Multi-radar based target detection device and detection method

Abstract

To provide a multi-radar based detection device and detection method for a target object.SOLUTION: In a detection method, a first detection result corresponding to a first detection space and a second detection result corresponding to a second detection space are received. The first detection space entering a first status is determined in response to the first detection result indicating that the target object in the first detection space moves to an overlapping area between the first detection space and the second detection space. First information is output in response to determining that the first detection space enters the first status. The second detection space entering a second status is determined in response to the second detection result indicating that the target object not in the second detection space appears in the overlapping area. Second information is output in response to determining that the second detection space enters the second status.SELECTED DRAWING: Figure 15

Description

[Technical Field]

[0001] The present invention relates to radar detection technology, and more particularly to a target detection device and method based on multi-radar. [Background technology]

[0002] Due to social changes and aging, more and more elderly people live alone. How to care for these elderly people living alone is a major challenge facing today's society. In addition to utilizing the power of social groups such as social welfare workers and volunteers to conduct visits, care devices with remote detection functions can provide immediate assistance to care recipients when emergencies occur. Existing remote detection devices are mainly based on identification using wearable devices or imaging devices. When installing location detection devices in private indoor spaces, radar can be used to detect care recipients and prevent their images from being recorded, thereby reducing the invasion of their privacy. Summary of the Invention [Problem to be solved by the invention]

[0003] However, since the living environment of a care recipient may include multiple spaces, the detection range of a single radar may not be able to completely cover the living environment. Therefore, it is usually necessary to install a dedicated radar in each independent space. Therefore, one of the key challenges is how to integrate the information from multiple radar systems to effectively detect the care recipient. [Means for solving the problem]

[0004] The present invention provides a multi-radar based target detection apparatus and method that can integrate detection results from multiple radars to provide accurate information to the user.

[0005] One embodiment of the present invention provides a multi-radar-based target detection device. The detection device includes a human-machine interface, a transceiver, and a processor. A first detection result corresponding to a first detection space and a second detection result corresponding to a second detection space are received from multiple radars. The processor is coupled to the human-machine interface and the transceiver and is configured to: determine that the first detection space has entered a first state in response to the first detection result indicating that a target object previously in the first detection space has moved into an overlapping region between the first detection space and the second detection space; output first information indicating that the target object is located within the first detection space via the human-machine interface in response to determining that the first detection space has entered the first state; determine that the second detection space has entered a second state in response to the second detection result indicating that a target object previously not in the second detection space has appeared in the overlapping region; and output second information indicating that the target object is not present in the second detection space via the human-machine interface in response to determining that the second detection space has entered the second state.

[0006] One embodiment of the present invention provides a multi-radar-based target detection method, which includes receiving a first detection result corresponding to a first detection space and a second detection result corresponding to a second detection space from multiple radars, determining that the first detection space has entered a first state in response to the first detection result indicating that a target object in the first detection space has moved into an overlapping region between the first detection space and the second detection space, outputting first information indicating that the target object is located within the first detection space via a human-machine interface in response to determining that the first detection space has entered the first state, determining that the second detection space has entered a second state in response to the second detection result indicating that a target object that was not in the second detection space has appeared in the overlapping region, and outputting second information indicating that the target object is not in the second detection space via the human-machine interface in response to determining that the second detection space has entered the second state. [Effects of the Invention]

[0007] Based on the above, the present invention can combine the detection results of multiple radars to quickly and accurately determine the space in which the target is located and the state of the target. The present invention can also provide various information for the user to view through a human-machine interface, helping the user quickly understand the state of the target or the detected space. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a schematic diagram of a multi-radar based target detection device according to one embodiment of the present invention; [Figure 2] FIG. 2 is a schematic diagram of a detection space according to one embodiment of the present invention. [Figure 3] FIG. 1 is a flow diagram of a method for detecting a subject according to one embodiment of the present invention. [Figure 4A] FIG. 2 is a schematic diagram of bounding box evolution according to one embodiment of the present invention; [Figure 4B] FIG. 2 is a schematic diagram of bounding box evolution according to one embodiment of the present invention; [Figure 4C] FIG. 2 is a schematic diagram of bounding box evolution according to one embodiment of the present invention; [Figure 5] FIG. 2 is a schematic diagram of a first state machine of a detection space according to one embodiment of the present invention. [Figure 6] FIG. 10 is a schematic diagram of a second state machine of the detection space according to one embodiment of the present invention. [Figure 7A] FIG. 2 is a schematic diagram of a first example of state transitions in a detection space according to one embodiment of the present invention. [Figure 7B] FIG. 2 is a schematic diagram of a first example of state transitions in a detection space according to one embodiment of the present invention. [Figure 8A] FIG. 10 is a schematic diagram of a second example of state transitions in a detection space according to one embodiment of the present invention. [Figure 8B] FIG. 10 is a schematic diagram of a second example of state transitions in a detection space according to one embodiment of the present invention. [Figure 9A]FIG. 10 is a schematic diagram of a third example of state transitions in a detection space according to one embodiment of the present invention. [Figure 9B] FIG. 10 is a schematic diagram of a third example of state transitions in a detection space according to one embodiment of the present invention. [Figure 10A] FIG. 10 is a schematic diagram of a fourth example of state transitions in a detection space according to one embodiment of the present invention. [Figure 10B] FIG. 10 is a schematic diagram of a fourth example of state transitions in a detection space according to one embodiment of the present invention. [Figure 11A] FIG. 10 is a schematic diagram of a fifth example of state transitions in a detection space according to one embodiment of the present invention. [Figure 11B] FIG. 10 is a schematic diagram of a fifth example of state transitions in a detection space according to one embodiment of the present invention. [Figure 12] FIG. 2 is a schematic diagram of a graphical user interface display by a human machine interface according to one embodiment of the present invention. [Figure 13] 1 is a schematic diagram of an icon for a graphical user interface according to one embodiment of the present invention. [Figure 14] 1 is a schematic diagram of an icon for a graphical user interface according to one embodiment of the present invention. [Figure 15] 1 is a flow diagram of a multi-radar based target object detection method according to one embodiment of the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0009] 1 is a schematic diagram of a multi-radar based target detection apparatus 100 according to one embodiment of the present invention. The detection apparatus 100 may include a processor 110, a human-machine interface 120, and a transceiver 130.

[0010] The processor 110 may be, for example, a central processing unit (CPU) or other programmable general-purpose or special-purpose microcontroller unit (MCU), microprocessor, digital signal processor (DSP), programmable controller, application specific integrated circuit (ASIC), graphical processing unit (GPU), image signal processor (ISP), image processing unit (IPU), arithmetic logic unit (ALU), complex programmable logic device (CPLD), field programmable gate array (FPGA), or other similar component, or a combination of the above components. The processor 110 is coupled to the human-machine interface 120 and the transceiver 130 and may access and execute various modules and applications stored on storage media.

[0011] The human machine interface (HMI) 120 is used to receive information input by a user or to output information for the user's reference. The human machine interface 120 may include a device such as a touch screen.

[0012] The transceiver 130 transmits and receives signals wirelessly or via a wired system. The transceiver 130 may perform operations such as low-noise amplification, impedance matching, frequency mixing, frequency up- or down-conversion, filtering, amplification, etc. The detection device 100 may be communicatively connected to multiple radars installed at multiple locations via the transceiver 130 and receive detection results of each radar in the detection space from the multiple radars.

[0013] 2 is a schematic diagram of a detection space 200 according to one embodiment of the present invention. For example, the target object detected by the detection device 100 may be a subject (care recipient) 20, and the detection space 200 may be, for example, the area where the subject 20 lives. In the following embodiment, it is assumed that the detection space 200 includes two independent detection spaces 210 and 220, and the detection spaces 210 and 220 overlap with an overlapping area 300. The detection space 210 may include a default area 400. The default area 400 may be included in the detection space 210 or the detection space 220, and the default area 400 may be, for example, an entrance or exit of the detection space 220. The radar 11 may be configured to monitor the detection space 210 and generate a corresponding detection result. The radar 12 may be configured to monitor the detection space 220 and generate a corresponding detection result. The detection device 100 may receive detection results corresponding to the detection space 210 and detection results corresponding to the detection space 220 from the radar 11 and the radar 12, respectively, via the transceiver 130. The radar 11 or the radar 12 may be, for example, a continuous wave (CW) radar used to detect information including the physiological state (e.g., breathing or heart rate) of the subject 20, a frequency modulated continuous wave (FMCW) radar or an impulse radio ultra-wideband (IR-UWB) radar used to detect the movement of the subject 20 and generate detection results in the detection space, but the present invention is not limited thereto.

[0014] FIG. 3 is a flow diagram of a method for detecting a subject according to one embodiment of the present invention, which method is performed by the detection device 100 shown in FIG.

[0015] In step S301, the processor 110 may determine whether the subject 20 is in the detection space (e.g., the detection space 210 or 220) based on a detection result of a radar (e.g., the radar 11 or 12). If the subject 20 is in the detection space, step S302 is executed. If the subject 20 is not in the detection space, step S308 is executed. In one embodiment, the processor 110 may determine that the subject 20 is in the detection space based on a detection result of the radar that includes information related to the physiological state of the subject 20, and may determine that the subject 20 is not in the detection space based on a detection result that does not include information related to the physiological state of the subject.

[0016] In step S302, the processor 110 may display, for the user's reference, information indicating that the subject 20 is in the detection space via the human-machine interface 120.

[0017] In step S303, the processor 110 may count the staying time of the target person 20 in the detection space based on the target person 20 being in the detection space, and determine whether the staying time is excessively long (e.g., the staying time is longer than a threshold or an upper limit), excessively short (e.g., the staying time is shorter than a threshold or a lower limit), or normal (e.g., the staying time is below the upper limit or above the lower limit). If the staying time is excessively short, step S304 is executed. If the staying time is excessively long, step S305 is executed. If the staying time is normal, step S306 is executed.

[0018] In step S304, the processor 110 may display a warning message via the human-machine interface 120 indicating that the time spent by the subject 20 in the detection space is too short.

[0019] In step S305, the processor 110 may display a warning message via the human-machine interface 120 indicating that the subject 20 has stayed in the detection space for an excessively long time.

[0020] In step S306, the processor 110 may determine whether the subject 20 has fallen in the detection space based on the detection result. If it is determined that the subject 20 has fallen, step S307 is executed. If it is determined that the subject 20 has not fallen, step S301 is executed again after a certain time has elapsed.

[0021] Specifically, processor 110 may acquire a plurality of point cloud data corresponding to a plurality of time points from the detection results of subject 20 and the detection space by radar, generate a plurality of bounding boxes corresponding to the plurality of time points, and perform object detection on the plurality of point cloud data. Processor 110 may determine whether the subject has fallen based on a change in the bounding boxes.

[0022] In one embodiment, the processor 110 may determine a change in speed, height, or tilt angle of the subject 20 using multiple bounding boxes and determine whether the change in speed, height, or tilt angle has caused the subject 20 to fall.

[0023] 4A, 4B, and 4C are schematic diagrams illustrating changes in bounding boxes according to one embodiment of the present invention. Referring to FIG. 4A, assume that the processor 110 acquires bounding boxes 41, 42, and 43 corresponding to point clouds of the subject 20 in a time series. The processor 110 may determine the speed of the subject 20 for a default time (e.g., 5 seconds) using the bounding boxes 41, 42, and 43. If the speed of the subject 20 is faster than a threshold (e.g., 50 cm per second), the processor 110 may determine that the subject 20 may have fallen.

[0024] 4B , assume that the processor 110 acquires bounding boxes 44, 45, and 46 corresponding to the point cloud of the subject 20 in time series. The processor 110 may determine the height of the center of mass or center of gravity of the subject 20 according to the bounding boxes 44, 45, and 46. If the height of the center of mass or center of gravity of the subject 20 is less than a threshold value (e.g., 50 cm), the processor 110 may determine that the subject 20 may have fallen.

[0025] 4C , it is assumed that the processor 110 acquires bounding boxes 47, 48, and 49 corresponding to point clouds of the subject 20 in a time series. The processor 110 may determine a change in tilt angle of the subject 20 according to the bounding boxes 47, 48, and 49. If the change in tilt angle of the subject 20 is greater than a threshold value (e.g., 45 degrees), the processor 110 may determine that the subject 20 may have fallen.

[0026] Returning to FIG. 3, in step S307, the processor 110 may display, via the human-machine interface 120, information indicating that the subject 20 may have fallen, for the user's reference.

[0027] In step S308, the processor 110 may display information indicating that the subject 20 is not in the detection space via the human-machine interface 120 for the user's reference.

[0028] In step S309, the processor 110 may count the absence time after the target person 20 leaves the detection space, and determine whether the absence time is excessively long (e.g., the absence time is longer than a threshold or upper limit) based on the fact that the target person 20 is not in the detection space. If it is determined that the absence time is excessively long, step S310 is executed. If it is determined that the absence time is not excessively long, step S301 is executed again after a certain time has elapsed.

[0029] In step S310, the processor 110 may display information indicating that the subject 20 has been absent for an excessively long period of time via the human-machine interface 120 for the user's reference.

[0030] 5 is a schematic diagram of a first state machine 500 of a detection space (e.g., detection space 210 or 220) according to an embodiment of the present invention. The processor 110 may determine the current state of the detection space based on the detection results of each radar and the state machine 500, and output corresponding information through the human-machine interface 120 according to the state of the detection space, which is used to indicate whether a target is located within the detection space. Taking the detection space 210 as an example, the state machine 500 may include the following states and transition conditions:

[0031] State S1: The subject 20 fades out from the detection space 210. When the detection space 210 is in state S1, the processor 110 outputs information indicating that the subject 20 is in the detection space 210 via the human-machine interface 120.

[0032] State S2: The subject 20 fades into the detection space 210. When the detection space 210 is in state S2, the processor 110 outputs information indicating that the subject 20 is not in the detection space 210 via the human-machine interface 120.

[0033] State S3: The subject 20 is not detected in the detection space 210. When the detection space 210 is in state S3, the processor 110 outputs information indicating that the subject 20 is not in the detection space 210 via the human-machine interface 120.

[0034] State S4: The subject 20 is detected in the detection space 210. When the detection space 210 is in state S4, the processor 110 outputs, via the human-machine interface 120, information indicating that the subject 20 is in the detection space 210.

[0035] State S5: The subject 20 is about to leave the detection space 210. When the detection space 210 is in state S5, the processor 110 outputs information indicating that the subject 20 is in the detection space 210 via the human-machine interface 120.

[0036] State S6: The subject 20 is about to enter the detection space 210. When the detection space 210 is in state S6, the processor 110 outputs information indicating that the subject 20 is not in the detection space 210 via the human-machine interface 120.

[0037] Transition condition T1: When the detection space 210 is in state S4, the detection result indicates that the subject 20 located in the detection space 210 moves from the detection space 210 to the overlap region 300. When the transition condition T1 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S4 to state S1.

[0038] Transition condition T2: When the detection space 210 is in state S1, the detection result indicates that the subject 20 has left the detection space 210. When the transition condition T2 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S1 to state S3.

[0039] Transition condition T3: When the detection space 210 is in state S1, the detection result indicates that the subject 20 has moved from the overlap region 300 to the detection space 210. When the transition condition T3 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S1 to state S4.

[0040] Transition condition T4: When the detection space 210 is in state S2, the detection result indicates that the subject 20 has moved from the overlap region 300 to the detection space 210. When the transition condition T4 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S2 to state S4.

[0041] Transition condition T5: When the detection space 210 is in state S2, the detection result indicates that the subject 20 has left the detection space 210. When the transition condition T5 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S2 to state S3.

[0042] Transition condition T6: When the detection space 210 is in state S3, the detection result indicates that the subject 20 who was not in the detection space 210 has appeared in the overlap region 300. When the transition condition T6 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S3 to state S2.

[0043] Transition condition T7: When the detection space 210 is in state S4, the detection result indicates that the subject 20 who was located in the detection space 210 has left the detection space 210. When the transition condition T7 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S4 to state S5.

[0044] Transition condition T8: When the detection space 210 is in state S5, the detection result indicates that the subject 20 has appeared in the detection space 210. When the transition condition T8 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S5 to state S4.

[0045] Transition condition T9: When the detection space 210 is in state S5, the detection result indicates that the subject 20 is not detected in the detection space 210 (i.e., the subject 20 does not appear in the detection space 210). When the transition condition T9 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S5 to state S3.

[0046] Transition condition T10: When the detection space 210 is in state S6, the detection result indicates that the subject 20 has not left the detection space 210. When the transition condition T10 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S6 to state S4.

[0047] Transition condition T11: When the detection space 210 is in state S6, the detection result indicates that the subject 20 has left the detection space 210. When the transition condition T11 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S6 to state S3.

[0048] Transition condition T12: When the detection space 210 is in state S3, the detection result indicates that the subject 20 who was not in the detection space 210 has appeared in the detection space 210. When the transition condition T12 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S3 to state S6.

[0049] Transition condition T13: When the detection space 210 is in state S4, the detection result indicates that the subject 20 who was located in the detection space 210 has appeared in the detection space 210. When the transition condition T13 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S4 to state S4.

[0050] Transition condition T14: When the detection space 210 is in state S3, the detection result indicates that the subject 20 who was not in the detection space 210 has appeared in the detection space 210. When the transition condition T14 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S3 to state S3.

[0051] 6 is a schematic diagram of a second state machine 600 of a detection space (e.g., detection space 210 or 220) according to an embodiment of the present invention. The processor 110 may determine the current state of the detection space based on the detection results of each radar and the state machine 600, and output corresponding information through the human-machine interface 120 according to the state of the detection space, which is used to indicate whether a target is located within the detection space. Taking the detection space 210 as an example, compared with the state machine 500, the state machine 600 may further include the following transition conditions:

[0052] Transition condition T15: When the detection space 210 is in state S4, the detection result indicates that the subject 20 who was located in the detection space 210 has left the default region 400. When the transition condition T15 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S4 to state S3.

[0053] Transition condition T16: When the detection space 210 is in state S3, the detection result indicates that the subject 20 who was not in the detection space 210 has appeared in the default region 400. When the transition condition T16 is satisfied, the processor 110 may determine that the detection space 210 has switched from state S3 to state S4.

[0054] 7A and 7B are schematic diagrams of a first example of state transitions in a detection space according to one embodiment of the present invention, where FIG. 7A shows a state machine 500 corresponding to the detection space 210 and FIG. 7B shows a state machine 500 corresponding to the detection space 220. Assume that the subject 20 is in the detection space 210. State S4 of the detection space 210 indicates that the subject 20 is located in the detection space 210, and state S3 of the detection space 220 indicates that the subject 20 is not in the detection space 220. When the subject 20 is not moving, the transition conditions T13 of the detection space 210 and T14 of the detection space 220 are satisfied, respectively, and the state of the detection space 210 is returned to state S4, and the state of the detection space 220 is switched to state S3.

[0055] 8A and 8B are schematic diagrams of a second example of state transitions in the detection space according to one embodiment of the present invention, where FIG. 8A shows a state machine 500 corresponding to the detection space 210 and FIG. 8B shows a state machine 500 corresponding to the detection space 220. Assume that the subject 20 is in the detection space 210. State S4 of the detection space 210 indicates that the subject 20 is located in the detection space 210, and state S3 of the detection space 220 indicates that the subject 20 is not in the detection space 220. When the subject 20 appears in the overlap region 300, the transition conditions T1 of the detection space 210 and T6 of the detection space 220 are satisfied, respectively, and the state of the detection space 210 is switched to state S1, and the state of the detection space 220 is switched to state S2. Next, when the subject 20 leaves the detection space 210 and moves from the overlap area 300 to the detection space 220, the transition condition T2 of the detection space 210 and the transition condition T4 of the detection space 220 are satisfied, and the state of the detection space 210 is switched to state S3, and the state of the detection space 220 is switched to state S4.

[0056] 9A and 9B are schematic diagrams of a third example of state transitions in the detection space according to one embodiment of the present invention, where FIG. 9A shows a state machine 500 corresponding to the detection space 210 and FIG. 9B shows a state machine 500 corresponding to the detection space 220. Assume that the subject 20 is in the detection space 210. State S4 of the detection space 210 indicates that the subject 20 is located in the detection space 210, and state S3 of the detection space 220 indicates that the subject 20 is not in the detection space 220. When the subject 20 appears in the overlap region 300, the transition conditions T1 of the detection space 210 and T6 of the detection space 220 are satisfied, respectively, and the state of the detection space 210 is switched to state S1, and the state of the detection space 220 is switched to state S2. Next, when the subject 20 moves from the overlap area 300 to the detection space 210 and disappears from the detection space 220, the transition condition T3 of the detection space 210 and the transition condition T5 of the detection space 220 are satisfied, and the state of the detection space 210 is switched to state S4, and the state of the detection space 220 is switched to state S3.

[0057] 10A and 10B are schematic diagrams of a fourth example of state transitions in the detection space according to one embodiment of the present invention, where FIG. 10A shows a state machine 500 corresponding to the detection space 210 and FIG. 10B shows a state machine 500 corresponding to the detection space 220. Assume that the subject 20 is in the detection space 210. State S4 of the detection space 210 indicates that the subject 20 is located in the detection space 210, and state S3 of the detection space 220 indicates that the subject 20 is not in the detection space 220. When the subject 20 leaves the detection space 210 and appears in the detection space 220, the transition conditions T7 of the detection space 210 and T12 of the detection space 220 are satisfied, respectively, and the state of the detection space 210 is switched to state S5, and the state of the detection space 220 is switched to state S6. Next, if the subject 20 does not appear in the detection space 210 after a certain time, and if the subject 20 does not appear in the detection space 220 after a certain time, the transition condition T9 of the detection space 210 and the transition condition T10 of the detection space 220 are respectively satisfied, and the state of the detection space 210 is switched to state S3, and the state of the detection space 220 is switched to state S4.

[0058] 11A and 11B are schematic diagrams of a fifth example of state transitions in the detection space according to one embodiment of the present invention, where FIG. 11A shows a state machine 500 corresponding to the detection space 210 and FIG. 11B shows a state machine 500 corresponding to the detection space 220. Assume that the subject 20 is in the detection space 210. State S4 of the detection space 210 indicates that the subject 20 is located in the detection space 210, and state S3 of the detection space 220 indicates that the subject 20 is not in the detection space 220. If the subject 20 leaves the detection space 210 and appears in the detection space 220, the transition conditions T7 of the detection space 210 and T12 of the detection space 220 are satisfied, respectively, and the state of the detection space 210 is switched to state S5, and the state of the detection space 220 is switched to state S6. Next, when the subject 20 appears in the detection space 210 and disappears from the detection space 220, the transition condition T8 of the detection space 210 and the transition condition T11 of the detection space 220 are satisfied, respectively, and the state of the detection space 210 is switched to state S4, and the state of the detection space 220 is switched to state S3.

[0059] 12 is a schematic diagram of a graphical user interface 1100 displayed by the human-machine interface 120 according to one embodiment of the present invention. The graphical user interface 1100 may be used to display icons indicating the physiological state of the subject and may be used to display icons indicating the occupancy status of an area of ​​the detection space (e.g., detection space 210 or 220). For example, the graphical user interface 1100 may include an icon 121 used to indicate the number and name of the detection space, an icon 122 used to indicate the occupancy status of a bed in the detection space, an icon 123 used to indicate the physiological state of the subject 20, such as the heart rate or breathing, an icon 124 used to indicate the occupancy status of an area, such as a bedroom, living room, kitchen, or toilet, an icon 125 used to indicate whether the subject 20 is in the detection space, and an icon 126 used to indicate the time spent by the subject 20 in the detection space.

[0060] 13 is a schematic diagram of an icon 124 of a graphical user interface according to one embodiment of the present invention. For an area in the detection space (e.g., a bedroom, living room, kitchen, or toilet), the icon 124 may be used to indicate the occupancy status of the area or a related event, such as information such as the area is occupied, the area is not occupied, an emergency has occurred in the area, a fall event has occurred in the area, a radar used to monitor the area has been disconnected, or a subject has remained in the area for too long.

[0061] 14 is a schematic diagram of a graphical user interface icon 125 according to one embodiment of the present invention. For a detection space, the icon 125 may be used to indicate the usage status or related events of the detection space, such as information such as the detection space being in use, the detection space being unused, a subject remaining in the detection space for too long, the detection space being idle (e.g., unused) for too long, or a radar used to monitor the detection space being disconnected.

[0062] FIG. 15 is a flow diagram of a target detection method based on multiple radars according to one embodiment of the present invention, which may be implemented by the detection device 100 of FIG. 1. In step S151, a first detection result corresponding to a first detection space and a second detection result corresponding to a second detection space are received from multiple radars. In step S152, it is determined that the first detection space has entered a first state in response to the first detection result indicating that a target in the first detection space has moved into an overlapping area between the first detection space and the second detection space. In step S153, in response to determining that the first detection space has entered the first state, first information is output via a human-machine interface, where the first information indicates that the target is located in the first detection space. In step S154, it is determined that the second detection space has entered a second state in response to a second detection result indicating that a target that was not in the second detection space has appeared in the overlapping area. In step S155, in response to determining that the second detection space has entered the second state, second information is output via the human-machine interface, where the second information indicates that the target object is not in the second detection space.

[0063] Based on the above, the present invention has the following advantages: In the present invention, detection results from multiple detection devices can be integrated to provide users with real-time information on the location and status of targets. The present invention is highly scalable and flexible, and the number and location of radars can be flexibly configured according to different application environments to monitor spaces of different sizes. The human-machine interface of the present invention can display the status of multiple devices to allow users to quickly understand the situation on site. In the present invention, users can be immediately notified when the target's status is abnormal, allowing them to respond to emergencies as quickly as possible. In the present invention, the status of each space can be automatically detected to reduce human resource costs. The information integrated by the present invention can be applied to, for example, people flow statistics, behavior analysis, or activity record analysis to increase the application value of the detection results. [Industrial Applicability]

[0064] The multi-radar based target detection apparatus and method of the present invention can be applied to any radar detection technology. [Explanation of symbols]

[0065] 11: Radar 12: Radar 20: Target 41, 42, 43, 44, 45, 46, 47, 48, 49: Bounding box 100:Detection device 110: Processor 120: Human-machine interface 121, 122, 123, 124, 125, 126: Icons 130: Transmitter / Receiver 200: Detection area 210, 220: Detection space 300: Overlapping area 400: Default area 500, 600: State machine 1100: Graphical User Interface S1, S2, S3, S4, S5, S6: Status S151, S152, S153, S154, S155, S301, S302, S303, S304, S305, S306, S307, S308, S309, S310: Step T1, T2, T3, T4, T5, T6, T7, T8, T9, T10, T11, T12, T13, T14, T15, T16: Transition conditions

Claims

1. 1. A multi-radar based target detection device, comprising: Human machine interface and a transceiver that receives a first detection result corresponding to the first detection space and a second detection result corresponding to the second detection space; coupled to the human-machine interface and the transceiver; determining that the first detection space has entered a first state in response to the first detection result indicating that the target object in the first detection space has moved into an overlapping region between the first detection space and the second detection space; outputting first information indicating that the target object is located in the first detection space via the human-machine interface in response to determining that the first detection space has entered the first state; determining that the second detection space has entered a second state in response to a second detection result indicating that the target object that was not in the second detection space has appeared in the overlapping area; In response to determining that the second detection space has entered the second state, outputting second information indicating that the target object is not in the second detection space via the human-machine interface. and a processor configured to Including, The processor: determining that the first detection space has entered a fourth state from the first state in response to the first detection result indicating that the target object has moved from the overlap region into the first detection space; outputting the first information via the human-machine interface in response to determining that the first detection space has entered the fourth state; determining that the second detection space has entered a third state from the second state in response to the second detection result indicating that the target object has left the second detection space; outputting the second information via the human-machine interface in response to determining that the second detection space has entered the third state; further configured as follows: Detection device.

2. A multi-radar based target detection device, comprising: Human machine interface and a transceiver that receives a first detection result corresponding to the first detection space and a second detection result corresponding to the second detection space; coupled to the human-machine interface and the transceiver; determining that the first detection space has entered a first state in response to the first detection result indicating that the target object in the first detection space has moved into an overlapping region between the first detection space and the second detection space; outputting first information indicating that the target object is located in the first detection space via the human-machine interface in response to determining that the first detection space has entered the first state; determining that the second detection space has entered a second state in response to a second detection result indicating that the target object that was not in the second detection space has appeared in the overlapping area; In response to determining that the second detection space has entered the second state, outputting second information indicating that the target object is not in the second detection space via the human-machine interface. and a processor configured to Including, The processor: acquiring a plurality of point clouds corresponding to a plurality of time points from the first detection result; performing object detection on the plurality of point clouds to generate a plurality of bounding boxes corresponding to the plurality of point clouds; determining a change in tilt angle of the target object, a velocity of the target object, or a height of the target object for a default time based on the plurality of bounding boxes; determining that the target object has fallen over when the change in tilt angle is greater than a first threshold, the height is less than a second threshold, or the speed is greater than a third threshold; In response to determining that the target object has fallen, a warning message is output via the human-machine interface. further configured as follows: Detection device.

3. The processor: determining that the first detection space has entered a third state from the first state in response to the first detection result indicating that the target object has left the first detection space; outputting third information indicating that the target object is not in the first detection space via the human-machine interface in response to determining that the first detection space has entered the third state; determining that the second detection space has entered a fourth state from the second state in response to the second detection result indicating that the target object has moved from the overlap region into the second detection space; In response to determining that the second detection space has entered the fourth state, outputting fourth information indicating that the target object is located in the second detection space via the human-machine interface. further configured as follows:

3. The detection device according to claim 1 or 2.

4. A method for detecting a target based on multi-radar, comprising: receiving, from a plurality of radars, a first detection result corresponding to a first detection space and a second detection result corresponding to a second detection space; determining that the first detection space has entered a first state in response to the first detection result indicating that the target object in the first detection space has moved into an overlapping region between the first detection space and the second detection space; In response to determining that the first detection space has entered the first state, outputting first information indicating that the target object is located in the first detection space via a human-machine interface; determining that the second detection space has entered a second state in response to the second detection result indicating that the target object that was not in the second detection space has appeared in the overlapping region; In response to determining that the second detection space has entered the second state, outputting second information indicating that the target object is not in the second detection space via the human-machine interface; determining that the first detection space has entered a fourth state from the first state in response to the first detection result indicating that the target object has moved from the overlap region into the first detection space; outputting the first information via the human-machine interface in response to determining that the first detection space has entered the fourth state; determining that the second detection space has entered a third state from the second state in response to the second detection result indicating that the target object has left the second detection space; outputting the second information via the human-machine interface in response to determining that the second detection space has entered the third state; Including, A target object detection method based on multi-radar.

5. A method for detecting a target based on a multi-radar, comprising: receiving, from a plurality of radars, a first detection result corresponding to a first detection space and a second detection result corresponding to a second detection space; determining that the first detection space has entered a first state in response to the first detection result indicating that the target object in the first detection space has moved into an overlapping region between the first detection space and the second detection space; In response to determining that the first detection space has entered the first state, outputting first information indicating that the target object is located in the first detection space via a human-machine interface; determining that the second detection space has entered a second state in response to the second detection result indicating that the target object that was not in the second detection space has appeared in the overlapping region; outputting second information indicating that the target object is not in the second detection space through the human-machine interface in response to determining that the second detection space has entered the second state; and obtaining a plurality of point clouds corresponding to a plurality of time points from the first detection result. performing object detection on the plurality of point clouds to generate a plurality of bounding boxes corresponding to the plurality of point clouds; determining a change in tilt angle of the target object, a velocity of the target object, or a height of the target object for a default time based on the plurality of bounding boxes; determining that the target object has fallen over when the change in tilt angle is greater than a first threshold, the height is less than a second threshold, or the speed is greater than a third threshold; outputting a warning message via the human-machine interface in response to determining that the target object has fallen; Including, A target object detection method based on multi-radar.

Images