Radar-based bathroom fall detection method, apparatus, and radar

Abstract

The application provides a radar-based bathroom fall detection method and device and a radar. The method comprises the following steps: determining the track of a target based on point cloud data detected by the radar, and obtaining associated point cloud associated with the track of the target frame by frame; determining the centroid height of a human body and a water flow in each frame of associated point cloud; determining whether it is a non-standing state according to the centroid height of the human body and the water flow corresponding to the associated point cloud of a first preset number of continuous frames; in the non-standing state, determining whether the centroid height of the human body and the water flow decreases according to the historical centroid height of the human body and the water flow; and if the centroid height of the human body and the water flow decreases, determining that the human body falls. The application can improve the human body fall discrimination accuracy in a showering scene.

Description

Technical Field

[0001] This invention relates to the field of radar detection technology, and in particular to a radar-based method, device, and radar for detecting falls in bathrooms. Background Technology

[0002] Data on falls among the elderly shows that over 60% of falls in the home occur in the bathroom, and the shower area is the most frequent location for falls among the elderly. During shower time, only one person is typically in the bathroom; if an elderly person falls and is not detected in time, it could lead to a serious accident. To prevent such accidents, various products play a role, such as pressure sensors, which rely on the ground pressure during a fall to determine dangerous behavior. However, these products have issues with wiring, installation, and subsequent maintenance. Portable accelerometers are inconvenient for the elderly because they need to be carried around. Image processing based on deep learning could solve this problem, but privacy concerns limit the acceptance of such products.

[0003] Radar-based fall detection methods for bathrooms can solve the problems mentioned above, but their data processing difficulty lies in the fact that the point clouds of people and water cannot be separated. Deep learning processing based on radar data is mostly based on micro-Doppler images. In these images, the Doppler data of water is dominant, and the Doppler data of people is submerged and difficult to separate, making it difficult to accurately extract fall information. This results in poor fall detection accuracy in shower scenarios. Summary of the Invention

[0004] In view of this, the present invention provides a radar-based method, device and radar for detecting falls in bathrooms, which can solve the problem of poor accuracy in radar-based fall detection in a specific environment.

[0005] In a first aspect, embodiments of the present invention provide a radar-based method for detecting falls in bathrooms, comprising:

[0006] The target's trajectory is determined based on the point cloud data detected by radar, and the associated point cloud associated with the target's trajectory is obtained;

[0007] Determine the centroid height of the human body and water flow in each frame of the associated point cloud;

[0008] Based on the centroid height of the human body and water flow corresponding to the associated point cloud of the first preset number of consecutive frames, determine whether it is a non-standing state.

[0009] In a non-standing state, based on the historical center of mass height of the human body and the water flow, determine whether the center of mass height of the human body and the water flow has decreased;

[0010] If the height of the center of mass of the human body and the water flow decreases, it is determined that the human body has fallen.

[0011] In a possible implementation, the determining the centroid height of the human body and the water flow in the frame-associated point cloud comprises:

[0012] obtaining an amplitude value and a height value of each point cloud in the frame-associated point cloud;

[0013] for any point cloud, weighting the height value of the point cloud based on the amplitude value of the point cloud and the amplitude value of each point cloud in the frame-associated point cloud, to obtain a weighted height value of the point cloud, so that the weight of the height value of the human body point cloud in the frame-associated point cloud is increased, and the weight of the height value of the water flow point cloud in the frame-associated point cloud is decreased;

[0014] calculating the centroid height of the human body and the water flow in the frame-associated point cloud according to the weighted height value of each point cloud.

[0015] In a possible implementation, the calculating the centroid height of the human body and the water flow in the frame-associated point cloud according to the amplitude value of each point cloud and the weighted height value of each point cloud comprises:

[0016] calculating the centroid height of the human body and the water flow in the frame-associated point cloud according to a preset formula, the preset formula being

[0017]

[0018] wherein zCenter is used to represent the centroid height of the human body and the water flow in the frame-associated point cloud, N is used to represent the number of point clouds in the frame-associated point cloud, snr(i) is used to represent the point cloud amplitude value of the ith point cloud, z(i) is used to represent the point cloud height value of the ith point cloud, and z(i) is used to represent the weighted height value of the ith point cloud.

[0019] In a possible implementation, the determining the centroid height of the human body and the water flow in the frame-associated point cloud comprises:

[0020] if the average value of the centroid height of the human body and the water flow corresponding to the frame-associated point clouds in the first preset number of continuous frames is greater than the first preset threshold value and the standard deviation is less than the second preset threshold value, the non-standing state is determined.

[0021] In a possible implementation, the determining the centroid height of the human body and the water flow in the frame-associated point cloud comprises:

[0022] ​If the height of the center of mass of the human body and the water flow in the first target frame associated point cloud is less than a third preset threshold, and the difference between the height of the center of mass of the human body and the water flow in the second target frame associated point cloud and the height of the center of mass of the human body and the water flow in the first target frame associated point cloud is greater than a fourth preset threshold, it is determined that the height of the center of mass of the human body and the water flow has decreased, wherein the first target frame is located after the second target frame, the first target frame and the second target frame are separated by a second preset number of frames, and the time length corresponding to the second preset number of frames is determined by the time length taken by the human body to fall.

[0023] In a possible implementation, after it is determined whether the height of the center of mass of the human body and the water flow has decreased, the method further includes:

[0024] If the height of the center of mass of the human body and the water flow has not decreased, it is determined whether there is a low weak signal according to the height of the center of mass of the human body and the water flow in at least one frame of associated point cloud.

[0025] If there is a low weak signal, it is determined that the human body has fallen.

[0026] In a possible implementation, if the height of the center of mass of the human body and the water flow has not decreased, it is determined whether there is a low weak signal according to the height of the center of mass of the human body and the water flow in at least one frame of associated point cloud.

[0027] If the height of the center of mass of the human body and the water flow in the last frame of associated point cloud in the first continuous first preset number of frames of associated point cloud is less than a fifth preset threshold, it is determined that there is activity in the low place.

[0028] Alternatively, if it is determined that the height of the center of mass of the human body and the water flow has not decreased, the height of the center of mass of the human body and the water flow in a continuous third preset number of frames of associated point cloud is obtained, if the number of frames in which the height of the center of mass of the human body and the water flow is less than a sixth preset threshold in the continuous third preset number of frames of associated point cloud is greater than a preset percentage, it is determined that there is activity in the low place, and the continuous third preset number of frames of associated point cloud is obtained after the first continuous first preset number of frames of associated point cloud.

[0029] In a second aspect, an embodiment of the present application provides a radar-based bathroom fall detection device, which includes an acquisition module, a first determination module, a first judgment module, a second judgment module, and a second determination module.

[0030] The acquisition module is configured to determine a track of a target based on point cloud data detected by a radar, and to acquire associated point cloud related to the track of the target.

[0031] The first determination module is configured to determine the height of the center of mass of the human body and the water flow in each frame of associated point cloud.

[0032] The first determining module is configured to determine whether it is a non-standing state according to the height of the center of mass of the human body and the water flow corresponding to the associated point cloud of the continuous first preset frame number.

[0033] The second determining module is configured to determine whether the height of the center of mass of the human body and the water flow decreases according to the height of the center of mass of the historical human body and the water flow in the non-standing state.

[0034] The second determining module is configured to determine whether the height of the center of mass of the human body and the water flow decreases according to the height of the center of mass of the historical human body and the water flow in the non-standing state.

[0035] In a third aspect, an embodiment of the present application provides a radar, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the method according to the first aspect or any possible implementation manner of the first aspect when executing the computer program.

[0036] In a possible implementation manner, the radar is a millimeter wave radar.

[0037] Compared with the related art, the embodiment of the present application has the following beneficial effects:

[0038] The embodiment of the present application extracts the height of the center of mass of the human body and the water flow from each frame of associated point cloud of the target track, obtains the height time sequence array of the center of mass of the human body and the water flow, determines whether it is a non-standing state based on the height time sequence array of the center of mass of the human body and the water flow, traces and analyzes based on the height array of the center of mass of the human body and the water flow in the non-standing state, determines whether the height of the center of mass of the human body and the water flow decreases, and determines that the human body falls if the height of the center of mass of the human body and the water flow decreases. The method provided by the embodiment of the present application analyzes the human and the water in the shower state as a whole, solves the influence of the water in the related art based on the radar to determine whether the human falls in the shower state, and improves the determination accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments or related technical descriptions. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0040] Figure 1 is an implementation flowchart of a shower fall detection method based on a radar provided by the embodiment of the present application;

[0041] Figure 2 is a structural schematic diagram of a shower fall detection device based on a radar provided by the embodiment of the present application;

[0042] Figure 3 is a schematic diagram of a radar provided by an embodiment of the present application. DETAILED DESCRIPTION

[0043] In the following description, for the purpose of explanation and not limitation, specific details are set forth, such as particular system configurations, techniques, etc., in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

[0044] In order to make the objects, technical solutions and advantages of the present application clearer, the following will be described by specific embodiments in conjunction with the accompanying drawings.

[0045] Referring to Figure 1 which shows an implementation flowchart of a radar-based bathroom fall detection method provided by an embodiment of the present application, and is described in detail as follows:

[0046] In step 101, the track of the target is determined based on the point cloud data detected by the radar, and the associated point cloud associated with the track of the target is obtained.

[0047] In the shower state, the human body point cloud and the water flow point cloud usually have no obvious boundary, and it is also difficult to accurately distinguish the human body point cloud and the water flow point cloud through related technologies such as the cluster track process. Therefore, in the embodiment of the present application, the human body and the water flow are regarded as a whole for target detection and tracking, that is, the target referred to in this step is a whole composed of the human body and the water flow.

[0048] For example, it is judged frame by frame whether the point cloud data detected by the radar is associated with the track of the target, and if so, the frame of point cloud data is taken as the associated point cloud associated with the track of the target.

[0049] In the embodiment of the present application, the radar, such as the millimeter wave radar, is used for target detection and tracking, the track of the target is obtained, and each frame of associated point cloud is obtained.

[0050] Since the point cloud obtained by processing the radar signal contains a certain number of other noise point clouds in addition to the human body point cloud and the water flow point cloud, these noise point clouds greatly interfere with target detection and tracking. Optionally, in the embodiment of the present application, the noise point cloud is removed through the cluster track process.

[0051] The cluster track process can be three-dimensional track tracking, or the point cloud can be converted to two-dimensional space for two-dimensional track tracking. Finally, after the cluster track process, the point cloud of the target is as follows: the upper half contains most of the point clouds of the water flow and the human body, and the lower half contains a small number of point clouds of the water flow and the human body.

[0052] In step 102, the centroid height of the human body and the water flow associated with each frame of point cloud is determined.

[0053] After the track of the target is determined, the centroid height of the human body and the water flow associated with each frame of point cloud is obtained.

[0054] This step is mainly used to obtain the centroid height of the human body and the water flow by extracting the reliable centroid of the human body from the point cloud mixed with the human body and the water.

[0055] Optionally, for a frame of associated point cloud, the centroid height of the human body and the water flow associated with the frame of point cloud is obtained by calculating the average value of all point cloud height values.

[0056] Alternatively, the amplitude value and the height value of each point cloud in the frame of associated point cloud are obtained; for any point cloud, the height value of the point cloud is weighted based on the amplitude value of the point cloud and the amplitude value of each point cloud in the frame of associated point cloud, to obtain the weighted point cloud height; and the centroid height of the human body and the water flow associated with the frame of associated point cloud is calculated according to the height value of each weighted point cloud.

[0057] Through statistics, it is found that, in the shower state, the echo point cloud amplitude of the human body part is stronger than that of the point cloud of the shower water, and based on this rule, in the embodiments of the present application, the point cloud height value is weighted based on the amplitude value of the point cloud to determine the centroid height of the human body and the water flow. For any frame of associated point cloud, this method can increase the weight of the height value of the human body point cloud in the frame of associated point cloud, and decrease the weight of the height value of the water flow point cloud in the frame of associated point cloud, so that the centroid height of the human body and the water flow obtained is more consistent with the actual situation.

[0058] Optionally, the centroid height of the human body and the water flow associated with the frame of associated point cloud is calculated according to a preset formula, and the preset formula is

[0059]

[0060] wherein zCenter is used to represent the centroid height of the human body and the water flow associated with the frame of associated point cloud, N is used to represent the number of point clouds of the frame of associated point cloud, snr(i) is used to represent the point cloud amplitude value of the ith point cloud, and z(i) is used to represent the point cloud height value of the ith point cloud. is used to represent the height value of the ith weighted point cloud.

[0061] Each frame of associated point cloud is obtained according to the track of the target through the association algorithm, and the number of point clouds of each frame of associated point cloud can be the same or different.

[0062] In the embodiment of the present application, since the signal-to-noise ratio can represent the amplitude information of the radar, optionally, the amplitude value of the point cloud is represented by the signal-to-noise ratio of the point cloud. In the above preset formula, snr(i) is used to represent the signal-to-noise ratio of the i th point cloud.

[0063] In the embodiment of the present application, a world coordinate system is preset, and the zero point of the z-axis of the world coordinate system is located on the ground.

[0064] Optionally, a buffer array zCenterList is preset, and the array size is zCenterListSize. After the buffer array is initialized, the centroid height of the human body and the water flow of each frame of associated point cloud is calculated, and the centroid height of the human body and the water flow of each frame of associated point cloud is sequentially stored in the buffer array in a tail update manner, that is, after the centroid height of the human body and the water flow of the latest frame of associated point cloud is calculated, the obtained height value is placed in the last position of the buffer array, and the other positions are moved by one position. When the buffer array is filled with valid data, the centroid height of the human body and the water flow of the zCenterListSize frames of associated point cloud closest to the current time is obtained. For example, zCenterListSize = 500, and the buffer array stores the centroid height data of the human body and the water flow of the 500 frames of associated point cloud closest to the current time, and the buffer array always stores the centroid height of the human body and the water flow of the zCenterListSize frames of associated point cloud closest to the current time in a tail update manner.

[0065] In step 103, whether it is a non-standing state is determined according to the centroid height of the human body and the water flow corresponding to the associated point cloud of the continuous first preset number of frames.

[0066] When a person under water falls down in a shower scene, the person usually does not move or moves very weakly, at this time, the point cloud of the water dominates, and it is found through statistics that the centroid height of the human body and the water flow obtained through step 101 is in a high position for a long time and stably.

[0067] Based on this, in this step, whether the centroid height of the human body and the water flow of the associated point cloud of the continuous first preset number of frames meets the condition that the centroid height is in a high position for a long time and stably is analyzed. If yes, the non-standing state under water is determined.

[0068] Optionally, the condition that the centroid height is in a high position for a long time and stably can be that if the average value of the centroid height of the human body and the water flow corresponding to the associated point cloud of the continuous first preset number of frames is greater than a first preset threshold, and the standard deviation is less than a second preset threshold, the non-standing state is determined.

[0069] For example, the data in the buffer array zCenterList is taken as the centroid height of the human body and water flow associated with the point cloud of the first preset frame number, Mean(zCenterList) is used to represent the average value, std(zCenterList) is used to represent the standard deviation, thresh1 is used to represent the first preset threshold, and thresh2 is used to represent the second preset threshold. The following conditions need to be met:

[0070] Mean(zCenterList)>thresh1, and std(zCenterList)<thresh2.

[0071] Optionally, when the array size of the buffer array zCenterList is greater than the first preset frame number, the data of the buffer array zCenterList after the first preset frame number can also be taken as the centroid height of the human body and water flow corresponding to the associated point cloud of the first preset frame number in this step. The present embodiment does not limit this.

[0072] In addition, the centroid height of the human body and water flow in the continuous first preset frame number can also be filtered or subjected to an outlier removal operation, and then the average value is calculated as the basis for determining whether it is in a higher position. In addition, the variance size is used as the basis for judging whether it is stable.

[0073] In the present embodiment, any method based on the present concept, that is, the centroid height of the human body and water flow in a higher position for a period of time is stable, is used as the basis for determining the non-standing state of the underwater state, which belongs to the protection scope of the present application.

[0074] In step 104, in the non-standing state, whether the centroid height of the human body and water flow decreases is determined according to the historical centroid height of the human body and water flow.

[0075] In the present embodiment, optionally, in order to reduce the data jitter caused by operation and sensors and improve the judgment accuracy, the data in the buffer array zCenterList is subjected to smoothing filtering. Optionally, Kalman filtering is used. Kalman filtering is a common filtering method, which mainly includes the steps of prediction, new construction, association, and update. Of course, other filtering methods such as mean filtering can also be selected. The filtered data is saved in the buffer array zCenterFilterList, and the data is stored in the form of tail update.

[0076] In the embodiment of the present application, the historical center-of-mass height of the human body and the water flow is the center-of-mass height data of the human body and the water flow in a preset interval before the current time, which can be based on the data in the buffer array zCenterList as the historical center-of-mass height data, or based on the data in the buffer array zCenterFilterList as the historical center-of-mass height data; preferably, the data in the buffer array zCenterFilterList is used as the historical center-of-mass height data.

[0077] The judgment basis for the center-of-mass height of the human body and the water flow to decrease is that, in the historical center-of-mass height of the human body and the water flow, the center-of-mass height of the human body and the water flow is lower at the later time, and the center-of-mass height difference between the human body and the water flow at the previous time and the later time is greater than a preset height difference, and the time interval between the previous time and the later time is the time interval required for the falling behavior, and the corresponding frame number is calculated based on the time interval, so that the frame interval between the frame associated with the point cloud at the previous time and the frame associated with the point cloud at the later time is the frame number calculated based on the time interval, which can be represented by FallSize. For example, the time interval required for the falling behavior is 1 second, and the radar obtains 10 frames of point cloud data within 1 second, so FallSize = 10.

[0078] Based on the judgment basis, whether the center-of-mass height of the human body and the water flow decreases is judged according to the historical center-of-mass height of the human body and the water flow.

[0079] In an optional implementation, if the center-of-mass height of the human body and the water flow in the first target frame associated point cloud in the historical center-of-mass height of the human body and the water flow is less than a third preset threshold, and the difference between the center-of-mass height of the human body and the water flow in the second target frame associated point cloud and the center-of-mass height of the human body and the water flow in the first target frame associated point cloud is greater than a fourth preset threshold, it is judged that the center-of-mass height of the human body and the water flow decreases, wherein the first target frame is located after the second target frame, the first target frame and the second target frame are separated by a second preset frame number, and the time length corresponding to the second preset frame number is determined by the time length required for the human body to fall.

[0080] Taking the data in the buffer array zCenterFilterList as the historical center-of-mass height data as an example, when the following conditions are met, it is judged that the center-of-mass height of the human body and the water flow decreases:

[0081] zCenterFilterList(j) < thresh3, and zCenterFilterList(j-FallSize) > thresh4.

[0082] Wherein, zCenterFilterList(j) is used to represent the value of the center of mass height of the human body and the water flow corresponding to the first target frame in the buffer array zCenterFilterList, zCenterFilterList(j-FallSize) is used to represent the value of the center of mass height of the human body and the water flow corresponding to the second target frame in the buffer array zCenterFilterList, thresh3 is used to represent the third preset threshold, and thresh4 is used to represent the fourth preset threshold.

[0083] In step 105, if the center of mass height of the human body and the water flow decreases, it is determined that the human body falls.

[0084] If it is determined by step 103 that the human body is in a non-standing state, and the historical center of mass height data of the human body and the water flow is analyzed by step 104, it is determined that the center of mass height of the human body and the water flow decreases before the non-standing state, and the human body falls, and an alarm is given. In the embodiment of the present application, the target refers to a person taking a shower.

[0085] After determining whether the center of mass height of the human body and the water flow decreases, the method further comprises: if the center of mass height of the human body and the water flow does not decrease, determining whether there is a low weak signal according to the center of mass height of the human body and the water flow of at least one frame of associated point cloud; and if there is a low weak signal, determining that the human body falls.

[0086] If it is determined by step 103 that the human body is in a non-standing state, but the historical center of mass height data of the human body and the water flow is analyzed by step 104, it is determined that the center of mass height of the human body and the water flow does not decrease before the non-standing state, in order to avoid misjudgment caused by algorithm error and improve the accuracy of the human body bathroom fall determination, after determining that the center of mass height does not decrease, the center of mass height of the human body and the water flow of the associated point cloud is continuously analyzed to determine whether the low place has activity determination condition, if the non-standing state is met, and the low place has activity determination condition is met, it is determined that the human body falls.

[0087] Optionally, whether there is a weak signal in the low place is used as a basis for determining whether there is activity in the low place, and the determination condition is that when a person is in the low place, the center of mass height of the human body and the water flow of the associated point cloud is low. That is, compared with the normal shower state of the human body, when the human body falls and is in an active state in the low place, the center of mass height of the human body and the water flow is low.

[0088] In one optional implementation, if the centroid height of the human body and water flow in the last frame of the associated point cloud of the first preset number of consecutive frames is less than a fifth preset threshold, then it is determined that there is activity at a lower location; or, if it is determined that the centroid height of the human body and water flow has not decreased, then the centroid height of the human body and water flow in the associated point cloud of the third preset number of consecutive frames is obtained. If the number of frames in the associated point cloud of the third preset number of consecutive frames where the centroid height of the human body and water flow is less than a sixth preset threshold is greater than a preset percentage, then it is determined that there is activity at a lower location. The associated point cloud of the third preset number of consecutive frames is obtained after the associated point cloud of the first preset number of consecutive frames.

[0089] For example, using data from the buffer array zCenterList as the centroid height data, the following condition is met to determine if someone is active at a lower location:

[0090] zCenterList(zCenterListSize-1)<thresh5.

[0091] Among them, zCenterList(zCenterListSize-1) is used to represent the centroid height of the human body and water flow in the latest frame of the associated point cloud in the buffer array zCenterList, and thresh5 is used to represent the fifth preset threshold.

[0092] The radar-based bathroom fall detection method provided in this invention has at least the following advantages:

[0093] First, it does not need to be worn on the body; it can be installed in the corner of the bathroom to provide real-time safety protection for people in the bathroom.

[0094] Second, millimeter-wave radar operates in the overlapping region of far-infrared and microwave frequencies, so there is no privacy issue.

[0095] Third, the data processing solution can greatly reduce water interference, and is easy to operate, highly portable, and low in cost.

[0096] Fourth, the method provided in this embodiment of the invention is based on processing the point cloud height to obtain the judgment criteria, and compared with the Doppler scheme, it is not affected by the radar installation location.

[0097] The application extracts the center of mass height of the human body and the water flow for each frame of the associated point cloud of the target track, obtains a center of mass height time series of the human body and the water flow, judges whether it is a non-standing state based on the center of mass height time series of the human body and the water flow, performs tracing analysis based on the center of mass height array of the human body and the water flow in the non-standing state, judges whether the center of mass height of the human body and the water flow decreases, and if the center of mass height of the human body and the water flow decreases, it is judged that the human body falls. The method provided in the embodiment of the application analyzes the human and the water in the shower state as a whole, solves the influence of the water in the related art based on the radar in the shower state fall judgment process, and improves the judgment accuracy.

[0098] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiment of the application.

[0099] The following is the device embodiment of the application, and for the details not described in detail, reference can be made to the corresponding method embodiments described above.

[0100] Figure 2 The structure schematic diagram of the radar-based bathroom fall detection device provided by the embodiment of the application is shown, only the parts related to the embodiment of the application are shown for the convenience of description, and the details are described as follows:

[0101] As Figure 2 shown, the radar-based bathroom fall detection device 2 comprises an acquisition module 21, a first determination module 22, a first judgment module 23, a second judgment module 24 and a second determination module 25;

[0102] The acquisition module 21 is used for determining the track of the target based on the point cloud data detected by the radar, and acquiring the associated point cloud related to the track of the target frame by frame;

[0103] The first determination module 22 is used for determining the center of mass height of the human body and the water flow of each frame of the associated point cloud;

[0104] The first judgment module 23 is used for judging whether it is a non-standing state according to the center of mass height of the human body and the water flow corresponding to the associated point cloud of the continuous first preset frame number;

[0105] The second judgment module 24 is used for judging whether the center of mass height of the human body and the water flow decreases according to the historical center of mass height of the human body and the water flow in the non-standing state;

[0106] The second determination module 25 is used for determining that the human body falls if the center of mass height of the human body and the water flow decreases.

[0107] The application extracts the center of mass height of the human body and the water flow for each frame of the target track, obtains a center of mass height time series of the human body and the water flow, judges whether it is a non-standing state based on the center of mass height time series of the human body and the water flow, performs a trace analysis based on the center of mass height array of the human body and the water flow in the non-standing state, judges whether the center of mass height of the human body and the water flow decreases, and judges that the human body falls if the center of mass height of the human body and the water flow decreases. The method provided in the embodiment of the application analyzes the human and the water in the shower state as a whole, solves the influence of the water in the related art during the shower state fall judgment based on the radar, and improves the judgment accuracy.

[0108] In a possible implementation, the first determination module 22 is configured to:

[0109] obtain the amplitude value and the height value of each point cloud in the frame associated point cloud;

[0110] for any point cloud, weight the height value of the point cloud based on the amplitude value of the point cloud and the amplitude value of each point cloud in the frame associated point cloud, to obtain a weighted point cloud height, so that the weight of the height value of the human body point cloud in the frame associated point cloud increases, and the weight of the height value of the water flow point cloud in the frame associated point cloud decreases;

[0111] calculate the center of mass height of the human body and the water flow of the frame associated point cloud according to the height value of each weighted point cloud.

[0112] In a possible implementation, the first determination module 22 is configured to:

[0113] calculate the center of mass height of the human body and the water flow of the frame associated point cloud according to a preset formula, and the preset formula is

[0114]

[0115] wherein zCenter is used to represent the center of mass height of the human body and the water flow of the frame associated point cloud, N is used to represent the number of point clouds of the frame associated point cloud, snr(i) is used to represent the point cloud amplitude value of the ith point cloud, z(i) is used to represent the point cloud height value of the ith point cloud, is used to represent the height value of the ith weighted point cloud.

[0116] In a possible implementation, the first determination module 23 is configured to:

[0117] if the average value of the center of mass height of the human body and the water flow corresponding to the associated point clouds of the first preset number of continuous frames is greater than the first preset threshold value and the standard deviation is less than the second preset threshold value, the non-standing state is judged.

[0118] In a possible implementation, the second determination module 24 is configured to:

[0119] If the height of the center of mass of the human body and the water flow in the first target frame associated point cloud is less than the third preset threshold, and the difference between the height of the center of mass of the human body and the water flow in the second target frame associated point cloud and the height of the center of mass of the human body and the water flow in the first target frame associated point cloud is greater than the fourth preset threshold, it is judged that the height of the center of mass of the human body and the water flow decreases, wherein the first target frame is located after the second target frame, the first target frame and the second target frame are separated by a second preset frame number, and the time length corresponding to the second preset frame number is determined by the time length taken by the human body to fall.

[0120] In a possible implementation, the second determination module 25 is further configured to:

[0121] If the height of the center of mass of the human body and the water flow does not decrease, it is determined whether there is a low weak signal according to the height of the center of mass of the human body and the water flow in at least one frame of associated point cloud.

[0122] If there is a low weak signal, it is determined that the human body falls.

[0123] In a possible implementation, the second determination module 25 is further configured to:

[0124] If the height of the center of mass of the human body and the water flow in the last frame of associated point cloud in the first continuous first preset frame number of associated point cloud is less than the fifth preset threshold, it is judged that there is activity in the low place.

[0125] Alternatively, if it is judged that the height of the center of mass of the human body and the water flow does not decrease, the height of the center of mass of the human body and the water flow in the associated point cloud of the continuous third preset frame number is obtained, if the number of frames in which the height of the center of mass of the human body and the water flow is less than the sixth preset threshold in the continuous third preset frame number of associated point cloud is greater than a preset percentage, it is judged that there is activity in the low place, and the continuous third preset frame number of associated point cloud is obtained after the continuous first preset frame number of associated point cloud.

[0126] The radar-based bathroom fall detection device provided in the embodiment can be used to execute the radar-based bathroom fall detection method provided in the embodiment, and has similar implementation principles and technical effects, which will not be described here.

[0127] Figure 3 is a schematic diagram of a radar provided in an embodiment of the present application. As shown in Figure 3 The radar 3 in the embodiment includes a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30. The processor 30 implements the steps in each of the above radar-based bathroom fall detection method embodiments when executing the computer program 32, for example Figure 1The steps 101-105 are shown. Alternatively, the processor 30 implements the functions of the modules / units in the above-mentioned device embodiments when executing the computer program 32, for example Figure 2 The functions of the modules 21-25 are shown.

[0128] The computer program 32 can be divided into one or more modules / units for example, which are stored in the memory 31 and executed by the processor 30 to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program 32 in the radar 3.

[0129] The radar 3 can be a millimeter wave radar. The radar 3 can include, but is not limited to, the processor 30, the memory 31. Those skilled in the art can understand that the radar 3 can include more or fewer components than those shown, or combine certain components, or include different components, for example, the radar can also include an input / output device, a network access device, a bus, etc. Figure 3 The radar 3 is only an example and does not constitute a limitation on the radar 3, and can include more or fewer components than those shown, or combine certain components, or include different components, for example, the radar can also include an input / output device, a network access device, a bus, etc.

[0130] The processor 30 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic components, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0131] The memory 31 can be an internal storage unit of the radar 3, such as a hard disk or a memory of the radar 3. The memory 31 can also be an external storage device of the radar 3, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 31 can include both an internal storage unit and an external storage device of the radar 3. The memory 31 is used to store the computer program and other programs and data required by the radar. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0132] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of each functional unit and module is exemplified, and in actual application, the above-mentioned functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be realized in the form of hardware or software function unit. In addition, the specific name of each functional unit and module is only for the convenience of mutual distinction, and does not limit the protection scope of the present application. The specific working process of the unit and module in the above system can refer to the corresponding process in the foregoing method embodiment, which will not be described here.

[0133] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in detail in a certain embodiment can be referred to the related description of other embodiments.

[0134] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0135] In the embodiments provided by the present application, it should be understood that the disclosed device / radar and method can be implemented by other ways. For example, the device / radar embodiments described above are only schematic, and for example, the division of the modules or units is only a logical function division, and there can be another division way in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different parts can be indirect couplings or communication connections through some interfaces, devices or units, and can be electrical, mechanical or in other forms.

[0136] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0137] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0138] The integrated module / unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. When the computer program is executed by a processor, the steps of each radar-based bathroom fall detection method embodiment described above can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0139] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present application, and should be included in the protection scope of the present application.

Claims

1. A radar-based method for detecting falls in bathrooms, characterized in that, include: The target's trajectory is determined based on the point cloud data detected by radar, and the associated point cloud associated with the target's trajectory is obtained; Determine the centroid height of the human body and water flow in each frame of the associated point cloud; Based on the centroid height of the human body and water flow corresponding to the associated point cloud of the first preset number of consecutive frames, determine whether it is a non-standing state. In a non-standing state, based on the historical center of mass height of the human body and the water flow, determine whether the center of mass height of the human body and the water flow has decreased; If the height of the center of mass of the human body and the water flow decreases, it is determined that the human body has fallen. The determination of the centroid height of the human body and water flow in each frame of the associated point cloud includes: For each frame of associated point cloud, obtain the amplitude and height values ​​of each point cloud in that frame of associated point cloud; For any point cloud, the height value of the point cloud is weighted based on the amplitude value of the point cloud and the amplitude value of each point cloud in the associated point cloud of the frame, so as to increase the weight of the height value of the human body point cloud in the associated point cloud of the frame and decrease the weight of the height value of the water flow point cloud in the associated point cloud of the frame. Based on the weighted height value of each point cloud, calculate the centroid height of the human body and water flow in the associated point cloud of that frame; The step of determining whether the person is not standing based on the centroid height of the human body and water flow corresponding to the associated point cloud of a consecutive first preset number of frames includes: If the average height of the centroid of the human body and water flow corresponding to the associated point cloud of a consecutive first preset number of frames is greater than the first preset threshold and the standard deviation is less than the second preset threshold, then it is judged to be a non-standing state. Determining whether the center of mass of the human body and water flow has decreased based on historical data includes: If the centroid height of the human body and water flow in the historical human body and water flow associated point cloud of the first target frame is less than the third preset threshold, and the difference between the centroid height of the human body and water flow in the second target frame associated point cloud and the centroid height of the human body and water flow in the first target frame associated point cloud is greater than the fourth preset threshold, then it is determined that the centroid height of the human body and water flow has decreased. Here, the first target frame is located after the second target frame, and the first target frame is separated from the second target frame by a second preset number of frames. The duration corresponding to the second preset number of frames is determined by the duration of the human body falling.

2. The method according to claim 1, characterized in that, The step of calculating the centroid height of the human body and water flow in the associated point cloud of the frame based on the weighted height value of each point cloud includes: The centroid heights of the human body and water flow in the associated point cloud of this frame are calculated according to a preset formula, wherein the preset formula is: in, The height of the centroid of the human body and water flow is used to represent the associated point cloud of this frame. Used to indicate the number of point clouds associated with this frame. Used to indicate the first The point cloud amplitude value of a point cloud. Used to indicate the first The point cloud height value of a point cloud. Used to represent the weighted i-th The height value of a point cloud.

3. The method according to any one of claims 1 to 2, characterized in that, After determining whether the center of mass height of the human body and the water flow has decreased, the method also includes: If the height of the centroid of the human body and the water flow does not decrease, then determine whether there is a weak signal at a low location based on the height of the centroid of the human body and the water flow in at least one frame of associated point cloud. If a weak signal is detected at a low location, it confirms that the person has fallen.

4. The method according to claim 3, characterized in that, If the centroid heights of the human body and the water flow do not decrease, then determining whether there is a weak signal at a low altitude based on the centroid heights of the human body and the water flow from at least one frame of associated point cloud includes: If the centroid height of the human body and water flow in the last frame of the associated point cloud of the first preset number of consecutive frames is less than the fifth preset threshold, then it is determined that there is human activity at a lower location. Alternatively, if it is determined that the centroid height of the human body and the water flow has not decreased, the centroid height of the human body and the water flow in the associated point cloud for a consecutive third preset number of frames is obtained. If the number of frames in the associated point cloud for a consecutive third preset number of frames where the centroid height of the human body and the water flow is less than the sixth preset threshold is greater than a preset percentage, it is determined that there is activity at a lower location. The associated point cloud for a consecutive third preset number of frames is obtained after the associated point cloud for a consecutive first preset number of frames.

5. A radar-based bathroom fall detection device, characterized in that, include: The module includes an acquisition module, a first determination module, a first judgment module, a second judgment module, and a second determination module. The acquisition module is used to determine the target's trajectory based on the point cloud data detected by radar, and to acquire the associated point cloud related to the target's trajectory; The first determining module is used to determine the centroid height of the human body and water flow in each frame of the associated point cloud; The first judgment module is used to determine whether the person is not standing based on the centroid height of the human body and water flow corresponding to the associated point cloud of a consecutive first preset number of frames. The second judgment module is used to determine, in a non-standing state, whether the center of gravity height of the human body and the water flow has decreased based on the historical center of gravity height of the human body and the water flow. The second determining module is used to determine if the height of the center of mass of the human body and the water flow decreases; The determination of the centroid height of the human body and water flow in each frame of the associated point cloud includes: For each frame of associated point cloud, obtain the amplitude and height values ​​of each point cloud in that frame of associated point cloud; For any point cloud, the height value of the point cloud is weighted based on the amplitude value of the point cloud and the amplitude value of each point cloud in the associated point cloud of the frame, so as to increase the weight of the height value of the human body point cloud in the associated point cloud of the frame and decrease the weight of the height value of the water flow point cloud in the associated point cloud of the frame. Based on the weighted height value of each point cloud, calculate the centroid height of the human body and water flow in the associated point cloud of that frame; The step of determining whether the person is not standing based on the centroid height of the human body and water flow corresponding to the associated point cloud of a consecutive first preset number of frames includes: If the average height of the centroid of the human body and water flow corresponding to the associated point cloud of a consecutive first preset number of frames is greater than the first preset threshold and the standard deviation is less than the second preset threshold, then it is judged to be a non-standing state. Determining whether the center of mass of the human body and water flow has decreased based on historical data includes: If the centroid height of the human body and water flow in the historical human body and water flow associated point cloud of the first target frame is less than the third preset threshold, and the difference between the centroid height of the human body and water flow in the second target frame associated point cloud and the centroid height of the human body and water flow in the first target frame associated point cloud is greater than the fourth preset threshold, then it is determined that the centroid height of the human body and water flow has decreased. Here, the first target frame is located after the second target frame, and the first target frame is separated from the second target frame by a second preset number of frames. The duration corresponding to the second preset number of frames is determined by the duration of the human body falling.

6. A radar comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 4 above.

7. The radar according to claim 6, characterized in that, The radar in question is a millimeter-wave radar.

Images