organism falling" or "Y% probability of a non-living object falling". Alternatively, it can be expressed in the form of, for example, "X% probability of a person falling" or "Y% probability of an object falling". The number of decimal places of "X" and "Y" is not particularly limited, but can be, for example, a predetermined number of digits. In the comprehensive judgment unit 140, the probability is calculated, for example, using the output of a machine learning model or a confidence score based on rule-based judgment results. The machine learning model is a model that takes information including the motion state of an object that is likely to have fallen as input and is trained to output the probability of falling for each type of object.The type of object in this disclosure is not limited to being identifiable, but may be, for example, a living or non-living organism, or a person or an object. A method for calculating a confidence score based on rule-based judgment results may be, for example, a heuristic approach. In a heuristic approach, probabilities are calculated based on past data and experience as follows. It is also possible to calculate probabilities by combining various types of information. If vital information is below a predetermined value, there is a Y% probability that it is an object. If the time-series change in distance is above a predetermined value, there is a Y% probability that it is an object. The predetermined value is, for example, a pre-set threshold.

[0016] The fall detection device 100 (100A) is configured to include, in addition to the above configuration, a control unit (not shown), a storage unit (not shown), and a communication unit (not shown). The control unit (not shown) controls the entire fall detection device 100 (100A) and each of its components. The control unit (not shown) activates the fall detection device 100 (100A) according to an external command, for example. The control unit (not shown) also controls the state of the fall detection device 100 (100A) (operating state = state such as activated, shut down, or sleep). The storage unit (not shown) stores the data used by the fall detection device 100 (100A). The storage unit (not shown) stores the output (output data) from each component of the fall detection device 100 (100A), for example, and outputs the requested data to the requesting component for each component. The communication unit (not shown) communicates with external devices. For example, the fall detection device 100 (100A) communicates with a peripheral device (e.g., an external device). For example, if the fall detection device 100 (100A) and the external device are not connected by a wire, the communication unit (not shown) has the function of communicating between the fall detection device 100 (100A) and the external device. The communication unit (not shown) also has the function of communicating with a server device that has some of the functions of the fall detection device 100 (100A) or with an external server device. The control unit (not shown), storage unit (not shown), and communication unit (not shown) are the same in the embodiments described later.

[0017] Next, an example of processing of the fall detection device according to Embodiment 1 will be described. Figure 2 is a flowchart showing an example of processing of the fall detection device according to Embodiment 1 of this disclosure. The processing shown in Figure 2 is a fall detection method by the fall detection device. When the fall detection device 100 (100A) according to Embodiment 1 receives a monitoring command from, for example, an external monitoring system, it starts the processing shown in Figure 2 ("start").

[0018] The fall detection device 100 (100A) then performs a process to determine whether it has acquired a dynamic reflection component (dynamic component acquisition step) (step ST1010 "Dynamic reflection component acquired?"). In this process, the fall determination unit 120 of the fall detection device 100 (100A) acquires the dynamic component from, for example, a dynamic reflection component extraction unit (see the embodiment described later) provided outside or inside the device. If there are multiple moving objects in the detection target area, the fall determination unit 120 acquires the dynamic reflection component for each moving object.

[0019] The fall detection device 100 (100A) then performs a fall determination process (fall determination step) (step ST1020 "Fall?"). In the fall determination process, the fall determination unit 120 of the fall detection device 100 (100A) receives a motion component, which is a signal component indicating a moving object extracted from the sensor signal, and detects the fall of the object based on the motion component.

[0020] If the fall detection device 100 (100A) determines in the fall determination process of step ST1020 that it has not detected a fall (step ST1020 "Fall?"), it repeats the process of step ST1010. Alternatively, the fall detection device 100 (100A) may proceed to the termination determination process of step ST1060 (step ST1060 "Termination?").

[0021] The tipping detection device 100 (100A) then performs an immediate state detection process (immediate state detection step) (step ST1030 "immediate state detection"). In the immediate state detection process, the immediate state detection unit 130 of the tipping detection device 100 (100A) detects the motion state of the object after it has tipped over, if the tipping determination unit 120 has detected that the object has tipped over.

[0022] The fall detection device 100 (100A) then performs a comprehensive judgment process (comprehensive judgment step) (step ST1040 "Not applicable?"). In the comprehensive judgment process, the comprehensive judgment unit 140 of the fall detection device 100 (100A) determines whether the fall is that of a living person or a non-living person, based on the movement state detected by the immediate post-state detection unit 130.

[0023] If the fall detection device 100 (100A) determines in the comprehensive judgment process of step ST1040 that the fallen object is not a non-detectable object (i.e., it determines that the object is a detectable object) (step ST1040 "Non-detectable?" "NO"), it then performs a fall information output process (step ST1050 "Fall Information Output"). In the fall information output process, the comprehensive judgment unit 140 of the fall detection device 100 (100A) outputs fall information, which includes information identifying the fallen object, to an output device outside the fall detection device 100 (100A) as a result of the judgment. The fall information may include the probability of falling for each type of object, for example, information such as "person" - "X%", "object" - "Y%", or both. This allows external output devices to notify the user of the probability of each type of object tipping over (including alarms, image output, sound output, etc.), providing the user with confidence in the determination of whether the detected object has tipped over.

[0024] If the fall detection device 100 (100A) determines in the overall determination process of step ST1040 that the fallen object is not a target for detection (if it is determined that the object is not a target for detection) (step ST1040 "Not a target?" "YES"), or after executing the fall information output process of step ST1050, it then executes the termination determination process (step ST1060 "Termination?"). In the termination determination process, a control unit (not shown) of the fall detection device 100 (100A) determines whether to terminate the processing of the fall detection device 100 (100A). The control unit (not shown) determines whether to terminate the processing of the fall detection device 100 (100A) according to, for example, an external termination command or execution program.

[0025] If the fall detection device 100 (100A) determines in the termination determination process of step ST1060 that it is not going to terminate the process (step ST1060 "Termination?" "NO"), it proceeds to the process of step ST1010 and repeats from the process of step ST1010 thereafter. If the fall detection device 100 (100A) determines in the termination determination process of step ST1060 that it is going to terminate the process (step ST1060 "Termination?" "YES"), it then terminates the process ("Termination").

[0026] With the configuration described above, it becomes possible to determine falls of the target object with higher accuracy than before.

[0027] This embodiment describes a configuration including the following: [1] A fall detection device comprising: a motion component which is a signal component indicating a moving object extracted from a sensor signal, and a fall detection unit which detects the fall of an object based on the motion component; an immediate post-fall state detection unit which detects the motion state of the object after the fall when the fall detection unit detects the fall of an object; and an overall determination unit which determines whether the fall is of a living organism or a non-living organism based on the motion state detected by the immediate post-fall state detection unit. Thus, this disclosure has the effect of providing a fall detection device that can determine the fall of the object to be detected with higher accuracy than conventional devices.

[0028] This embodiment shows a configuration including the following:

[19] A fall detection method using a fall detection device, comprising: a fall determination step in which a fall determination unit of the fall detection device receives a motion component which is a signal component indicating a moving object extracted from a sensor signal and detects the fall of an object based on the motion component; an immediate post-state detection step in which, when the fall determination unit detects the fall of an object, the motion state of the object after the fall is detected; and an overall determination step in which the overall determination unit of the fall detection device determines whether the fall is of a living organism or a non-living organism based on the motion state detected by the immediate post-state detection unit. Thus, the present disclosure has the effect of providing a fall detection method that enables the detection of falls of the object to be detected with higher accuracy than conventional methods.

[0029] This embodiment further describes a configuration including the following:

[16] The overall determination unit outputs fall detection information including the probability for each type of object that is likely to have fallen as a result of the determination, characterized in that the fall detection device is according to any one of [1], [2], [3], [4], [5], [6], [7], [8], [9],

[10] ,

[11] ,

[12] ,

[13] ,

[14] , and

[15] . Thus, the present disclosure can further provide the user with an indication of the reliability of the determination result and can provide a fall detection device that can determine the fall of the object to be detected with high accuracy. Furthermore, the present disclosure can achieve the same effects as above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0030] Embodiment 2. Embodiment 1 described above provided an example of the basic configuration of the fall detection device of the present disclosure. Embodiment 2 provides a more detailed example of the configuration of the fall detection device. In Embodiment 2, components of Embodiment 2 that are the same as those of Embodiment 1 described above are denoted by the same reference numerals or similar numerals, and redundant explanations are omitted as appropriate.

[0031] Next, an example of the configuration of a fall detection system including a fall detection device according to Embodiment 2 of this disclosure will be described. Figure 3 is a diagram showing an example of the configuration of a fall detection device and a fall detection system including said fall detection device according to Embodiment 2 of this disclosure. The fall detection system including the fall detection device according to Embodiment 2 is the same as the fall detection system including the fall detection device according to Embodiment 3 or Embodiment 4 described later in the drawings. Therefore, in order to distinguish it from the configuration of the other embodiments, the fall detection system 1 according to Embodiment 2 will be described as "fall detection system 1 (1B)", and similarly, the fall detection device 100 according to Embodiment 2 will be described as "fall detection device 100 (100B)". The fall detection system 1 (1B) shown in Figure 3 is composed of a fall detection device 100 (100B), a sensor (first sensor) 300, and an output destination device 400.

[0032] The sensor (first sensor) 300 is a sensor capable of detecting moving objects in a detection target area, which is a detectable space. The sensor (first sensor) 300 is, for example, a radio wave sensor (radar) such as a millimeter wave sensor. In this case, the sensor (first sensor) 300 transmits millimeter waves toward the detection target area, which is a detectable space, receives the reflected waves, and outputs a sensor signal.

[0033] The output device 400 receives and processes the detection result from the fall detection device 100 (100B). The output device 400, for example, notifies or alarms the object that has fallen over. The output device 400, for example, notifies or alarms the object that has fallen over to the terminal devices of pre-registered users. Users include, for example, staff of a nursing home, security guards in the nursing home's security room, and family members of care recipients who are at risk of falling. Terminal devices include, for example, smartphones and tablet devices.

[0034] The fall detection device 100 (100B) according to this second embodiment detects a fall of the object to be detected. The fall detection device 100 (100B) relates to, for example, a bio-fall detection sensor using a radio wave sensor (radar) and a detection method thereof. When the sensor detects an action that may cause a fall, it determines whether the action that may cause a fall is of human origin or object origin, depending on the state immediately afterward. The fall detection device 100 (100B) shown in the referenced Figure 3 includes a dynamic reflection component extraction unit 110, a fall determination unit 120, an immediate state detection unit 130, and an overall determination unit 140.

[0035] The motion reflection component extraction unit 110 of the fall detection device 100 (100B) extracts the motion component contained in the sensor signal. The motion reflection component extraction unit 110 receives the sensor signal output by the sensor (first sensor) 300 and extracts the motion component based on the sensor signal. The motion component is a signal component that indicates a motion object that is likely to exist in the space detectable by the sensor (first sensor) 300. The motion reflection component extraction unit 110 extracts only the reflection component from the motion object from the sensor signal. For example, it performs processing such as suppressing the signal component from a target with a velocity of 0 that is present around the sensor. This can be achieved by applying existing methods such as MTI (Moving Target Indicator).

[0036] The fall detection unit 120 of the fall detection device 100 (100B), similar to the fall detection unit 120 already described, receives a motion component, which is a signal component indicating a moving object extracted from the sensor signal, and detects the fall of an object based on the motion component. Based on the motion component, the fall detection unit 120 determines that an object has fallen if the movement speed of the moving object in the space detectable by the sensor (first sensor) 300 is greater than or equal to a preset threshold. This determination may be rule-based or may use methods such as machine learning. Because this determination is a simple method using only velocity information, it can detect all actions that may cause a fall without fail. The fall detection unit 120 determines the position (point) where the fall of the object was detected and outputs fall information including that position. The position where the fall of the detected object was detected is expressed in the form of spatial coordinates in the detection target area of ​​the sensor (first sensor) 300. The method for calculating spatial coordinates can be implemented using existing methods.

[0037] The immediate post-tip state detection unit 130 of the tipping detection device 100 (100B), similar to the immediate post-tip state detection unit 130 already described, detects the motion state of an object after it has tipped over when the tipping determination unit 120 detects that the object has tipped over. When the immediate post-tip state detection unit 130 receives information of the detection target area including the position (point) where the tipping of the object was detected by the tipping determination unit 120, it detects the motion state of the moving object in the detection target area in chronological order from immediately after the tipping of the object was detected. If the tipping of multiple objects is detected, the immediate post-tip state detection unit 130 detects the motion state of each object.

[0038] Here, the motion state detected by the immediate post-tumble state detection unit 130 is, for example, a motion state obtained using the time-series changes of the velocity, distance, or both velocity and distance of an object that may have fallen over. The time-series changes in velocity and distance include information that determines the state immediately after the fall, leading to improved accuracy of the determination. For example, the immediate post-tumble state detection unit 130 acquires the time-series changes of the velocity of the target (object) immediately after the fall for, for example, 3 seconds. Alternatively, the immediate post-tumble state detection unit 130 acquires the time-series changes of the distance of the target immediately after the fall for, for example, 3 seconds. From the information acquired in this way, results such as those shown in Figures 4 and 5 can be obtained. Figure 4 is a diagram showing a first example of criteria for determining the type of object in the fall detection device of this disclosure. Figure 5 is a diagram showing a second example of criteria for determining the type of object in the fall detection device of this disclosure. Here, we will explain the case where the type of object is a person (living being) and an object (non-living being). If an object bounces or spins immediately after falling, it may exhibit a periodic change in velocity (time change of the state (velocity) of an object (non-living thing) immediately after falling, as shown by the dotted line graph in Figure 4 (2020). Alternatively, if the object rolls immediately after falling, it may exhibit a change in distance (time change of the state (distance) of an object (non-living thing) immediately after falling, as shown by the dotted line graph in Figure 5 (2120). On the other hand, since a person is likely to remain stationary immediately after falling, the time changes in velocity and distance are likely to be a change in velocity (time change of the state (velocity) of a person (living thing) immediately after falling, as shown by the solid line graph in Figure 4 (2010)), or a change in distance (time change of the state (distance) of a person (living thing) immediately after falling, as shown by the solid line graph in Figure 5 (2110).

[0039] Furthermore, if the living organism is a person, the motion state detected by the immediate post-state detection unit 130 is, for example, a movement characteristic of the human body that lasts for a predetermined period of time or longer. Specifically, this includes movements such as breathing, heart rate, arms, and legs. Complex movements of the human body that last for a certain period of time or longer are important information that distinguishes humans from objects (non-living things) and lead to higher accuracy in judgment. Complex movements of the human body that last for a certain period of time or longer include respiratory rate, heart rate (vital information), and limb movements. The methods for acquiring vital information and limb movements can be achieved using existing techniques.

[0040] Here, we will explain the difference between falls of living beings and falls of non-living beings. It is thought that immediately after a fall an object transitions to one of the following states, which are clearly different from those after a person falls: - Bouncing (if the object is elastic) - Rolling (if unstable on the surface it is placed on, or if it is cylindrical, etc.) - Spinning (if it is disc-shaped) - No vital information is detected (because it is not a living being) When a state different from that of a person falling is detected immediately after a fall, the movement that may cause a fall is determined to be of object origin. In this way, it is possible to distinguish with high accuracy whether a person or an object has fallen.

[0041] Returning to the explanation of Figure 3, the comprehensive determination unit 140 of the fall detection device 100 (100B) determines whether the detected object has fallen, similar to the comprehensive determination unit 140 already described. Based on the motion state detected by the immediate post-fall state detection unit 130, the comprehensive determination unit 140 determines, for example, whether it is a fall of a living being or a fall of a non-living being. Specifically, the comprehensive determination unit 140 determines, for example, whether it is a fall of a person or a fall of an object. In this case, a fall of a living being determined by the comprehensive determination unit 140 is a fall of a person, and a fall of a non-living being determined by the comprehensive determination unit 140 is a fall of an object. The comprehensive determination unit 140 determines that it is a fall of an object if, for example, a periodic change is obtained in the time series change of velocity, such as repeating positive and negative values. Also, the comprehensive determination unit 140 determines that it is a fall of an object if, for example, in the time series change of distance, movement of more than a predetermined value (for example, 50 cm or more) occurs immediately after the fall. Furthermore, the comprehensive determination unit 140 determines that an object has fallen if the vital information is below a predetermined value (for example, a heart rate of 0). Also, the comprehensive determination unit 140 determines that anything other than the above is a person's fall. The comprehensive determination unit 140 may use a rule-based or machine learning-based method for determining whether it is a person or an object. The comprehensive determination unit 140 outputs a determination result, similar to the comprehensive determination unit 140 in the previously described embodiment. The determination result is fall information, similar to the determination result in the previously described embodiment, for example, information that identifies the object that fell. The fall information may include the probability for each type of object that likely fell, similar to the fall information in the previously described embodiment.

[0042] In this explanation, the fall detection device 100 (100B), the sensor (first sensor) 300, and the output device 400 are shown as separate devices. However, the fall detection device 100 (100B) may be configured to include the sensor (first sensor) 300 and the output device 400.

[0043] Next, a processing example of the fall detection device according to Embodiment 2 of the present disclosure will be described. FIG. 6 is a flowchart showing an example of the processing of the fall detection device according to Embodiment 2 of the present disclosure. The processing shown in FIG. 6 is a fall detection method by the fall detection device. When the fall detection device 100 (100B) according to the present Embodiment 2 receives a fall monitoring command from, for example, a fall detection system or an external monitoring system, etc., it starts the processing shown in FIG. 6 ("Start").

[0044] The fall detection device 100 (100B) then executes a sensor signal reception process (step ST2010 "Sensor signal reception"). In the sensor signal reception process, the moving body reflection component extraction unit 110 of the fall detection device 100 (100B) receives the sensor signal output by the sensor (first sensor) 30). Specifically, the moving body reflection component extraction unit 110 receives the sensor signal output by the sensor (first sensor) 300.

[0045] The fall detection device 100 (100B) then executes a moving body reflection component extraction process (step ST2020 "Moving body reflection component extraction"). In the moving body reflection component extraction process, the moving body reflection component extraction unit 110 of the fall detection device 100 (100B) extracts a moving body component, which is a signal component indicating a moving body, based on the sensor signal.

[0046] The fall detection device 100 (100B) then executes a fall determination process (step ST2030 "Fall?"). In the fall determination process, the fall determination unit 120 of the fall detection device 100 (100B) receives the moving body component, which is a signal component indicating a moving body extracted from the sensor signal, and detects the fall of an object based on the moving body component. The fall determination unit 120 acquires the moving body reflection component for each moving body output by the moving body reflection component extraction unit 110. Specifically, for example, if the fall determination unit 120 detects a speed faster than 3 [m / s], it determines that there is a possibility of a fall.

[0047] When the fall detection device 100 (100B) determines that a fall has not been detected in the fall determination process of step ST2030 (step ST2030 "Fall?"), it repeatedly executes the process of step ST2010. Alternatively, the fall detection device 100 (100B) may proceed to the end determination process of step ST2070 (step ST2070 "End?").

[0048] Next, the fall detection device 100 (100B) executes a post - state detection process (step ST2040 "Post - state detection"). In the post - state detection process, when the post - state detection unit 130 of the fall detection device 100 (100B) detects the fall of an object by the fall determination unit 120, it detects the motion state of the object after the fall. The post - state detection unit 130 receives a moving - object component, which is a signal component indicating a moving object extracted from the sensor signal, and uses the moving - object component to calculate the motion state of the object after the fall for each fallen object, and outputs the motion state of each fallen object to the comprehensive determination unit 140.

[0049] Next, the fall detection device 100 (100B) executes a comprehensive determination process (step ST2050 "Non - target?"). In the comprehensive determination process, the comprehensive determination unit 140 of the fall detection device 100 (100B) determines whether the fall is of a living body or a non - living body based on the motion state detected by the post - state detection unit 130.

[0050] If the fall detection device 100 (100B) determines in the comprehensive judgment process of step ST2050 that the fallen object is not a non-detectable object (i.e., it determines that the object is a detectable object) (step ST2050 "Non-detectable?" "NO"), it then performs fall information output processing (step ST2060 "Fall Information Output"). In the fall information output processing, the comprehensive judgment unit 140 of the fall detection device 100 (100B) outputs fall information to the output destination device 400, which includes, for example, the location where the fall was detected, information identifying the fallen object, or both. Specifically, if the comprehensive judgment unit 140 determines that the detected object is a living organism, it outputs fall information to the output destination device 400 indicating that a fall of a living organism has been detected. The fall information may include the location where the fall was detected, information identifying the fallen living organism, etc. Specifically, if the integrated determination unit 140 determines that the detected object is a person and that it has fallen, it outputs fall information to the output device 400 indicating that a person has fallen. The fall information may include the location where the fall was detected, information identifying the person who fell, etc. Also, the fall information may include the probability (fall probability) for each type of object that is likely to have fallen, similar to the fall information in the embodiments described above. In this case, the output device 400 can, for example, notify the user of the fall probability for each type of object, and can provide the user with the reliability of the determination result of whether the detected object has fallen.

[0051] If the fall detection device 100 (100B) determines in the overall determination process of step ST2050 that the fallen object is not a target for detection (if it is determined that the object is not a target for detection) (step ST2050 "Not a target?" "YES"), or after executing the fall information output process of step ST2060, it then executes the termination determination process (step ST2070 "Termination?"). In the termination determination process, a control unit (not shown) of the fall detection device 100 (100B) determines whether to terminate the processing of the fall detection device 100 (100B). The control unit (not shown) determines whether to terminate the processing of the fall detection device 100 (100B) according to, for example, an external termination command or execution program.

[0052] If the fall detection device 100 (100B) determines in the termination determination process of step ST2070 that it is not going to terminate the process (step ST2070 "Termination?" "NO"), it proceeds to the process of step ST2010 and repeats from the process of step ST2010 thereafter. If the fall detection device 100 (100B) determines in the termination determination process of step ST2070 that it is going to terminate the process (step ST2070 "Termination?" "YES"), it then terminates the process ("Termination").

[0053] The fall detection technology disclosed herein can be applied as appropriate in situations where there is a potential for falls. For example, it can be used in elevators, nursing homes, and senior housing to detect falls, notify a management center, and allow for verification of camera footage. It can also be used in vehicles such as buses and trains, offices, bathrooms, and toilets. In other words, the fall detection technology disclosed herein is applicable to general monitoring systems.

[0054] This embodiment further describes a configuration including the following: [2] The fall detection device according to [1], further comprising: a dynamic reflection component extraction unit that receives a sensor signal output by a sensor and extracts the dynamic component (a dynamic component which is a signal component indicating a moving object that is likely to exist in the space detectable by the sensor) based on the sensor signal. This provides the effect that the present disclosure can further extract movement indicating a fall of the object to be detected from the sensor signal and use it for determination, thereby providing a fall detection device that can determine the fall of the object to be detected with higher accuracy than conventional devices. Furthermore, the present disclosure can achieve the same effect as above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0055] This embodiment further describes a configuration including the following: [3] A fall detection device according to either one of [1] or [2], characterized in that a fall of a living being determined by the comprehensive determination unit is a fall of a person, and a fall of a non-living being determined by the comprehensive determination unit is a fall of an object. Thus, this disclosure further provides a fall detection device that can determine the fall of the object to be detected with higher accuracy than conventional devices. Furthermore, this disclosure can achieve the same effect as above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0056] This embodiment further describes a configuration including the following: [4] The fall detection device according to any one of [1], [2], or [3], characterized in that the fall detection unit determines, based on the motion component, that an object has fallen if the motion speed of the motion in a space detectable by the sensor is greater than or equal to a preset threshold. This further enables the present disclosure to provide a fall detection device that performs a simple determination process using only velocity information and enables the determination of falls of the object to be detected with higher accuracy than conventional devices. Furthermore, the present disclosure achieves the same effects as described above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0057] This embodiment further describes a configuration including the following: [5] The motion state detected by the immediate post-motion state detection unit is a motion state obtained using the time-series changes of the velocity, distance, or velocity and distance of an object that may have fallen over, as described in any one of [1], [2], [3], and [4]. This provides the effect that the present disclosure can further provide a fall detection device that can determine the fall of the object to be detected with high accuracy. Furthermore, the present disclosure can achieve the same effect as described above by applying the above configuration to a system including a fall detection device or to the fall detection method described above.

[0058] This embodiment further describes a configuration including the following: [6] The motion state detected by the immediate post-motion state detection unit is a movement characteristic of the human body that lasts for a predetermined period of time or longer, characterized in that it is the fall detection device according to any one of [1], [2], [3], [4], or [5]. Thus, this disclosure has the effect of providing a fall detection device that can determine a fall of the object to be detected with high accuracy. Furthermore, this disclosure can achieve the same effect as above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0059] Embodiment 3. In the embodiments described above, a configuration was described in which the immediate post-state detection unit performs calculations using the sensor information of the entire detection area of ​​the sensor. Embodiment 3 describes an example of a configuration that enables a reduction in the amount of computation. In Embodiment 3, among the components related to Embodiment 3, components that are the same as those related to Embodiment 1 or Embodiment 2 already described are denoted by the same reference numerals or similar reference numerals, and redundant explanations are omitted as appropriate.

[0060] Next, an example of the configuration of a fall detection system including a fall detection device according to Embodiment 3 of this disclosure will be described. The fall detection system including the fall detection device according to Embodiment 3 is the same as the fall detection system including the fall detection device according to Embodiment 2 which has already been described in the drawings, so it will be described with reference to Figure 3. In the description of this embodiment, in order to distinguish it from the configuration of other embodiments, the fall detection system 1 according to Embodiment 3 will be described as "fall detection system 1 (1C)", and similarly, the fall detection device 100 according to Embodiment 3 will be described as "fall detection device 100 (100C)". The fall detection system 1 (1C) according to Embodiment 3 is configured to include a fall detection device 100 (100C), a sensor (first sensor) 300, and an output destination device 400. The sensor (first sensor) 300 is configured in the same way as the sensor (first sensor) 300 which has already been described. The output destination device 400 is configured in the same way as the output destination device 400 which has already been described.

[0061] The fall detection device 100 (100C) according to this third embodiment differs from the fall detection device 100 according to the previously described embodiment in that it sets a detection target area centered on the position where a fall is detected and performs a determination process to determine whether an object that may fall is the target of detection. The fall detection device 100 (100C) is composed of a motion reflection component extraction unit 110, a fall determination unit 120, an immediate post-fall state detection unit 130, and an overall determination unit 140.

[0062] The motion reflection component extraction unit 110 of the fall detection device 100 (100C) receives a sensor signal output by the sensor (first sensor) 300, similar to the motion reflection component extraction unit 110 already described, and extracts the motion component based on the sensor signal.

[0063] The tipping detection unit 120 of the tipping detection device 100 (100C), similar to the tipping detection unit 120 already described, receives a motion component, which is a signal component indicating a moving object extracted from the sensor signal, and detects the tipping of an object based on the motion component. Furthermore, the tipping detection unit 120 determines a processing area of ​​a three-dimensional shape based on the point where the tipping of the object was detected. The determined processing area is also called the sensing area. Figure 7 is an image diagram showing an example of a sensing area set in the tipping detection device according to Embodiment 3 of this disclosure. The tipping determination position 3000 shown in Figure 7 is the position (spatial coordinate) in the detection area of ​​the detected object that the tipping detection unit 120 determined to have tipped over. The method for calculating the point where a potentially tipping motion is detected (for example, the point where a speed faster than 3 [m / s] is detected) can be realized by using existing methods. Examples of three-dimensional shapes include a rectangular prism, a sphere, and a cylinder. The sensing area 3010 shown in Figure 7 is a rectangular parallelepiped processing area with three sides of 1 m, centered on the position within the detection target area. By limiting the processing area to only the minimum necessary range in the subsequent immediate state detection unit 130, the computational load can be reduced.

[0064] The immediate post-tip state detection unit 130 of the tipping detection device 100 (100C), similar to the immediate post-tip state detection unit 130 already described, detects the motion state of the object after it has tipped over when the tipping determination unit 120 detects that the object has tipped over. Furthermore, the immediate post-tip state detection unit 130 performs processing on the processing target area within the space detectable by the sensor and detects the motion state of the object after it has tipped over.

[0065] The comprehensive determination unit 140 of the fall detection device 100 (100C), similar to the comprehensive determination unit 140 already described, determines, for example, whether it is a fall of a living person or a fall of a non-living person, based on the movement state detected by the immediate post-state detection unit 130.

[0066] Next, an example of processing of a fall detection device according to Embodiment 3 of the present disclosure will be described. Figure 8 is a flowchart showing an example of processing of a fall detection device according to Embodiment 3 of the present disclosure. The processing shown in Figure 8 is a fall detection method by the fall detection device. When the fall detection device 100 (100C) according to Embodiment 3 receives a monitoring command from, for example, an external monitoring system, it starts the processing shown in Figure 8 ("start"). The fall detection device 100 (100C) then performs a sensor signal reception process (step ST3010 "sensor signal reception"). In the sensor signal reception process, the motion reflection component extraction unit 110 of the fall detection device 100 (100C) receives a sensor signal output by the sensor (first sensor) 300, similar to the sensor signal reception process already described.

[0067] The fall detection device 100 (100C) then performs a motion reflection component extraction process (step ST3020 "motion reflection component extraction"). In the motion reflection component extraction process, the motion reflection component extraction unit 110 of the fall detection device 100 (100C) extracts the motion component based on the sensor signal, similar to the motion reflection component extraction process described earlier.

[0068] The fall detection device 100 (100C) then performs a fall determination process (step ST3030 "Fall?"). In the fall determination process, the fall determination unit 120 of the fall detection device 100 (100C) receives a motion component, which is a signal component indicating a moving object extracted from the sensor signal, similar to the fall determination process already described, and detects the fall of the object based on the motion component. Specifically, the fall determination unit 120 acquires the motion reflection component for each moving object output by the motion reflection component extraction unit 110.

[0069] If the fall detection device 100 (100C) determines in the fall determination process of step ST3030 that it has not detected a fall (step ST3030 "Fall?"), it repeats the process of step ST3010. Alternatively, the fall detection device 100 (100C) may proceed to the termination determination process of step ST3080 (step ST3080 "Termination?").

[0070] The fall detection device 100 (100C) then performs a sensing area setting process (step ST3040 "sensing area setting"). In the sensing area setting process, the fall determination unit 120 of the fall detection device 100 (100C) determines a three-dimensional processing target area (sensing area) based on the point where the fall of the object was detected.

[0071] The fall detection device 100 (100C) then performs immediate state detection processing (step ST3050 "immediate state detection"). In the immediate state detection processing, the immediate state detection unit 130 of the fall detection device 100 (100C) performs processing on the processing target area within the space detectable by the sensor. The immediate state detection unit 130 detects the motion state of the object after it has fallen and outputs the motion state for each object to the comprehensive determination unit 140. The processing on the processing target area in the immediate state detection unit 130 is the same as the immediate state detection processing already described. As a result, since processing is performed only on the processing target area out of the entire detection target area of ​​the sensor (first sensor) 300, the amount of computation can be reduced.

[0072] The fall detection device 100 (100C) then performs an overall determination process (step ST3060 "Not applicable?"). In the overall determination process, the overall determination unit 140 of the fall detection device 100 (100C) determines whether the fall is that of a living person or a non-living person, based on the movement state detected by the immediate post-state detection unit 130, similar to the overall determination process already described.

[0073] If the fall detection device 100 (100C) determines in the comprehensive judgment process of step ST3060 that the fallen object is not a non-detectable object (i.e., it determines that the object is a detectable object) (step ST3060 "Non-detectable?" "NO"), it then performs a fall information output process (step ST3070 "Fall Information Output"). In the fall information output process, the comprehensive judgment unit 140 of the fall detection device 100 (100C), similar to the fall information output process already described, outputs fall information, including information identifying the fallen object, as a result of the judgment, to the output destination device 400.

[0074] If the fall detection device 100 (100C) determines in the overall determination process of step ST3060 that the fallen object is not a target for detection (if it is determined that the object is not a target for detection) (step ST3060 "Not a target?" "YES"), or after executing the fall information output process of step ST3070, it then executes the termination determination process (step ST3080 "Termination?"). In the termination determination process, a control unit (not shown) of the fall detection device 100 (100C) determines whether to terminate the processing of the fall detection device 100 (100C). The control unit (not shown) determines whether to terminate the processing of the fall detection device 100 (100C) according to, for example, an external termination command or execution program.

[0075] If the fall detection device 100 (100C) determines in the termination determination process of step ST3080 that it is not going to terminate the process (step ST3080 "Termination?" "NO"), it proceeds to the process of step ST3010 and repeats from the process of step ST3010 thereafter. If the fall detection device 100 (100C) determines in the termination determination process of step ST3080 that it is going to terminate the process (step ST3080 "Termination?" "YES"), it then terminates the process ("Termination").

[0076] The fall detection device according to this embodiment can reduce the amount of computation by limiting the processing area of ​​the immediate state detection unit 130 to only the minimum necessary range.

[0077] This embodiment further describes a configuration including the following: [7] The fall detection device according to any one of [1], [2], [3], [4], [5], or [6], characterized in that the fall detection unit determines a three-dimensional processing target area (sensing area) based on the point where the fall of the object is detected, and the immediate post-fall state detection unit performs processing on the processing target area in the space detectable by the sensor to detect the motion state of the object after it has fallen. Thus, this disclosure has the effect of providing a fall detection device that can further reduce the amount of computation. Furthermore, this disclosure can achieve the same effect as above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0078] Embodiment 4. In the embodiments described above, a configuration was described in which a single motion state is used to determine whether an object that may have fallen over is the target of detection. Embodiment 4 describes an example of a configuration in which multiple motion states are used to detect the falling of the target of detection. In Embodiment 4, among the components related to Embodiment 4, components that are the same as those related to Embodiment 1, Embodiment 2, or Embodiment 3 already described are denoted by the same reference numerals or similar reference numerals, and redundant explanations are omitted as appropriate.

[0079] Next, an example of the configuration of a fall detection system including a fall detection device according to Embodiment 4 of this disclosure will be described. The fall detection system including the fall detection device according to Embodiment 4 is the same as the fall detection system including the fall detection device according to Embodiment 2 which has already been described in the drawings, so it will be described with reference to Figure 3. In the description of this embodiment, in order to distinguish it from the configuration of other embodiments, the fall detection system 1 according to Embodiment 4 will be described as "fall detection system 1 (1D)", and similarly, the fall detection device 100 according to Embodiment 4 will be described as "fall detection device 100 (100D)". The fall detection device 100 (100D) is composed of a sensor (first sensor) 300 and an output destination device 400. The sensor (first sensor) 300 is configured in the same way as the sensor (first sensor) 300 which has already been described. The output destination device 400 is configured in the same way as the output destination device 400 which has already been described.

[0080] The fall detection device 100 (100D) according to this fourth embodiment differs from the fall detection device 100 according to the previously described embodiment in that it detects falls of the target using multiple types of motion states. The fall detection device 100 (100D) is composed of a motion reflection component extraction unit 110, a fall determination unit 120, an immediate post-fall state detection unit 130, and an overall determination unit 140.

[0081] The motion reflection component extraction unit 110 of the fall detection device 100 (100D), similar to the motion reflection component extraction unit 110 already described, receives a sensor signal output by the sensor (first sensor) 300 and extracts the motion component based on the sensor signal.

[0082] The tipping detection unit 120 of the tipping detection device 100 (100D), similar to the tipping detection unit 120 already described, receives a motion component, which is a signal component indicating a moving object extracted from the sensor signal, and detects the tipping of the object based on this motion component.

[0083] The immediate post-tip state detection unit 130 of the fall detection device 100 (100D), similar to the immediate post-tip state detection unit 130 already described, detects the motion state of the object after it has fallen when the fall determination unit 120 detects that the object has fallen. The immediate post-tip state detection unit 130 according to this fourth embodiment detects multiple types of motion states. For example, the immediate post-tip state detection unit 130 detects the motion state using the time-series change in the velocity of the object after it has fallen, and also detects the motion state of the object using the time-series change in distance. Furthermore, the immediate post-tip state detection unit 130 detects the motion state of the object based on movements characteristic of the human body, such as breathing, heart rate, arms, and legs.

[0084] The comprehensive determination unit 140 of the fall detection device 100 (100D) determines whether the detected object has fallen, similar to the comprehensive determination unit 140 already described. The comprehensive determination unit 140 determines, for example, whether it is a fall of a living being or a fall of a non-living being, based on the motion state detected by the immediate post-state detection unit 130. The comprehensive determination unit 140 according to this fourth embodiment determines whether the detected object has fallen based on a plurality of types of motion states. Specifically, the comprehensive determination unit 140 determines whether it is a fall of a living being or a fall of a non-living being, based on a plurality of types of motion states. For example, the comprehensive determination unit 140 determines whether it is a fall of the detected object by combining the motion state based on the time-series change of velocity and the motion state based on the time-series change of distance. In addition, the comprehensive determination unit 140 determines whether it is a fall of the detected object by combining the motion states of an object based on movements characteristic of the human body, such as breathing, heart rate, arms, and legs. The comprehensive determination unit 140 outputs a determination result, similar to the comprehensive determination unit 140 in the previously described embodiment. The determination result is, similar to the determination result in the embodiments already described, tipping information that includes, for example, information identifying the tipping object. The tipping information may also include the probability for each type of object that would have tipped over, similar to the tipping information in the embodiments already described.

[0085] Next, an example of the processing of a fall detection device according to Embodiment 4 of this disclosure will be described. The processing of the fall detection device 100 (100D) according to Embodiment 4 differs from the comprehensive determination processing already described. When the fall detection device 100 (100D) performs the immediate state detection processing (for example, step ST2040 "immediate state detection" shown in Figure 6), it then performs the comprehensive determination processing (for example, step ST2050 "not applicable?" shown in Figure 6). In the comprehensive determination processing, the comprehensive determination unit 140 of the fall detection device 100 (100D) calculates multiple types of time-series changes based on the motion state detected by the immediate state detection unit 130, and uses these time-series changes to determine whether it is a fall of a living being or a fall of a non-living being. For example, the comprehensive determination unit 140 uses a combination of the motion state based on the time-series change of velocity and the motion state based on the time-series change of distance to determine whether it is a fall of the object being detected. Furthermore, the comprehensive determination unit 140 determines whether the detected object has fallen by combining the aforementioned motion state of the object based on each of the movements characteristic of the human body, such as breathing (vital information), heart rate (vital information), arms, legs, etc. A specific example of the determination method is shown below. Various types of information can be combined as appropriate. The determination may be rule-based or based on machine learning. ・If periodic changes in velocity are obtained in the time series of velocity changes, such as repeating positive and negative values, the movement that may cause a fall is of object origin. ・If distance changes are obtained in the time series of distance changes, such as repeating approaches and moves away, the movement that may cause a fall is of object origin. ・If movement of a predetermined value or more (e.g., 50 cm or more) occurs in the time series of distance changes, the movement that may cause a fall is of object origin. ・If, for example, the respiratory rate is detected at a typical adult value (12-20 breaths / min) from the vital information, the heart rate is detected at a typical adult value (60-100 beats), or both, the movement that may cause a fall is of human origin.

[0086] By combining various pieces of information using the configuration described above, it is possible to determine with high accuracy whether a movement that could cause a fall is human-induced or object-induced.

[0087] This embodiment further describes a configuration including the following: [8] The fall detection device according to any one of [1], [2], [3], [4], [5], [6], and [7], characterized in that the comprehensive determination unit determines whether the fall is that of a living person or a non-living person based on a plurality of types of movement states. Thus, the present disclosure has the effect of providing a fall detection device that can determine the fall of the object to be detected with high accuracy. Furthermore, the present disclosure can achieve the same effect as above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0088] Embodiment 5. In the embodiments described above, we have described forms in which the placement position of the sensors and the output destination device in the fall detection system may be arbitrary. Embodiment 5 describes a specific configuration example of the placement position of the sensors and the output destination device. In Embodiment 5, among the components related to Embodiment 5, components that are the same as those related to Embodiments 1, 2, 3, or 4 already described are denoted by the same reference numerals or similar reference numerals, and redundant explanations are omitted as appropriate.

[0089] Next, an example of the configuration of a fall detection system including a fall detection device according to Embodiment 5 of the present disclosure will be described. Figure 9 is a diagram showing an example of the configuration of a fall detection device and a fall detection system including the fall detection device according to Embodiment 5 of the present disclosure. The fall detection system 1 (1E) according to Embodiment 5 differs from the fall detection system 1 according to the previously described embodiment in that the arrangement of the sensor (first sensor) 300 and the function of the output destination device 400 are specified. The fall detection system 1 (1E) shown in Figure 9 is configured to include a fall detection device 100 (100E), a sensor (first sensor) 300, and an output destination device 400.

[0090] The sensor (first sensor) 300 according to this embodiment 5 is assumed to be installed above the detection target area, which is the area where an object that may fall over is likely to exist. Specifically, the sensor (first sensor) 300 is installed at a position higher than, for example, the height of an adult's head. For example, the radar is installed at a height of the average adult male's height (1.7 [m]) or higher. This is because a fall is thought to involve a movement in the vertical downward direction. This makes it possible to measure the movement during a fall with high accuracy. The sensor (first sensor) 300 outputs a sensor signal to the fall detection device 100 (100E).

[0091] The output destination device 400 according to this embodiment 5 includes a control device 410. The control device 410 outputs video of the space where an object has fallen over, as detected by the fall detection device 100 (100E), to a terminal device of a pre-registered user. Furthermore, if the fall of the object is a fall that is subject to detection, the output destination device 400 notifies the pre-registered user's terminal device of an alarm in accordance with the command of the fall detection device 100 (100E). That is, in accordance with the command of the comprehensive determination unit 140, if the fall of an object is detected by the fall determination unit 120, the control device 410 outputs video of the space where the object has fallen over to a pre-registered user's terminal device, and if a fall of a living person is detected, it notifies the pre-registered user's terminal device of an alarm. Users include, for example, staff of a nursing home, security guards in the nursing home's security room, and family members of care recipients who are at risk of falling. Terminal devices include, for example, smartphones and tablet devices. This allows for visual confirmation via video if a fall occurs. Furthermore, by issuing an alarm, it encourages quick rescue efforts in the event of a fall.

[0092] The fall detection device 100 (100E) according to this embodiment 5 is applied to a fall detection system that issues an alarm when a fall of the target object is detected. The fall detection device 100 (100E) is composed of a motion reflection component extraction unit 110, a fall determination unit 120, an immediate post-fall state detection unit 130, and an overall determination unit 140.

[0093] The motion reflection component extraction unit 110 of the fall detection device 100 (100E), similar to the motion reflection component extraction unit 110 already described, receives the sensor signal output by the sensor (first sensor) 300 and extracts a motion component, which is a signal component indicating a moving object that is likely to be present in the detection target area, which is a space detectable by the sensor (first sensor) 300, based on the sensor signal.

[0094] The fall detection unit 120 of the fall detection device 100 (100E), similar to the fall detection unit 120 already described, receives a motion component, which is a signal component indicating a moving object extracted from the sensor signal, and detects the fall of an object based on the motion component. The fall detection unit 120 detects the fall of an object using the motion component based on the sensor signal output by the sensor (first sensor) 300 installed above the detection target area, which is the area where an object that may fall may exist.

[0095] The immediate post-tip state detection unit 130 of the tipping detection device 100 (100E) detects the motion state of the object after it has tipped over, similar to the function of the immediate post-tip state detection unit 130 already described, when the tipping determination unit 120 detects that the object has tipped over.

[0096] The comprehensive determination unit 140 of the fall detection device 100 (100E) has the following functions in addition to the functions of the embodiment already described. When the detection target is a fall of a living being, if the fall of an object is detected by the fall determination unit 120, the comprehensive determination unit 140 commands an external control device to output video of the space in which the fall of the object was detected to a terminal device of a user that has been registered in advance. If a fall of a living being is detected, the comprehensive determination unit 140 further commands the control device 410 to send an alarm to a terminal device of a user that has been registered in advance.

[0097] If the detected object is a person falling, the comprehensive determination unit 140, if the fall determination unit 120 detects that an object has fallen, commands the control device 410 to output video of the space where the object fell to a pre-registered user's terminal device. If a person falls, the comprehensive determination unit 140 further commands the control device 410 to send an alarm to a pre-registered user's terminal device.

[0098] The comprehensive determination unit 140 determines whether the fall is that of a living being or a non-living being based on the motion state detected by the immediate post-state detection unit 130. If the fall determination unit 120 detects that an object has fallen, the comprehensive determination unit 140 instructs the control device 410 to output video of the space where the object's fall was detected to a pre-registered user's terminal device. If the fall of a living being is detected, the comprehensive determination unit 140 further instructs the control device 410 to send an alarm to a pre-registered user's terminal device.

[0099] Next, an example of processing of a fall detection device according to Embodiment 5 of this disclosure will be described. Figure 10 is a flowchart showing an example of processing by a control device (or fall detection device) in a fall detection system according to Embodiment 5 of this disclosure. The processing shown in Figure 10 is a fall detection method by the control device (or fall detection device) of the fall detection system. The processing by the fall detection device 100 (100E) according to Embodiment 5 differs from the processing already described in that it specifies the details of the output destination in the fall information output processing. If the fall detection device 100 (100E) determines in the comprehensive judgment processing already described (see, for example, step ST2050 shown in Figure 6) that the fallen object is not a non-detection target (if it is determined that the object is a detection target) (see, for example, step ST2050 "Non-target?" "NO" shown in Figure 6), it then executes fall information output processing (see, for example, step ST2060 "Fall Information Output" shown in Figure 6). In the fall information output processing, the comprehensive determination unit 140 of the fall detection device 100 (100E) determines whether the fall is that of a living being or a non-living being based on the motion state detected by the immediate post-state detection unit 130. Based on the determination result, the comprehensive determination unit 140 generates fall information and outputs the fall information to the control device 410 of the output destination device 400. The comprehensive determination unit 140 may also output fall information to the control device 410 of the output destination device 400, including the probability of fall for each type of object that is likely to have fallen, similar to the fall information in the embodiment described above.

[0100] Next, the processing of the control device 410 will be described. When the control device 410 receives, for example, tipping information output by the tipping detection device 100 (100E), it starts the processing shown in Figure 10. The control device 410 then performs tipping information acquisition processing (step ST5010 "Tipping Information Acquisition"). In the tipping information acquisition processing, the control device 410 acquires the tipping information output by the tipping detection device 100 (100E). The control device 410 then performs alarm command processing (step ST5020 "Alarm Command"). In the alarm command processing, the control device 410 commands an alarm device (not shown) to issue an alarm according to the tipping information. If the tipping information includes the tipping probability for each type of object, the control device 410 may, in the alarm command processing, command an alarm device (not shown) to issue an alarm according to the tipping probability for each type of object. In this case, for example, alarms can be issued in stages according to the probability of whether the object is living or non-living. Furthermore, the control device 410, which is the output destination device 400, can, for example, notify the user of the probability of each type of object falling over, and can provide the user with the reliability of the determination result of whether the detected object has fallen over. An alarm device (not shown) transmits an alarm to, for example, a pre-registered user's terminal device, that the detected object has fallen over, in accordance with a command. The alarm may include, for example, the location where the fall was detected, information identifying the detected object (e.g., name), the state of the fall, or a combination of these. After executing the alarm command processing, the control device 410 then terminates the processing shown in Figure 10 ("termination").

[0101] This embodiment further describes a configuration including the following: [9] The fall detection device according to any one of [1], [2], [3], [4], [5], [6], [7], and [8], characterized in that the fall detection unit detects the fall of an object using the motion component based on a sensor signal output by a sensor installed above the detection target area, which is an area in which an object that may fall may exist. Thus, the present disclosure has the effect of providing a fall detection device that can determine the fall of an object with high accuracy. Furthermore, the present disclosure can achieve the same effect as above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0102] This embodiment further describes a configuration including the following:

[10] The fall detection device according to any one of [1], [2], [3], [4], [5], [6], [7], [8], and [9], characterized in that when the fall detection unit detects that an object has fallen, the comprehensive determination unit instructs an external control device to output an image of the space in which the object fell to a terminal device of a pre-registered user, and when it detects that a living person has fallen, it further instructs an external control device to notify a pre-registered user's terminal device of an alarm.

[0103] This embodiment further describes a configuration including the following:

[11] The fall detection device according to [3], characterized in that when the fall detection unit detects that an object has fallen, it commands a control device to output an image of the space in which the object fell to a terminal device of a pre-registered user, and when it detects that a person has fallen, it commands a control device to send an alarm to a terminal device of a pre-registered user. Thus, the present disclosure can further provide a fall detection device that enables the user to visually confirm the situation through the image. Furthermore, the present disclosure can achieve the same effect as above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0104] This embodiment further describes a configuration including the following:

[17] A fall detection device comprising: a sensor installed above a detection target area which is a space in which an object that may fall may exist; a dynamic reflection component extraction unit which receives a sensor signal output by the sensor and extracts a dynamic component which is a signal component indicating a moving body that may exist in the detection target area (a space detectable by the sensor) based on the sensor signal; a fall determination unit which receives a dynamic component which is a signal component indicating a moving body extracted from the sensor signal and detects the fall of an object based on the dynamic component; an immediate post-fall state detection unit which detects the motion state of the object after the fall if the fall determination unit has detected the fall of an object; and a comprehensive determination unit which determines whether the fall is that of a living being or a non-living being based on the motion state detected by the immediate post-fall state detection unit; and a fall detection system comprising: a fall detection device as described in [2], a sensor installed above a detection target area which is a space in which an object that may fall may exist and which outputs a sensor signal to the fall detection device. Furthermore, the fall detection device included in the fall detection system may be any one of the fall detection devices from [3] to

[16] . This provides the effect that the present disclosure can further provide a fall detection system that can determine the fall of the object to be detected with high accuracy. The present disclosure also provides the same effect as above by applying the above configuration to the fall detection method.

[0105] This embodiment further describes a configuration including the following:

[18] A fall detection device having: a motion reflection component extraction unit that receives a sensor signal output by a sensor and extracts a motion component, which is a signal component indicating a motion that is likely to be present in a detection target area, which is a space detectable by the sensor, based on the sensor signal; a fall determination unit that receives a motion component, which is a signal component indicating a motion extracted from the sensor signal and detects the fall of an object based on the motion component; an immediate post-fall state detection unit that detects the motion state of the object after the fall when the fall determination unit detects the fall of an object; and an overall determination unit that determines whether the fall is that of a living being or a non-living being based on the motion state detected by the immediate post-fall state detection unit, wherein when the fall determination unit detects the fall of an object, the overall determination unit commands the control device to output an image of the space in which the fall of the object was detected to a terminal device of a pre-registered user, and when the fall of a living being is detected, the overall determination unit further commands the control device to send an alarm to a terminal device of a pre-registered user; A fall detection system comprising: a control device that, in accordance with a command from the comprehensive determination unit, outputs video of the space where the fall of an object was detected to a pre-registered user's terminal device when the fall of an object is detected by the fall determination unit, and when the fall of a living person is detected, an alarm is sent to a pre-registered user's terminal device. Thus, this disclosure further provides a fall detection system that enables highly accurate determination of object falls. Furthermore, this disclosure achieves the same effect as above by applying the above configuration to the fall detection method.

[0106] Embodiment 6. In the embodiments described above, a configuration was described in which only the sensor signal of the sensor (first sensor) is used. Embodiment 6 describes a configuration example in which the sensor signal used for fall detection and the sensor signal used for immediate state detection are different. In Embodiment 6, among the components related to Embodiment 6, components that are the same as those related to Embodiments 1, 2, 3, 4, or 5 already described are denoted by the same reference numerals or similar reference numerals, and redundant explanations are omitted as appropriate.

[0107] Next, an example of the configuration of a target tracking system including a fall detection device according to Embodiment 6 of this disclosure will be described. Figure 11 is a diagram showing an example of the configuration of a fall detection device according to Embodiment 6 of this disclosure. The fall detection system 1 (1F) according to Embodiment 6 differs from the fall detection system 1 according to the previously described embodiment in that it is equipped with multiple sensors and determines whether the target has fallen using sensor signals output by the multiple sensors. The fall detection system 1 (1F) shown in Figure 11 is composed of a fall detection device 100 (100F), a sensor (first sensor) 300, an output destination device 400, and a sensor (second sensor) 500. The sensor (first sensor) 300 is configured in the same way as the sensor (first sensor) 300 previously described. The output destination device 400 is configured in the same way as the output destination device 400 previously described.

[0108] The sensor (second sensor) 500 is a different type of sensor from the sensor (first sensor) 300. If the sensor (first sensor) 300 is a millimeter-wave radar, then the sensor (second sensor) 500 may include, for example, an image sensor, a temperature sensor, or a sound sensor. The sensor (second sensor) 500 only needs to consist of at least one sensor. The sensor (second sensor) 500 may also be composed of a combination of multiple sensors.

[0109] The sensor (second sensor) 500 is, for example, an image sensor. If the sensor (second sensor) 500 is an image sensor, it acquires an image of the area where a fall may occur and is used to determine whether there is a living being (person) in a fallen state within that image. Motion (falling) is detected using the sensor (first sensor) 300, and the determination of whether the fallen object is a living being (person) or a non-living object is made using the image, which allows for more accurate determination.

[0110] The sensor (second sensor) 500 is, for example, a temperature sensor. If the sensor (second sensor) 500 is a temperature sensor, it acquires temperature information in the range where an action that could cause a fall occurs, and if it detects a temperature that could be body temperature (for example, 35°C or higher), it determines that it is a living organism (person). Motion (falling) is detected using the sensor (first sensor) 300, and the determination of whether what fell is a person or an object is made using temperature information, which enables more accurate determination.

[0111] The aforementioned sensor (second sensor) 500 is, for example, a sound sensor. If the sensor (second sensor) 500 is a sound sensor, it acquires sound within the range where an action that could cause a fall occurs, and if it detects a person's groan or the sound of clothes rustling, it determines that it is a person. Motion (falling) is detected by the sensor (first sensor) 300, and the determination of whether what fell is a person or an object is made by sound, thereby enabling more accurate determination.

[0112] The fall detection device 100 (100F) according to this sixth embodiment differs from the fall detection device 100 according to the previously described embodiment in that the immediate post-fall state detection unit 130 detects the motion state of an object after it has fallen using a second sensor signal output by a second sensor. The fall detection device 100 (100F) is configured to include a motion reflection component extraction unit 110, a fall determination unit 120, an immediate post-fall state detection unit 130, and an overall determination unit 140.

[0113] The motion reflection component extraction unit 110 of the fall detection device 100 (100F), similar to the motion reflection component extraction unit 110 already described, receives a sensor signal output by the sensor (first sensor) 300 and extracts a motion component, which is a signal component indicating a moving object that is likely to be present in the detection target area, which is a space detectable by the sensor (first sensor) 300, based on the sensor signal.

[0114] The tipping detection unit 120 of the tipping detection device 100 (100F), similar to the tipping detection unit 120 already described, receives a motion component, which is a signal component indicating a moving object extracted from the sensor signal, and detects the tipping of the object based on the motion component.

[0115] The immediate post-tip state detection unit 130 of the fall detection device 100 (100F) has the following functions in addition to the functions of the immediate post-tip state detection unit 130 already described. The immediate post-tip state detection unit 130 detects the motion state of an object after it has fallen using a second sensor signal output by a second sensor, which is a different type of sensor from the sensor that output the sensor signal used to extract the motion component.

[0116] The comprehensive determination unit 140 of the fall detection device 100 (100F) determines whether the detected object has fallen, similar to the comprehensive determination unit 140 already described. Based on the motion state detected by the immediate post-state detection unit 130, the comprehensive determination unit 140 determines, for example, whether it is a fall of a living being or a fall of a non-living being. The comprehensive determination unit 140 outputs a determination result, similar to the comprehensive determination unit 140 in the previously described embodiment. The determination result is fall information, similar to the determination result in the previously described embodiment, including, for example, information identifying the object that fell. The fall information may include the probability for each type of object that likely fell, similar to the fall information in the previously described embodiment.

[0117] Next, an example of processing of a fall detection device according to Embodiment 6 of the present disclosure will be described. Figure 12 is a flowchart showing an example of processing of a fall detection device according to Embodiment 6 of the present disclosure. The processing shown in Figure 12 is a fall detection method by the fall detection device. When the fall detection device 100 (100F) according to Embodiment 6 receives a fall monitoring command from a fall detection system or an external monitoring system, it starts the processing shown in Figure 12 ("start").

[0118] The fall detection device 100 (100F) then performs a first sensor signal reception process (step ST6010 "receiving a sensor signal from the first sensor"). In the first sensor signal reception process, the motion reflection component extraction unit 110 of the fall detection device 100 (100F) receives the sensor signal output by the sensor (first sensor) 300. Specifically, the motion reflection component extraction unit 110 receives the sensor signal output by the first sensor.

[0119] The fall detection device 100 (100F) then performs a motion reflection component extraction process (step ST6020 "motion reflection component extraction"). In the motion reflection component extraction process, the motion reflection component extraction unit 110 of the fall detection device 100 (100F) extracts the motion component based on the sensor signal. The motion reflection component extraction unit 110 outputs the extracted motion component to the fall determination unit 120.

[0120] The fall detection device 100 (100F) then performs a fall determination process (step ST6030 "Fall?"). In the fall determination process, the fall determination unit 120 of the fall detection device 100 (100F) receives a motion component, which is a signal component indicating a moving object extracted from the sensor signal, and detects the fall of the object based on the motion component. The fall determination unit 120 acquires the motion component (motion reflection component) for each moving object output by the motion reflection component extraction unit 110. Using the motion component, the fall determination unit 120 determines that there is a possibility of a fall if it detects that the moving object is moving at a speed faster than, for example, 3 [m / s] (threshold).

[0121] If the fall detection device 100 (100F) determines in the fall determination process of step ST2030 that it has not detected a fall (step ST6030 "Fall?"), it repeatedly executes the process of step ST6010. Alternatively, the fall detection device 100 (100F) may proceed to the termination determination process of step ST6070 (step ST6070 "Termination?").

[0122] The fall detection device 100 (100F) then performs a second sensor signal reception process (step ST6040 "receive sensor signal from second sensor"). In the second sensor signal reception process, the motion reflection component extraction unit 110 of the fall detection device 100 (100F) receives the sensor signal output by the sensor (second sensor) 500.

[0123] The fall detection device 100 (100F) then performs immediate state detection processing (step ST6050 "immediate state detection"). In the immediate state detection processing, the immediate state detection unit 130 of the fall detection device 100 (100F) detects the motion state of the object after it has fallen if the fall determination unit 120 has detected that the object has fallen. The immediate state detection unit 130 uses the sensor signal received from the sensor (second sensor) 500 to calculate the motion state for each fallen object and outputs the motion state for each fallen object to the comprehensive determination unit 140.

[0124] The fall detection device 100 (100F) then performs an overall determination process (step ST6060 "Not applicable?"). In the overall determination process, the overall determination unit 140 of the fall detection device 100 (100F) determines whether the fall is that of a living person or a non-living person, based on the movement state detected by the immediate post-state detection unit 130.

[0125] If the fall detection device 100 (100F) determines in the comprehensive judgment process of step ST6060 that the fallen object is not a non-detectable object (i.e., it determines that the object is a detectable object) (step ST6060 "Non-detectable?" "NO"), it then performs a fall information output process (step ST6070 "Fall Information Output"). In the fall information output process, the comprehensive judgment unit 140 of the fall detection device 100 (100F) outputs fall information to the output destination device 400, which includes, for example, the location where the fall was detected, information identifying the fallen object, or both. Specifically, if the comprehensive judgment unit 140 determines that the detected object is a living organism, it outputs fall information to the output destination device 400 indicating that a fall of a living organism has been detected. The fall information may include the location where the fall was detected, information identifying the fallen living organism, etc. Specifically, if the integrated determination unit 140 determines that the detected object is a person and that it has fallen, it outputs fall information to the output device 400 indicating that a person has fallen. The fall information may include the location where the fall was detected, information identifying the person who fell, etc. Also, the fall information may include the probability of each type of object that likely fell, similar to the fall information in the embodiment described above. In this case, the output device 400 can, for example, notify the user of the probability of each type of object that likely fell, thereby providing the user with the reliability of the determination result regarding whether the detected object fell.

[0126] The fall detection device 100 (100F) determines in the comprehensive judgment process of step ST6060 that the fallen object is not a target for detection (it is determined that the object is not a target for detection) (step ST6060 "Not a target?" "YES"), or after executing the fall information output process of step ST6070, it then executes the termination judgment process (step ST6080 "Termination?"). In the termination judgment process, a control unit (not shown) of the fall detection device 100 (100F) determines whether to terminate the processing of the fall detection device 100 (100F). The control unit (not shown) determines whether to terminate the processing of the fall detection device 100 (100F) according to, for example, an external termination command or execution program.

[0127] If the fall detection device 100 (100F) determines in the termination determination process of step ST6080 that it is not going to terminate the process (step ST6080 "Termination?" "NO"), it proceeds to the process of step ST6010 and repeats from the process of step ST6010 thereafter. If the fall detection device 100 (100F) determines in the termination determination process of step ST6080 that it is going to terminate the process (step ST6080 "Termination?" "YES"), it then terminates the process ("Termination").

[0128] With the configuration described above, the system uses a combination of sensor signals from different types of sensors to determine if the target has fallen, enabling more accurate detection.

[0129] This embodiment further describes a configuration including the following:

[12] The immediate post-tipping state detection unit detects the motion state of an object after it has fallen over using a second sensor signal output by a second sensor which is a different type of sensor from the sensor that output the sensor signal used to extract the motion component, as described in any one of [1], [2], [3], [4], [5], [6], [7], [8], [9],

[10] , and

[11] . Thus, this disclosure has the effect of providing a fall detection device that can determine whether an object has fallen over with high accuracy. Furthermore, this disclosure can achieve the same effect as described above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0130] This embodiment further describes a configuration including the following:

[13] The fall detection device according to the description, characterized in that the second sensor is an image sensor. This provides the effect that the present disclosure can further provide a fall detection device that can determine whether an object has fallen with high accuracy. Furthermore, the present disclosure can achieve the same effect as above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0131] This embodiment further describes a configuration including the following:

[14] The tipping detection device is characterized in that the second sensor is a temperature sensor. This provides the effect that the present disclosure can further provide a tipping detection device that can determine whether an object has tipped over with high accuracy. Furthermore, the present disclosure can achieve the same effect as above by applying the above configuration to a system including a tipping detection device or to the above tipping detection method.

[0132] This embodiment further describes a configuration including the following:

[15] The fall detection device according to the description, characterized in that the second sensor is a sound sensor. This provides the effect that the present disclosure can further provide a fall detection device that can determine whether an object has fallen with high accuracy. Furthermore, the present disclosure can achieve the same effect as above by applying the above configuration to a system including a fall detection device or to the above fall detection method.

[0133] Here, we will describe the details of the hardware configuration for realizing the functions of the present disclosure. Figure 13 is a diagram showing a first example of a hardware configuration for realizing the functions according to the configuration of the present disclosure. Figure 14 is a diagram showing a second example of a hardware configuration for realizing the functions according to the configuration of the present disclosure. The fall detection device 100 (100A, 100B, 100C, 100D, 100E, 100F) and the control device 410 of the present disclosure are realized by the hardware shown in Figure 13 or Figure 14, respectively.

[0134] The fall detection device 100 (100A, 100B, 100C, 100D, 100E, 100F) is composed of, for example, a processor 10001, a memory 10002, an input / output interface 10003, and a communication circuit 10004, as shown in Figure 13. The processor 10001 and memory 10002 are, for example, mounted on a computer. The memory 10002 stores a program that causes the computer to function as a dynamic reflection component extraction unit 110, a fall determination unit 120, an immediate post-state detection unit 130, an overall determination unit 140, and a control unit (not shown). By the processor 10001 reading and executing the program stored in the memory 10002, the functions of the dynamic reflection component extraction unit 110, the fall determination unit 120, the immediate post-state detection unit 130, the overall determination unit 140, and the control unit (not shown) are realized. Furthermore, a storage unit (not shown) is realized by memory 10002 or other memory (not shown). Additionally, a communication unit (not shown) is realized by communication circuit 10004.

[0135] The processor 10001 uses, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a microprocessor, a microcontroller, or a DSP (Digital Signal Processor). The memory 10002 may be a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable Read Only Memory), or flash memory; it may be a magnetic disk such as a hard disk or flexible disk; it may be an optical disk such as a CD (Compact Disc) or DVD (Digital Versatile Disc); or it may be a magneto-optical disk. The processor 10001 and the memory 10002 or the communication circuit 10004 are connected in a manner that enables them to transmit data to each other. Furthermore, the processor 10001, the memory 10002, and the communication circuit 10004 are connected in a manner that allows them to mutually transmit data with other hardware via the input / output interface 10003.

[0136] Alternatively, the functions of the motion reflection component extraction unit 110, the fall determination unit 120, the immediate post-fall state detection unit 130, the overall determination unit 140, and the control unit (not shown) in the fall detection device 100 (100A, 100B, 100C, 100D, 100E, 100F) may be realized by a dedicated processing circuit 20001, as shown in Figure 14.

[0137] The processing circuit 20001 may be a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), an FPGA (Field-Programmable Gate Array), a SoC (System-on-a-Chip), or a system LSI (Large-Scale Integration), etc. A storage unit (not shown) is realized by memory 20002 or other memory (not shown). The memory 20002 may be a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable Read Only Memory), or flash memory; it may be a magnetic disk such as a hard disk or flexible disk; it may be an optical disk such as a CD (Compact Disc) or DVD (Digital Versatile Disc); or it may be a magneto-optical disk. Furthermore, a communication unit (not shown) is realized by the communication circuit 20004. The processing circuit 20001 and the memory 20002 or the communication circuit 20004 are connected in a manner that allows them to transmit data to each other. Furthermore, the processing circuit 20001, the memory 20002, and the communication circuit 20004 are connected in a manner that allows them to transmit data to each other with other hardware via the input / output interface 20003. Note that the functions of the motion reflection component extraction unit 110, the fall determination unit 120, the immediate post-fall state detection unit 130, the comprehensive determination unit 140, and the control unit (not shown) in the fall detection device 100 (100A, 100B, 100C, 100D, 100E, 100F) may be implemented by separate processing circuits, or they may be implemented together in a single processing circuit.

[0138] Alternatively, some functions of the fall detection device 100 (100A, 100B, 100C, 100D, 100E, 100F), including the motion reflection component extraction unit 110, the fall determination unit 120, the immediate post-fall state detection unit 130, the comprehensive determination unit 140, and the control unit (not shown), may be implemented by the processor 10001 and memory 10002, while the remaining functions may be implemented by the processing circuit 20001.

[0139] Within the scope of this disclosure, it is possible to freely combine the embodiments, modify any component of each embodiment, or omit any component of each embodiment.

[0140] This disclosure is suitable for use in fall detection devices or systems including fall detection devices, as it can determine falls of the target object with higher accuracy than conventional methods.

[0141] 1 (1B, 1C, 1D, 1E, 1F) Fall detection system, 100 (100A, 100B, 100C, 100D, 100E, 100F) Fall detection device, 110 Dynamic reflection component extraction unit, 120 Fall judgment unit, 130 Immediate state detection unit, 140 Comprehensive judgment unit, 300 Sensor (first sensor), 400 Output destination device, 410 Control device, 500 Sensor (second sensor), 2010 Time change of immediate state (velocity) of a person (living organism) after falling, 2020 Time change of immediate state (velocity) of an object (non-living organism) after falling, 2110 Time change of immediate state (distance) of a person (living organism) after falling, 2120 Time change of immediate state (distance) of an object (non-living organism) after falling, 3000 Fall judgment position, 3010 Sensing area, 10001 Processor, 10002 memory, 10003 input / output interface, 10004 communication circuit, 20001 processing circuit, 20002 memory, 20003 input / output interface, 20004 communication circuit.

Claims

1. A fall detection device comprising: a motion component which is a signal component indicating a moving object extracted from a sensor signal, and a fall detection unit which detects the fall of an object based on the motion component; an immediate post-fall state detection unit which detects the motion state of the object after the fall if the fall detection unit has detected the fall of an object; and an overall determination unit which determines whether the fall is of a living organism or a non-living organism based on the motion state detected by the immediate post-fall state detection unit.

2. The fall detection device according to claim 1, further comprising a dynamic reflection component extraction unit that receives a sensor signal output by a sensor and extracts the dynamic component based on the sensor signal.

3. The fall detection device according to claim 1 or 2, characterized in that the fall of a living being determined by the comprehensive determination unit is a fall of a person, and the fall of a non-living being determined by the comprehensive determination unit is a fall of an object.

4. The fall detection device according to claim 1 or 2, characterized in that the fall detection unit determines, based on the motion component, that an object has fallen if the motion speed of the motion in a space detectable by the sensor is equal to or greater than a preset threshold.

5. The fall detection device according to claim 1 or 2, characterized in that the motion state detected by the immediate post-motion state detection unit is a motion state obtained using the time-series changes of the velocity, distance, or velocity and distance of an object that may have fallen over.

6. The fall detection device according to claim 1 or 2, characterized in that the motion state detected by the immediate post-motion state detection unit is a movement characteristic of the human body that lasts for a predetermined period of time or longer.

7. The fall detection device according to claim 1 or 2, characterized in that the fall detection unit determines a three-dimensional processing area based on the point where the fall of the object is detected, and the immediate post-fall state detection unit performs processing on the processing area within the space detectable by the sensor to detect the motion state of the object after it has fallen.

8. The fall detection device according to claim 1 or 2, characterized in that the comprehensive determination unit determines whether the fall is that of a living person or a non-living person based on a plurality of types of movement states.

9. The fall detection device according to claim 1 or 2, characterized in that the fall detection unit detects the fall of an object using the motion component based on a sensor signal output by a sensor installed above the detection target area, which is an area in which an object that may fall may exist.

10. The fall detection device according to claim 1 or 2, characterized in that the comprehensive determination unit, when the fall detection unit detects that an object has fallen, instructs an external control device to output video of the space in which the fall of the object was detected to a terminal device of a pre-registered user, and when the fall of a living person is detected, further instructs an external control device to send an alarm to a terminal device of a pre-registered user.

11. The fall detection device according to claim 3, characterized in that the comprehensive determination unit, when the fall detection unit detects that an object has fallen, instructs the control device to output video of the space in which the object fell to a terminal device of a pre-registered user, and when the fall of a person is detected, further instructs the control device to send an alarm to a terminal device of a pre-registered user.

12. The fall detection device according to claim 1 or 2, characterized in that the immediate post-motion state detection unit detects the motion state using a second sensor signal output by a second sensor which is a different type of sensor from the sensor that output the sensor signal used to extract the motion component.

13. The fall detection device according to claim 12, characterized in that the second sensor is a video sensor.

14. The tip-over detection device according to claim 12, characterized in that the second sensor is a temperature sensor.

15. The fall detection device according to claim 12, characterized in that the second sensor is a sound sensor.

16. The fall detection device according to claim 1 or 2, characterized in that the comprehensive determination unit outputs fall information including the probability for each type of object that is likely to have fallen as a result of the determination.

17. A fall detection system comprising: a fall detection device according to claim 2; and a sensor installed above a detection target area which is a space in which an object that may fall may exist, and which outputs a sensor signal to the fall detection device.

18. A fall detection device having: a motion reflection component extraction unit that receives a sensor signal output by a sensor and extracts a motion component that is a signal component indicating a motion that is likely to be present in a detection target area which is a space detectable by the sensor, based on the sensor signal; a fall determination unit that receives the motion component and detects the fall of an object based on the motion component; an immediate post-fall state detection unit that detects the motion state of the object after the fall if the fall determination unit has detected the fall of an object; and an overall determination unit that determines whether the fall is that of a living being or a non-living being based on the motion state detected by the immediate post-fall state detection unit, wherein if the fall determination unit has detected the fall of an object, the overall determination unit commands the control device to output an image of the space in which the fall of the object was detected to a terminal device of a pre-registered user, and if the fall of a living being is detected, the overall determination unit further commands the control device to send an alarm to the terminal device of a pre-registered user; A fall detection system comprising: a control device that, in accordance with a command from the comprehensive determination unit, outputs video of the space where the fall of an object was detected to a pre-registered user's terminal device when the fall of an object is detected by the fall determination unit, and when the fall of a living person is detected, an alarm is sent to a pre-registered user's terminal device.

19. A fall detection method using a fall detection device, comprising: a fall determination step in which a fall determination unit of the fall detection device receives a motion component, which is a signal component indicating a moving object extracted from a sensor signal, and detects that an object has fallen based on the motion component; an immediate post-state detection step in which, if the fall determination unit has detected that an object has fallen, the immediate post-state detection unit of the fall detection device detects the motion state of the object after it has fallen; and an overall determination step in which the overall determination unit of the fall detection device determines, based on the motion state detected by the immediate post-state detection unit, whether the fall is that of a living being or a non-living being.

Images