Human body tumble detection method and device in bathroom environment and readable storage medium

Abstract

The invention provides a human body tumble detection method and device in a bathroom environment and a readable storage medium. The method comprises the following steps: extracting Doppler spectrum width features from a distance-Doppler spectrum corresponding to a bathroom radar echo signal; wherein the Doppler spectrum width feature is used for representing the breadth of speed distribution of the detection target; determining whether shower equipment in the current bathroom is in a working state or not based on the Doppler spectrum width features; if the current shower equipment is in the working state, generating a 4D point cloud image representing human body action information based on a distance-Doppler spectrum; and determining a detection result of the human body falling event according to the 4D point cloud image. Through the implementation of the scheme of the invention, the Doppler spectrum width features correspondingly extracted by the radar echo signals are utilized to judge the water flow condition of the shower head in the bathroom, the human body target and the water flow target are distinguished, the human body action is identified based on the 4D point cloud imaging technology, and high-precision detection of human body tumble can be realized.

Description

Technical Field

[0001] The present application relates to the field of radar data processing technology, and in particular to a method, device, and readable storage medium for detecting a human fall in a bathroom environment. Background Art

[0002] Falls are a common occurrence in the daily lives of the elderly due to the deterioration of their physical functions. If falls can be detected promptly and treated effectively, the health risks and pressure on medical resources can be reduced.

[0003] In bathroom showers, the risk of falls for the elderly is significantly increased due to the slippery floor, pervasive steam, confined space, and limited support. When a fall occurs, it's often difficult for outsiders to detect it and provide timely assistance due to the bathroom's privacy and soundproofing. Therefore, developing a fall detection system capable of real-time monitoring in this high-risk bathroom setting has significant practical and social value.

[0004] In the relevant technology, common fall detection devices can be mainly divided into two types: contact type (such as accelerometers, gyroscopes, etc.) and non-contact type (such as visual sensors, audio sensors, infrared sensors, Wi-Fi devices, etc.). Although contact-type sensors are highly sensitive and have strong real-time performance, they have disadvantages such as needing to be worn and poor comfort for long periods of time. While most non-contact sensors can avoid direct contact with users, they have problems such as privacy leakage and susceptibility to environmental influences. Non-contact radar sensors have the advantages of 24 / 7 operation and privacy protection, and are unaffected by external factors such as light, temperature, and sound. However, radar signals obtained in bathroom environments are easily affected by environmental factors such as shower water flow, resulting in low accuracy in detecting human fall movements in bathroom environments. Summary of the Invention

[0005] The embodiments of the present application provide a method, device, and readable storage medium for detecting a human fall in a bathroom environment, which can at least solve the problem of low accuracy in detecting a human fall based on radar signals in a bathroom environment.

[0006] A first aspect of an embodiment of the present application provides a method for detecting a human fall in a bathroom environment, comprising: The Doppler spectrum width feature is extracted from the range-Doppler spectrum corresponding to the bathroom radar echo signal. The Doppler spectrum width feature is used to characterize the width of the velocity distribution of the detected target. Based on the Doppler spectrum width characteristics, determine whether the shower device in the current bathroom is in working state; If the shower device is currently in working state, a 4D point cloud image representing human motion information is generated based on the range-Doppler spectrum; Based on the 4D point cloud image, the detection result of the human fall event is determined.

[0007] A second aspect of an embodiment of the present application provides an electronic device, comprising: a memory and a processor, wherein the processor is configured to execute a computer program stored in the memory. When the processor executes the computer program, the processor implements the steps of the method for detecting a human fall in a bathroom environment provided in the first aspect of the embodiment of the present application.

[0008] A third aspect of an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method for detecting a human fall in a bathroom environment provided in the first aspect of the embodiment of the present application are implemented.

[0009] As can be seen from the above, according to the human fall detection method, device, and readable storage medium in a bathroom environment provided by the present application, a Doppler spectrum width feature is extracted from the range-Doppler spectrum corresponding to the bathroom radar echo signal; wherein the Doppler spectrum width feature is used to characterize the breadth of the detection target velocity distribution; based on the Doppler spectrum width feature, it is determined whether the shower device in the current bathroom is in an operating state; if the current shower device is in an operating state, a 4D point cloud image representing human motion information is generated based on the range-Doppler spectrum; and based on the 4D point cloud image, a detection result of a human fall event is determined. Through the implementation of the present application, the on and off states of the shower in the bathroom environment are identified based on the range-Doppler spectrum of the radar echo signal. When it is determined that the shower device is in an operating state, that is, when there is a continuous shower flow in the bathroom, 4D human body point cloud imaging technology is used to identify the specific posture adjustment state of the human body; such data processing can effectively suppress the interference of the shower device shower water flow on human motion recognition, greatly improving the accuracy of human fall detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of the basic flow of a method for detecting a human fall in a bathroom environment provided in the first embodiment of the present application; Figure 2 The distance-Doppler spectrum when the bathroom is empty and the shower is closed; Figure 3 The distance-Doppler spectrum when the bathroom is empty and the shower is in working condition; Figure 4 The distance-Doppler spectrum when the bathroom is occupied and the shower is closed; Figure 5 The distance-Doppler spectrum when there are people in the bathroom and the shower equipment is in working condition; Figure 6 The range-Doppler spectrum to be processed in the first embodiment of the present application; Figure 7 This is the spectrum corresponding to the statistical matrix in the first embodiment of this application; Figure 8 The range-Doppler spectrum with labels in the first embodiment of the present application; Figure 9 A schematic diagram of a detailed flow chart of a method for detecting a human fall in a bathroom environment provided in the second embodiment of the present application; Figure 10 A schematic structural diagram of an electronic device provided in the third embodiment of the present application. DETAILED DESCRIPTION

[0011] In order to make the purpose, features, and advantages of the invention of this application more obvious and easy to understand, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of this application.

[0012] In the description of the embodiments of the present application, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0013] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of this application, the meaning of "plurality" is two or more, unless otherwise clearly specified.

[0014] In the embodiments of the present application, unless otherwise expressly specified or limited, the terms "installed," "connected," "connected," "fixed," etc. should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integration; mechanical connections, electrical connections; direct connections, or indirect connections through an intermediate medium; and can refer to internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the embodiments of the present application based on specific circumstances.

[0015] In order to solve the problem of low accuracy in detecting human fall actions based on radar signals in the related art, the first embodiment of the present application provides a method for detecting human fall in a bathroom environment, such as Figure 1 This is a basic flow chart of the human fall detection method in a bathroom environment provided in this embodiment. The human fall detection method in a bathroom environment includes the following steps: Step 101: extracting a Doppler spectrum width feature from a range-Doppler spectrum corresponding to a bathroom radar echo signal.

[0016] Specifically, the Doppler spectrum width feature is used to characterize the breadth of the target velocity distribution. In this embodiment, in a bathroom shower environment, millimeter-wave radar can be used to detect human falls and distinguish between showerhead status (on / off). When the shower is on, the continuous flow of water causes the Doppler frequency distribution to become wider, increasing the Doppler spectrum width feature value. When the shower is off but the user makes a quick movement, a brief period of high-speed movement occurs, which may also produce a larger Doppler spectrum width feature. Therefore, when detecting human falls, it is necessary to first eliminate the interference of the showerhead water flow on the detection.

[0017] Step 102: Determine whether the shower equipment in the current bathroom is in working state based on the Doppler spectrum width feature.

[0018] Specifically, if the shower in the bathroom is on, 4D point cloud imaging and other technologies can be used to further identify falls. If the shower is off, conventional fall detection methods (such as posture angle change analysis) can be used directly to detect falls, ensuring detection accuracy while reducing detection processing complexity. Of course, in other implementations, even when the shower is off, 4D point cloud imaging and other technologies can still be used for fall detection, and this is not a limitation here.

[0019] Step 103: If the shower device is currently in working state, a 4D point cloud image representing human motion information is generated based on the range-Doppler spectrum.

[0020] Specifically, millimeter-wave radar information is used to image human body movements into 4D point clouds, which can obtain information such as the target's distance, azimuth, pitch angle, and speed. By constructing dynamic point cloud images, the movement of the human body can be continuously tracked from the time and space dimensions, and the movement characteristics of the human body can be accurately captured, providing high-precision data support for subsequent analysis of human behavior patterns.

[0021] Step 104: Determine the detection result of the human fall event based on the 4D point cloud image.

[0022] Specifically, in the process of analyzing human behavior patterns based on 4D point cloud images, it is possible to first determine whether the human body is in a low posture state; if the human body is in a low posture state, the point cloud image features can be further extracted to perform human fall recognition.

[0023] In one implementation of this embodiment, the above-mentioned radar echo signal is a discrete echo signal associated with a slow-time pile, a fast-time pile, and an antenna pile; accordingly, before the step of extracting the Doppler spectrum width feature from the range-Doppler spectrum corresponding to the bathroom radar echo signal, it also includes: performing an FFT transform on the bathroom radar echo signal in the fast time dimension to generate an initial range-dimensional spectrum; performing static clutter suppression processing on the range-dimensional spectrum to obtain a target range-dimensional spectrum; performing an FFT transform on the target range-dimensional spectrum in a slow-time pile to obtain range-Doppler spectrum data; performing incoherent accumulation processing on the range-Doppler spectrum data to output a range-Doppler spectrum.

[0024] Specifically, the radar continuously transmits electromagnetic wave signals into space. The electromagnetic wave signals are scattered or reflected by objects and then received by the radar receiver. After passing through the signal amplifier, mixer and ADC sampling, a discrete echo signal containing distance, speed and angle dimension information is obtained. The echo signal received by the radar module can be expressed as ,in, is the slow time dimension, indicating the A linear frequency modulated continuous wave signal; is the fast time dimension, indicating the sampling points; is the antenna dimension, indicating the The received signal of each channel is processed by fast Fourier transform along the fast time and slow time of the echo signal, and static clutter suppression and incoherent accumulation are performed to obtain the range-Doppler spectrum containing the target information. .

[0025] It should be noted that the distance-Doppler spectra in the bathroom under different conditions have different spectrum width characteristics, such as Figure 2-5 As shown in the figure, by comparing the two, we can see that: (1) the numerical difference between the shower water interference and the human target in the range-Doppler spectrum is small, which is due to the similar radar cross-section (RCS) of the two; (2) the shower water interference is distributed over a wide range in the spectrum and is distributed in the positive Doppler frequency range, while the human target is more concentrated near the zero Doppler frequency in the spectrum. Therefore, the shower status can be detected based on the range-Doppler spectrum to determine whether the shower equipment in the bathroom is in working condition.

[0026] In one implementation of this embodiment, the Doppler spectrum width feature may be extracted from the range-Doppler spectrum; the Doppler spectrum width feature may be expressed as:

[0027]

[0028]

[0029] in, represents the Doppler spectrum width characteristic, represents the range-Doppler spectrum, represents the index of the Doppler unit, Represents the index of the distance unit, represents the mean coefficient, represents the number of Doppler units, Indicates the number of distance units.

[0030] Furthermore, in one implementation of this embodiment, the step of determining whether the shower device in the current bathroom is in an operating state based on the Doppler spectrum width feature includes: performing sliding window detection on the range-Doppler spectrum based on the Doppler spectrum width feature; and determining that the shower device in the current bathroom is in an operating state when the number of target data frames that meet a first preset condition within the time window is greater than a preset frame number threshold; wherein the target data is range-Doppler spectrum data, and the first preset condition is: the Doppler spectrum width feature corresponding to the target data is greater than a preset feature threshold.

[0031] Specifically, when When , it indicates that there is a large Doppler bandwidth at this time, where is the preset feature threshold. For the Doppler spectrum width feature, the feature value is sometimes larger when the shower is turned off. This may be due to the continuous Doppler frequency component caused by the human body squatting or swinging the arms quickly. Therefore, this embodiment detects the shower state by sliding the window. The frame data contains satisfy When the shower equipment is in working condition, Indicates the window length, Indicates the trigger threshold.

[0032] In one implementation of this embodiment, the step of generating a 4D point cloud image representing human motion information based on the range-Doppler spectrum includes: detecting interference distribution information corresponding to the shower water flow of the shower device in the range-Doppler spectrum, and removing the interference distribution information from the range-Doppler spectrum to obtain a target range-Doppler signal; and processing the target range-Doppler signal based on a digital beamforming algorithm to obtain a 4D point cloud image representing human motion information.

[0033] Specifically, due to the interference of shower water flow, before performing point cloud imaging, it is necessary to first determine the corresponding information of the water flow target and the human target, and separate the water flow target and the human target to suppress the interference of the shower water flow on recognition and improve the accuracy of human fall recognition.

[0034] Furthermore, in one implementation of this embodiment, the step of detecting interference distribution information corresponding to the shower head water flow of the shower device in the range-Doppler spectrum includes: creating a statistical matrix corresponding to the range-Doppler spectrum; wherein the statistical matrix is ​​the same size as the range-Doppler spectrum; based on the statistical matrix, detecting interference targets for each frame of data of the range-Doppler spectrum, determining all target positions where the interference targets appear in the range-Doppler spectrum and the frequency of occurrence of the interference targets in each target position; marking the range-Doppler spectrum in combination with the target position and frequency to obtain a marked range-Doppler spectrum; and determining the interference distribution information corresponding to the shower head water flow of the shower device based on the marking information in the marked range-Doppler spectrum.

[0035] For example, when performing 4D point cloud imaging, the following steps may be followed: (1) Create a statistical matrix with the same size as the range-Doppler spectrum and the initial value of the element is zero , and set the statistical duration to ; Among them, the range-Doppler spectrum is as follows Figure 6 As shown, the statistical matrix like Figure 7 As shown; (2) A two-dimensional OS-CFAR detector is used to detect the target for each frame of the range-Doppler spectrum. The detected interference target can be expressed as ,in represents the Doppler index of the interference target, Represents the distance index, is the number of interference targets detected. The detection results within the time are re-counted, and the frequency of interference targets appearing at each position in the range-Doppler spectrum is recorded. ; (3) In The breadth-first search algorithm is used to find and mark different connected areas and obtain the marked distance-Doppler spectrum. Since the shower water flow target in the shower state is widely distributed on the distance-Doppler spectrum, the connected domain with the largest area is the interference distribution of the shower water flow. ,like Figure 8 As shown; (4) The interference distribution information corresponding to the shower water flow is converted from Eliminate the target range-Doppler signal from the human target information to obtain the target range-Doppler signal containing only the human target information ; (5) Based on the target range-Doppler signal, the Digital Beamforming (DBF) algorithm is used to estimate the angle of the detected target unit; after coordinate transformation, a 4D point cloud image containing rich human body physical information is obtained. ,in, Represents the three-dimensional Cartesian coordinate system Axis coordinates, Represents the three-dimensional Cartesian coordinate system Axis coordinates, Represents the three-dimensional Cartesian coordinate system Axis coordinates, Indicates speed, Indicates energy.

[0036] In one implementation of this embodiment, the above-mentioned step of determining the detection result of the human fall event based on the 4D point cloud image includes: judging whether the user performs a specific posture adjustment within a preset time based on the 4D point cloud image; wherein the process of the specific posture adjustment is a process of changing from a high posture to a low posture; if it is determined that the user performs a specific posture adjustment within the preset time, then determining the target point cloud image corresponding to the process of the user performing the specific posture adjustment; respectively extracting the width information and height information of the point cloud distribution in each frame of the target point cloud image; and combining all the width information and height information to determine the detection result of the human fall event.

[0037] Specifically, the process of specific posture adjustment can refer to the process of a human target changing from a high posture, such as standing or walking, to a low posture, such as falling or squatting. The most obvious change in height during a fall is the change in height. Common human movements, such as sitting, squatting, and bending over to pick up something, also exhibit similar characteristics. Therefore, before identifying a fall, specific posture adjustment (i.e., changing from a high specific posture to a low posture) can be detected and judged.

[0038] Furthermore, in one implementation of the present embodiment, the above-mentioned step of judging whether the user has made a specific posture adjustment within a preset time based on the 4D point cloud image includes: determining the associated point cloud number sequence and the center of mass height sequence of the 4D point cloud image; performing sliding window detection on the 4D point cloud image based on the associated point cloud number sequence and the center of mass height sequence; when the associated point cloud number and the center of mass height corresponding to the target area in the timing window meet the second preset condition, determining that the user has made a specific posture adjustment within the preset time; wherein the target area includes a first area and a second area, and the first area is later than the second area in time sequence; the second preset condition includes: the proportion of data frames in the first area of ​​the timing window whose center of mass height is lower than the preset height threshold is greater than a first proportion threshold, the proportion of data frames in the first area whose associated point cloud number is higher than the preset point number threshold is greater than a second proportion threshold, and the proportion of data frames in the second area of ​​the timing window whose center of mass height is higher than the preset height threshold is greater than a third proportion threshold.

[0039] Specifically, the first region may refer to the rear portion of the time window (e.g., the second half of the time window), and the second region may refer to the front portion of the time window (e.g., the first half of the time window). In a specific implementation, a determination may be made as to whether the number of data frames in the rear portion of the time window where the center of mass height is below a height threshold (i.e., the aforementioned preset height threshold) exceeds a first ratio threshold. If so, the human target is considered to be in a low posture. Next, a determination is made as to whether the number of data frames in the rear portion of the time window where the number of associated point clouds exceeds a point count threshold (i.e., the aforementioned preset point count threshold) exceeds a second ratio threshold. If so, the human target is considered to be present; otherwise, the human target is considered to have left the detection area. Finally, a determination is made as to whether the number of data frames in the front portion of the time window where the center of mass height is above the height threshold exceeds a third ratio threshold. If so, the human target is considered to be in a high posture. If all three of these determinations are met, the human target is considered to have transitioned from a high posture to a low posture.

[0040] Furthermore, in one implementation of the present embodiment, the above-mentioned step of combining all width information and height information to determine the detection result of a human fall event includes: determining the frame-level features of the target point cloud image by combining the width information and the height information; wherein the frame-level features include the width distribution of the point cloud on the x-axis, the width distribution on the y-axis, the maximum height on the z-axis, and the average height of the point cloud; extracting extended features and height features from the frame-level features; wherein the extended features are used to characterize the distribution changes of the human body projection in the horizontal direction, and the height features are used to characterize the distribution changes of the human body projection in the height direction; and inputting the extended features and the height features into a preset classifier to output the detection result of the human fall event.

[0041] Specifically, for example, the distribution width information and height information of each frame of point cloud can be expressed as:

[0042]

[0043]

[0044]

[0045] in, Indicates the Frame point cloud along The width of the distribution of the axis, Indicates the Frame point cloud along The width of the distribution of the axis, Indicates the The maximum height of the frame point cloud, Indicates the The average height of the frame point cloud, is the frame index, , is the data duration used for fall status recognition, Represents the number of points in a single frame point cloud, .

[0046] When a person falls and lies on the ground, their height remains low for a long time, and the horizontal extension of their trunk and limbs increases significantly. In contrast, squatting, bending over, and picking up objects also cause a decrease in height, but the height after the decrease is still higher than the height of the fall, and the trunk and limbs do not expand significantly in the horizontal direction.

[0047] Therefore, it can be obtained from 、 、 Extract extended features 、 、 、 :

[0048]

[0049]

[0050]

[0051] And, from and Extract height features and :

[0052]

[0053] in, express The normalized ratio of the axis width to the maximum height, express The normalized ratio of the axis width to the maximum height, Indicates the distribution range of the horizontal plane projection (Euclidean distance), It represents the normalized ratio of the horizontal distribution range to the maximum height. represents the median of the maximum height, Indicates the median of the average height.

[0054] Based on the extracted extended features and height features, the SVM classifier is used to identify the human fall state. The SVM classifier is suitable for binary classification tasks under small samples, can effectively handle nonlinear feature relationships, and improve the accuracy of human fall state judgment.

[0055] Based on the technical solution of the embodiment of the present application described above, a Doppler spectrum width feature is extracted from the range-Doppler spectrum corresponding to the bathroom radar echo signal; wherein the Doppler spectrum width feature is used to characterize the breadth of the velocity distribution of the detected target; based on the Doppler spectrum width feature, it is determined whether the shower equipment in the current bathroom is in an operating state; if the current shower equipment is in an operating state, a 4D point cloud image representing human motion information is generated based on the range-Doppler spectrum; and based on the 4D point cloud image, the detection result of the human fall event is determined. Through the implementation of the solution of the present application, the on and off states of the shower in the bathroom environment are identified on the range-Doppler spectrum of the radar echo signal. When it is determined that the shower equipment shower is in an operating state, that is, when there is a continuous shower water flow in the bathroom, 4D human body point cloud imaging technology is used to identify the specific posture adjustment state of the human body. This can effectively suppress the interference of the shower equipment shower water flow on human motion recognition, greatly improving the accuracy of human fall detection.

[0056] Figure 9 The method in the second embodiment of the present application is a refined method for detecting human falls in a bathroom environment, and the method for detecting human falls in a bathroom environment includes: Step 901: Perform range-Doppler imaging on the radar echo signal of the bathroom environment to obtain a range-Doppler spectrum.

[0057] Step 902: Extract Doppler spectrum width features from the range-Doppler spectrum.

[0058] Step 903: Perform sliding window detection on the range-Doppler spectrum based on the Doppler spectrum width feature.

[0059] Step 904: When the number of target data frames that meet the first preset condition within the timing window is greater than a preset frame number threshold, it is determined that the shower device in the current bathroom is in a working state.

[0060] Specifically, in this embodiment, the target data is range-Doppler spectrum data, and the first preset condition is that the Doppler spectrum width feature corresponding to the target data is greater than a preset feature threshold.

[0061] Step 905: Detect interference distribution information corresponding to the shower water flow in the range-Doppler spectrum, and remove the interference distribution information from the range-Doppler spectrum to obtain a target range-Doppler signal.

[0062] Step 906: Process the target range-Doppler signal based on a digital beamforming algorithm to obtain a 4D point cloud image representing human motion information.

[0063] Step 907: Determine whether the user has made a specific posture adjustment within a preset time based on the 4D point cloud image.

[0064] Specifically, the process of adjusting the specific posture is a process of changing from a high posture to a low posture.

[0065] Step 908: If it is determined that the user has made a specific posture adjustment within the preset time, a target point cloud image corresponding to the process of the user making the specific posture adjustment is determined.

[0066] Step 909 : Determine a detection result of a human fall event based on the width information and height information of the point cloud distribution in the target point cloud image.

[0067] Specifically, the width information and height information can be combined to determine the frame-level features of the target point cloud image; wherein the frame-level features include the width distribution of the point cloud on the x-axis, the width distribution on the y-axis, the maximum height on the z-axis, and the average height of the point cloud; the extended features and the height features are extracted from the frame-level features; wherein the extended features are used to characterize the distribution changes of the human body projection in the horizontal direction, and the height features are used to characterize the distribution changes of the human body projection in the height direction; the extended features and the height features are input into a preset classifier to output the detection results of the human fall event.

[0068] In this embodiment, the radar echo signal of the bathroom environment is imaged by range-Doppler imaging, and the shower status is detected; after determining that the device is in the working state, the human target and the water flow target are separated on the range-Doppler spectrum, and 4D human point cloud imaging is performed; based on the 4D point cloud image, the specific posture adjustment of the user in the bathroom environment is detected to determine whether the human body is in a low posture state, and human fall recognition is completed efficiently. Through the implementation of the scheme of this embodiment, the following beneficial effects can be obtained: (1) The user does not need to contact the sensor device or wear any device; personal privacy is not violated; it is not sensitive to light, dust, smoke and temperature; the device can work all day and all weather; (2) It can work in a variety of complex bathroom shower environments and can penetrate the bathroom glass to achieve robust detection of different fall postures; (3) The computational complexity is low and it can be run on a general ARM processor, which is easy to perform edge computing and perception.

[0069] It should be understood that the size of the serial numbers of the steps in this embodiment does not mean the order in which the steps are executed. The order in which the steps are executed should be determined by their functions and internal logic, and should not constitute a sole limitation on the implementation process of the embodiments of this application.

[0070] Figure 10 This is an electronic device provided in the third embodiment of the present application. This electronic device can be used to implement the human fall detection method in the bathroom environment of the aforementioned embodiment, and mainly includes: Memory 1001, processor 1002, and computer program 1003 stored in memory 1001 and executable on processor 1002. Memory 1001 and processor 1002 are connected via communication. When processor 1002 executes computer program 1003, the method of Embodiment 1 or 2 described above is implemented. The number of processors may be one or more.

[0071] The memory 1001 can be a high-speed random access memory (RAM) memory or a non-volatile memory such as a disk drive. The memory 1001 is used to store executable program code. The processor 1002 is coupled to the memory 1001 .

[0072] Furthermore, the embodiment of the present application also provides a computer-readable storage medium, which can be set in the above electronic device, and the computer-readable storage medium can be the above Figure 10 Memory in the illustrated embodiment.

[0073] The computer-readable storage medium stores a computer program that, when executed by a processor, implements the human fall detection method in a bathroom environment described in the aforementioned embodiment. Furthermore, the computer-readable storage medium may be a USB flash drive, a mobile hard drive, a read-only memory (ROM), RAM, a magnetic disk, or an optical disk, among other media capable of storing program code.

[0074] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules is only a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0075] Modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed across multiple network modules. Some or all of these modules may be selected to achieve the purpose of this embodiment based on actual needs.

[0076] In addition, the functional modules in the various embodiments of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module. The above-mentioned integrated modules may be implemented in the form of hardware or software functional modules.

[0077] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of this application. The aforementioned readable storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks, or optical disks.

[0078] It should be noted that for the aforementioned method embodiments, for ease of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0079] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0080] The above is a description of the human fall detection method, device, and readable storage medium in a bathroom environment provided by this application. For those skilled in the art, based on the ideas of the embodiments of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for detecting human falls in a bathroom environment, characterized in that: include: Extracting a Doppler spectrum width feature from the range-Doppler spectrum corresponding to the bathroom radar echo signal; wherein the Doppler spectrum width feature is used to characterize the breadth of the detected target velocity distribution; determining whether the shower device in the bathroom is currently in operation based on the Doppler spectrum width feature; If the shower device is currently in working state, generating a 4D point cloud image representing human motion information based on the range-Doppler spectrum; A detection result of a human fall event is determined based on the 4D point cloud image.

2. The human fall detection method according to claim 1, characterized in that: The Doppler spectrum width characteristic is expressed as: in, represents the Doppler spectrum width characteristic, represents the range-Doppler spectrum, represents the index of the Doppler unit, Represents the index of the distance unit, represents the mean coefficient, represents the number of Doppler units, Indicates the number of distance units.

3. The human fall detection method according to claim 2, characterized in that: The determining whether the shower device in the bathroom is currently in a working state based on the Doppler spectrum width feature includes: performing sliding window detection on the range-Doppler spectrum based on the Doppler spectrum width feature; When the number of target data frames that meet a first preset condition within the time sequence window is greater than a preset frame number threshold, it is determined that the shower device in the current bathroom is in an operating state; wherein the target data is range-Doppler spectrum data, and the first preset condition is that the Doppler spectrum width feature corresponding to the target data is greater than a preset feature threshold.

4. The human fall detection method according to claim 1, characterized in that: The generating of a 4D point cloud image representing human motion information based on the range-Doppler spectrum includes: detecting interference distribution information corresponding to the shower water flow of the shower device in the range-Doppler spectrum, and removing the interference distribution information from the range-Doppler spectrum to obtain a target range-Doppler signal; The target range-Doppler signal is processed based on a digital beamforming algorithm to obtain a 4D point cloud image representing human motion information.

5. The human fall detection method according to claim 4, characterized in that: The detecting of interference distribution information corresponding to the shower water flow of the shower device in the range-Doppler spectrum includes: Creating a statistical matrix corresponding to the range-Doppler spectrum; wherein the statistical matrix has the same size as the range-Doppler spectrum; Based on the statistical matrix, detecting interference targets for each frame of the range-Doppler spectrum, determining all target positions where interference targets appear in the range-Doppler spectrum and the frequency of occurrence of the interference targets in each target position; marking the range-Doppler spectrum in combination with the target position and the frequency to obtain a marked range-Doppler spectrum; According to the marking information in the marked range-Doppler spectrum, interference distribution information corresponding to the shower head water flow of the shower device is determined.

6. The human fall detection method according to claim 1, characterized in that: Determining a detection result of a human fall event based on the 4D point cloud image includes: Determining whether the user performs a specific posture adjustment within a preset time based on the 4D point cloud image; wherein the specific posture adjustment process is a process of changing from a high posture to a low posture; If it is determined that the user has made a specific posture adjustment within the preset time, then determining a target point cloud image corresponding to the process of the user making the specific posture adjustment; Extracting width information and height information of the point cloud distribution in each frame of the target point cloud image respectively; Combine all the width information and the height information to determine a detection result of a human fall event.

7. The human fall detection method according to claim 6, characterized in that: The determining, based on the 4D point cloud image, whether the user performs a specific posture adjustment within a preset time includes: Determining a sequence of associated point cloud quantities and a sequence of centroid heights of the 4D point cloud image; Performing sliding window detection on the 4D point cloud image based on the sequence of associated point cloud quantities and the sequence of centroid heights; When the number of associated point clouds and the centroid height corresponding to the target area in the time sequence window meet the second preset condition, it is determined that the user has made a specific posture adjustment within the preset time; wherein, the target area includes a first area and a second area, and the first area is later than the second area in time sequence; the second preset condition includes: the proportion of data frames with a centroid height lower than the preset height threshold in the first area of ​​the time sequence window is greater than a first proportion threshold, the proportion of data frames with a number of associated point clouds higher than the preset point number threshold in the first area is greater than a second proportion threshold, and the proportion of data frames with a centroid height higher than the preset height threshold in the second area of ​​the time sequence window is greater than a third proportion threshold.

8. The human fall detection method according to claim 6, characterized in that: Combining all of the width information and the height information to determine a detection result of a human fall event includes: Determining frame-level features of the target point cloud image by combining the width information and the height information; wherein the frame-level features include the width distribution of the point cloud on the x-axis, the width distribution on the y-axis, the maximum height on the z-axis, and the average height of the point cloud; Extracting an extended feature and a height feature from the frame-level feature; wherein the extended feature is used to characterize the distribution change of the human body projection in the horizontal direction, and the height feature is used to characterize the distribution change of the human body projection in the height direction; The extended features and the height features are input into a preset classifier to output a detection result of a human fall event.

9. An electronic device, characterized in that: Comprising a memory and a processor, wherein: The processor is configured to execute a computer program stored in the memory; When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images