Human body balance ability detection method, device and equipment and readable medium

Abstract

The invention provides a human body balance ability detection method, device and equipment and a readable medium, and relates to the technical field of millimeter wave detection. A human body posture is determined by performing visual detection on a subject, a static echo signal of the subject is collected through a millimeter wave radar when the human body posture is static, and then the static balance ability of the subject is analyzed; when the human body posture is dynamic, dynamic echo signals of the subject in a preset dynamic posture are collected through a millimeter wave radar, then point cloud information is generated, and the dynamic balance ability of the subject is analyzed based on the point cloud information; and comprehensively analyzing the human body balance ability of the subject based on the static and dynamic balance ability. The non-contact detection mode of the scheme enables the scheme deployment to be flexible, the detection to be convenient, and the process to be sustainable; static and dynamic balance analysis scenes are distinguished through visual detection, balance detection can be comprehensively and fully carried out on a subject, dependence of artificial experience is avoided, the detection result is standardized, and the detection accuracy is improved.

Description

Technical Field

[0001] This disclosure relates to the field of millimeter-wave detection technology, specifically to a method for detecting human balance ability, a device for detecting human balance ability, an electronic device, and a computer-readable storage medium. Background Technology

[0002] Human balance refers to the body's ability to maintain stability through its own regulatory mechanisms while maintaining a static posture, engaging in voluntary movement, or experiencing external disturbances. Balance ability testing helps identify and prevent individual fall risks in a timely manner, avoiding potential secondary injuries and effectively improving quality of life. The demand for balance ability testing is becoming increasingly common in workplaces, homes, healthcare settings, and elderly care facilities.

[0003] Currently, common methods for testing human balance ability include observation, rating scales, and balance testing devices. Observation involves experienced personnel visually assessing a subject's balance in a specific posture. Rating scales assess a subject's performance on each functional task, such as sitting-to-standing or standing on one leg, by having experienced personnel evaluate and score their performance to obtain balance test results. Balance testing devices utilize pressure sensors to construct force-measuring devices, such as force plates or pressure pads, to obtain changes in pressure distribution or the oscillation of the pressure center in a specific posture, thereby evaluating the subject's balance ability.

[0004] Of the above methods, the observation method relies on the relevant professional knowledge and experience of personnel, the assessment process lacks standardized indicators, and is costly, inefficient, and its accuracy and stability are difficult to guarantee; the scale method obtains corresponding scores by scoring each task and determines the human balance ability based on the scores, but this method still relies on the judgment of personnel experience, has insufficient accuracy, the assessment process is time-consuming, and continuous monitoring cannot be achieved; the balance test instrument assessment method can provide quantitative assessment, but the measurement data is singular, the maintenance and calibration costs are high, the use environment requires a flat ground, and the subject needs to actively measure in a limited area, making continuous measurement impossible and limiting the application scenarios.

[0005] It should be noted that the information in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this disclosure is to provide a method, device, electronic device, and computer-readable storage medium for detecting human balance ability, which can comprehensively assess the human balance ability of a subject in a non-contact setting by combining dynamic and static balance ability with visual detection of human posture and millimeter-wave radar.

[0007] According to a first aspect of this disclosure, a method for detecting human balance ability is provided, comprising: determining the subject's human posture based on a human image of the subject; acquiring static echo signals of the subject using millimeter-wave radar when the human posture is static; analyzing the subject's static balance ability based on the static echo signals; acquiring dynamic echo signals of the subject under a preset dynamic posture using millimeter-wave radar when the human posture is dynamic, and generating point clouds of the subject based on the dynamic echo signals to obtain point cloud information; analyzing the subject's dynamic balance ability based on the point cloud information; and analyzing the subject's corresponding human balance ability based on the static balance ability and dynamic balance ability.

[0008] In an exemplary embodiment, analyzing a subject's static balance ability based on a static echo signal includes: filtering out periodic micro-motion signals from the static echo signal and extracting phase information from consecutive frames to obtain human micro-motion signals corresponding to the static echo signal; extracting static stability parameters of the subject when the human body posture is static based on the human micro-motion signals; and analyzing the subject's static balance ability based on the static stability parameters.

[0009] In an exemplary embodiment, the static echo signal includes a global static echo signal and a local static echo signal. The local static echo signal corresponding to the subject's body position is obtained by combining the human body image, including: obtaining the global static echo signal collected by the millimeter-wave radar from the subject in the frontal view direction; and, based on the human body joint information corresponding to the subject in the human body image, pointing the millimeter-wave radar to the human body position corresponding to the subject to obtain the local static echo signal.

[0010] In one exemplary embodiment, analyzing a subject's static balance ability based on a static stability parameter includes: determining the numerical range to which the static stability parameter belongs; and determining the subject's static balance ability based on the static balance level corresponding to the numerical range.

[0011] In an exemplary embodiment, analyzing a subject's dynamic balance ability based on point cloud information includes: obtaining the point cloud centroid corresponding to the subject based on the point cloud information; extracting the subject's dynamic adaptive parameters under a preset dynamic posture based on the point cloud centroid; and analyzing the subject's dynamic balance ability based on the dynamic adaptive parameters.

[0012] In an exemplary embodiment, point cloud generation of the subject is performed based on the dynamic echo signal to obtain point cloud information, including: extracting radial distance, radial velocity, point azimuth angle and point pitch angle from the dynamic echo signal to generate initial point cloud information corresponding to the subject; and converting the initial point cloud information from the polar coordinate system to the rectangular coordinate system to obtain point cloud information.

[0013] In an exemplary embodiment, analyzing a subject's dynamic balance ability based on dynamic adaptive parameters includes: inputting dynamic adaptive parameters into a pre-trained model, obtaining the dynamic balance level output by the pre-trained model, and determining the subject's dynamic balance ability.

[0014] In one exemplary embodiment, when the human body posture is dynamic, the dynamic echo signal of the subject under the preset dynamic posture is collected by millimeter-wave radar, including: prompting the subject with the preset dynamic posture when the human body posture is dynamic; and collecting the dynamic echo signal of the subject under the preset dynamic posture by millimeter-wave radar when the subject performs the preset dynamic posture.

[0015] In one exemplary embodiment, determining the subject's human posture based on a human body image of the subject includes: extracting human joint information of the subject based on the human body image of the subject; determining the joint displacement of the subject based on the joint information; and determining the subject's human posture based on the joint displacement.

[0016] In one exemplary embodiment, after analyzing the subject's balance ability based on static and dynamic balance abilities, the method further includes: triggering a risk warning corresponding to the risk conditions when the subject's balance ability meets the risk conditions.

[0017] In one exemplary embodiment, the risk condition includes: in human balance ability, the static balance ability measured in a single measurement does not meet a first static balance index, and the dynamic balance ability does not meet a first dynamic balance index.

[0018] In one exemplary embodiment, the risk condition includes: in human balance ability, at least two measurements of static balance ability consecutively fail to meet a second static balance index.

[0019] In one exemplary embodiment, the risk condition includes: in human balance ability, at least one measurement of static balance ability undergoes a sudden change, and the magnitude of the change does not conform to a third static balance index.

[0020] In one exemplary embodiment, the risk condition includes: in human balance ability, the slope of the static balance ability curve deviates from a fourth static balance index.

[0021] In one exemplary embodiment, the risk condition includes: in human balance ability, dynamic balance ability measured at least twice consecutively fails to meet a second dynamic balance index.

[0022] In one exemplary embodiment, the risk condition includes: in human balance ability, at least one measurement of dynamic balance ability undergoes a sudden change, and the magnitude of the change does not conform to a third dynamic balance index.

[0023] In one exemplary embodiment, the risk condition includes: in human balance ability, the slope of the dynamic balance ability curve deviates from a fourth dynamic balance index.

[0024] According to a second aspect of this disclosure, a human balance ability detection device is provided, comprising: a visual detection module and a balance detection module, the balance detection module including a static balance detection submodule, a dynamic balance detection submodule, and a comprehensive balance detection submodule; wherein, the visual detection module is used to determine the human posture of the subject based on a human image of the subject; the static balance detection submodule is used to acquire the static echo signal of the subject using millimeter-wave radar when the human posture is static; and analyze the static balance ability of the subject based on the static echo signal; the dynamic balance detection submodule is used to acquire the dynamic echo signal of the subject under a preset dynamic posture using millimeter-wave radar when the human posture is dynamic, and generate a point cloud of the subject based on the dynamic echo signal to obtain point cloud information; and analyze the dynamic balance ability of the subject based on the point cloud information; the comprehensive balance detection submodule is used to analyze the corresponding human balance ability of the subject based on the static balance ability and the dynamic balance ability.

[0025] In an exemplary embodiment, the integrated balance detection submodule is specifically used to filter out periodic micro-motion signals from the static echo signal and extract phase information from consecutive frames to obtain the human micro-motion signal corresponding to the static echo signal; extract the static stability parameters of the subject when the human posture is static based on the human micro-motion signal; and analyze the subject's static balance ability based on the static stability parameters.

[0026] In one exemplary embodiment, the static balance detection submodule is specifically used to point the millimeter-wave radar at the corresponding human body position of the subject based on the human body joint information corresponding to the subject in the human body image, and obtain a local static echo signal.

[0027] In one exemplary embodiment, the static balance detection submodule is specifically used to determine the numerical range to which the static stability parameter belongs; and to determine the subject's static balance ability based on the static balance level corresponding to the numerical range.

[0028] In an exemplary embodiment, the dynamic balance detection submodule is specifically used to obtain the centroid of the point cloud corresponding to the subject based on point cloud information; extract the dynamic adaptability parameters of the subject under a preset dynamic posture based on the centroid of the point cloud; and analyze the subject's dynamic balance ability based on the dynamic adaptability parameters.

[0029] In an exemplary embodiment, the dynamic balance detection submodule is specifically used to extract radial distance, radial velocity, point azimuth angle and point pitch angle from the dynamic echo signal to generate initial point cloud information corresponding to the subject; and to convert the initial point cloud information from the polar coordinate system to the rectangular coordinate system to obtain the point cloud information.

[0030] In one exemplary embodiment, the dynamic balance detection submodule is specifically used to input dynamic adaptability parameters into a pre-trained model, obtain the dynamic balance level output by the pre-trained model, and determine the subject's dynamic balance ability.

[0031] In one exemplary embodiment, the dynamic balance detection submodule is specifically used to prompt the subject with a preset dynamic posture when the human body posture is dynamic; and to collect the dynamic echo signal of the subject in the preset dynamic posture by millimeter-wave radar when the subject performs the preset dynamic posture.

[0032] In one exemplary embodiment, the visual detection module is used to extract human joint information of the subject based on a human body image of the subject; determine the joint displacement of the subject based on the human joint information; and determine the human posture of the subject based on the joint displacement.

[0033] In one exemplary embodiment, after analyzing the subject's balance ability based on static and dynamic balance abilities, the method further includes: triggering a risk warning corresponding to the risk conditions when the subject's balance ability meets the risk conditions.

[0034] In one exemplary embodiment, the risk condition includes: in human balance ability, the static balance ability measured in a single measurement does not meet a first static balance index, and the dynamic balance ability does not meet a first dynamic balance index.

[0035] In one exemplary embodiment, the risk condition includes: in human balance ability, at least two measurements of static balance ability consecutively fail to meet a second static balance index.

[0036] In one exemplary embodiment, the risk condition includes: in human balance ability, at least one measurement of static balance ability undergoes a sudden change, and the magnitude of the change does not conform to a third static balance index.

[0037] In one exemplary embodiment, the risk condition includes: in human balance ability, the slope of the static balance ability curve deviates from a fourth static balance index.

[0038] In one exemplary embodiment, the risk condition includes: in human balance ability, dynamic balance ability measured at least twice consecutively fails to meet a second dynamic balance index.

[0039] In one exemplary embodiment, the risk condition includes: in human balance ability, at least one measurement of dynamic balance ability undergoes a sudden change, and the magnitude of the change does not conform to a third dynamic balance index.

[0040] In one exemplary embodiment, the risk condition includes: in human balance ability, the slope of the dynamic balance ability curve deviates from a fourth dynamic balance index.

[0041] According to a third aspect of this disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to implement the above-described method by executing the executable instructions.

[0042] According to a fourth aspect of this disclosure, a computer-readable storage medium is provided that stores a computer program thereon, which, when executed by a processor, implements the method described above.

[0043] This disclosure provides a method, device, electronic device, and computer-readable storage medium for detecting human balance ability. The scheme determines human posture through visual detection of the subject, then collects static echo signals from the subject using millimeter-wave radar when the posture is static, and analyzes the subject's static balance ability based on these signals. Conversely, when the posture is dynamic, it collects dynamic echo signals from the subject under a preset dynamic posture using millimeter-wave radar, generates point clouds based on these signals to obtain point cloud information, and analyzes the subject's dynamic balance ability based on this information. Finally, it comprehensively analyzes the subject's balance ability based on both static and dynamic balance capabilities. This non-contact detection method allows for flexible deployment, convenient detection, and sustainable processing. Based on visual detection, it distinguishes between static and dynamic balance analysis scenarios using millimeter-wave radar, enabling comprehensive and thorough balance detection of the subject. It eliminates reliance on human experience, standardizes the detection results, and integrates static and dynamic balance analysis results, making the results more multidimensional and improving detection accuracy.

[0044] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0045] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0046] Figure 1 This schematically illustrates one of the steps of a human balance ability detection method in an exemplary embodiment of the present disclosure.

[0047] Figure 2 This schematically illustrates a second flowchart of the steps of a human balance ability detection method in an exemplary embodiment of this disclosure.

[0048] Figure 3The schematic diagram illustrates the structure of a human balance ability detection device in an exemplary embodiment of the present disclosure.

[0049] Figure 4 A schematic diagram of the composition of an electronic device to which the exemplary embodiments of the present disclosure may be applied is shown. Detailed Implementation

[0050] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0051] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0052] It should be noted that all data and information obtained in this public disclosure are accessed, collected, stored, and used for subsequent analysis and processing after the user or relevant data owner has been clearly informed of the content of the data collection, the purpose of the data, the processing method, etc., and with the consent and authorization of the user or relevant data owner. Furthermore, the public may send the user or relevant data owner access, correction, or deletion of the data, as well as the method of revoking consent or authorization.

[0053] Figure 1 One of the flowcharts of a method for detecting human balance ability according to an exemplary embodiment of the present disclosure is shown. Figure 1 The method may include steps 101 to 106 as follows.

[0054] In step 101, the subject's human posture is determined based on the subject's human body image.

[0055] In this embodiment, the subject may include at least one user active in the detection scenario. This can be any user entering the detection scenario who directly becomes a subject, or a user who has pre-registered their identity information upon entering the detection scenario. In this case, the subject's identity can be identified using the pre-registered identity information. Human images can be captured by cameras, depth cameras, monitoring equipment, etc., deployed in the detection scenario, or by mobile devices equipped with camera modules. Those skilled in the art can choose according to actual deployment conditions and measurement needs; this embodiment does not impose specific limitations in this regard.

[0056] Human body images can be used to determine the subject's posture. Human posture can be categorized as static or dynamic, and these two concepts are relative. Continuous-time human body images can reflect the maintenance and changes in the subject's posture. Based on this, static and dynamic postures can be distinguished according to measurement accuracy and requirements, thus corresponding to different millimeter-wave radar testing scenarios.

[0057] In step 102, with the human body in a static posture, the static echo signal of the subject is acquired by millimeter-wave radar.

[0058] In this embodiment of the disclosure, the human posture is static, which can be a human image reflecting changes in the subject's human posture, or the maintenance of the human posture conforming to the definition of static. In this case, the static echo signal generated by the subject in a static human posture can be acquired by the millimeter-wave radar.

[0059] Millimeter-wave radar can be deployed in a detection scenario, with at least one millimeter-wave radar deployed at one or more different locations within the scenario. The pointing of the millimeter-wave radar can be fixed or adjustable. Static echo signals of the subject can be acquired by activating millimeter-wave radars with different pointing directions or by adjusting the pointing of an adjustable millimeter-wave radar. Adjusting the pointing of the millimeter-wave radar can be achieved through a mechanical structure or based on beamforming technology. For example, depending on the location of the subject, a millimeter-wave radar pointing to the corresponding location can be activated, or the pointing of the millimeter-wave radar can be adjusted to a corresponding location. This disclosure does not impose specific limitations on these embodiments.

[0060] Millimeter-wave radar is a type of radar that operates at radio frequencies in the millimeter-wave band (30-300 GHz) with wavelengths of 1-10 mm, such as 24 GHz, 60 GHz, 77 GHz, and 80 GHz. Millimeter-wave radar has shorter wavelengths, higher resolution and accuracy, and a larger Doppler frequency shift, making it more effective at detecting low-speed targets. Its wider frequency band allows it to use narrow pulse or wideband frequency-modulated signals to distinguish fine features, and it also offers higher range resolution. Therefore, in the detection of human balance ability, millimeter-wave radar, combined with visual detection of human images, can more accurately leverage its performance characteristics, improving measurement resolution and accuracy.

[0061] When the human body is in a static posture, the millimeter-wave radar is activated to detect the static echo signal of the subject. Depending on the type of millimeter-wave radar, it can transmit a frequency of f to the subject. c The frequency-modulated continuous wave is used to generate a high-frequency millimeter-wave signal through linear frequency modulation, thereby acquiring the static echo signal reflected by the subject in a static posture. Those skilled in the art can choose other types of millimeter-wave radar according to actual needs, such as pulse or other continuous wave types, and the embodiments disclosed herein do not impose specific limitations in this regard.

[0062] In step 103, the subject's static balance ability is analyzed based on the static echo signal.

[0063] In this embodiment, the static echo signal contains static balance characteristics of the subject when the human body is in a static posture, such as distance, velocity, and micro-movements. Based on the static echo signal, these static balance characteristics can be extracted and standardized to determine the subject's static balance ability.

[0064] In step 104, when the human body is in a dynamic posture, the dynamic echo signal of the subject under the preset dynamic posture is collected by millimeter-wave radar, and point cloud is generated based on the dynamic echo signal to obtain point cloud information.

[0065] In this embodiment, the human posture is dynamic, which may be a change in the subject's posture reflected in a human image, or the maintenance of the human posture may not conform to the definition of dynamic. In this case, the millimeter-wave radar can be used to collect the dynamic echo signal generated by the subject under a preset dynamic posture. The millimeter-wave radar can be referred to the relevant description in step 102 above, and will not be repeated here to avoid repetition.

[0066] Preset dynamic postures can be dynamic postures that the subject is currently performing that meet the testing requirements, such as walking, turning, squatting, etc.; or dynamic postures that the subject performs with prompting, such as walking, turning, squatting, etc. The selection and design of preset dynamic postures can be based on the subject's gender, age, height, body type, medical history, and other personal information, as well as on the difficulty of the movement. Subjects can also be provided with a choice of individual or combined preset dynamic postures.

[0067] During the process of the subject performing a preset posture, the millimeter-wave radar can sample the dynamic echo signal reflected by the human body through the ADC (Analog-to-Digital Converter). The dynamic echo signal includes the dynamic balance characteristics of the subject under the preset dynamic posture, such as distance, speed, and angle. Based on this, point cloud can be generated for the subject to construct point cloud information.

[0068] Point cloud information is a collection of discrete points, which may include point coordinates and other attribute information, such as the reflection intensity of dynamic echo signals. By extracting and analyzing the dynamic equilibrium features of the dynamic echo signals, the point coordinates on the subject's surface can be determined.

[0069] In step 105, the subject's dynamic balance ability is analyzed based on point cloud information.

[0070] In this embodiment, the point cloud information includes information on the changes in point coordinates when the subject performs a preset dynamic posture while the human body is in a dynamic state. Based on the point cloud information, the point coordinates can be aggregated and classified, and dynamic balance features such as distance, speed, angle, and time can be extracted and standardized for analysis to determine the subject's dynamic balance ability.

[0071] In step 106, the subject's balance ability is analyzed based on static balance ability and dynamic balance ability.

[0072] In this embodiment, the subject's balance ability can be comprehensively analyzed based on static balance ability and dynamic balance ability. By combining the static stability maintained by the subject in a static human posture and the dynamic adaptability in a dynamic human posture, the balance ability of the subject's core and lower limbs can be evaluated. The non-contact detection effectively improves the convenience and accuracy of human balance detection.

[0073] Figure 2 A second flowchart illustrating the steps of a human balance ability detection method according to an exemplary embodiment of this disclosure is shown. Figure 2 The method may include steps 201 to 210.

[0074] In step 201, the subject's human posture is determined based on the subject's human body image.

[0075] In this embodiment of the disclosure, step 201 can be referred to the relevant description of step 101 above. To avoid repetition, it will not be repeated here.

[0076] In an exemplary embodiment of this disclosure, step 201 may include steps A1 to A3.

[0077] In step A1, the subject's human joint information is extracted based on the subject's human body image.

[0078] In step A2, the joint displacement of the subject is determined based on the human joint information.

[0079] In this embodiment of the disclosure, based on the acquisition of human body images of the subject, the subject's joint information, such as the position information of the subject's head (H), shoulders (S), hips (Hip), and legs (L), can be extracted from the human body images. Based on this, the displacement of the subject's joints on consecutive human body images can be calculated, such as the displacement H of the subject's head. shift Shoulder displacement S ship Hip displacement ship Leg displacement L shift The displacement of the joint points can be obtained through pixel statistics or other distance calculation methods, and this disclosure does not impose specific limitations on this.

[0080] In step A3, the subject's posture is determined based on the joint displacement.

[0081] In this embodiment, human posture can be divided into relatively static and dynamic, as described in step 101 above. To avoid repetition, this will not be repeated here. Joint displacement reflects the maintenance and changes in the subject's human posture over continuous time. A large joint displacement indicates insufficient maintenance and significant changes in the subject's posture over continuous time; conversely, a small joint displacement indicates stable posture and minimal changes over continuous time. When determining the subject's human posture, the magnitude of the joint displacement can be used to determine the corresponding posture. For example, if the joint displacement is less than a static threshold, the subject's posture is determined to be static; if the joint displacement is greater than or equal to the static threshold, the subject's posture is determined to be dynamic. The above scheme is merely illustrative; those skilled in the art can choose appropriate evaluation methods according to actual needs. This embodiment does not impose specific limitations in this regard.

[0082] For example, taking hip displacement as an example, when it conforms to the following formula (1), the subject's corresponding human posture is static.

[0083] Hip shift =|P hip1 -P hip2 |<∈(pixels) (1)

[0084] Among them, P hip1 P represents the position information of the hip joint at time t1. hip2 This represents the hip joint position information at time t2, and ∈(pixels) represents the static threshold.

[0085] In step 202, when the human body is in a static posture, the static echo signal of the subject is acquired by millimeter-wave radar.

[0086] In this embodiment of the disclosure, step 202 can be referred to the relevant description of step 102 above. To avoid repetition, it will not be repeated here.

[0087] In an exemplary embodiment of this disclosure, the static echo signal includes a global static echo signal and a local static echo signal. The acquisition of the subject's static echo signal by millimeter-wave radar in step 202 includes the following steps B1 and B2.

[0088] In step B1, the global static echo signal collected by the millimeter-wave radar from the subject in the forward-looking direction is acquired.

[0089] In step B2, based on the human joint information corresponding to the subject in the human image, the millimeter-wave radar is directed to the corresponding human body position of the subject to obtain a local static echo signal.

[0090] In this embodiment of the disclosure, when the human body is in a static posture, the static echo signal may further include global static echo signals reflected from the entire body of the subject, as well as local static echo signals from specific areas of the subject's body. This allows for a combined global and local analysis of the subject, avoiding the loss of details and the neglect of the overall picture. It also helps in locating the source when predicting or experiencing imbalance problems, making the solution scalable. Specifically, pointing the millimeter-wave radar at the corresponding human body position can be achieved by activating the millimeter-wave radar pointing to the corresponding human body position from among multiple millimeter-wave radars pointing to different positions, or by adjusting the millimeter-wave radar's direction of the millimeter-wave radar to the corresponding human body position through mechanical structures or beamforming technology.

[0091] In this embodiment of the disclosure, the global static echo signal can be collected from the subject at a frontal viewing angle, i.e. without a pitch angle; the local static echo signal can, for example, include the static echo signal of the subject's upper body trunk and the static echo signal of the lower body legs. Those skilled in the art can choose according to actual needs, and this embodiment of the disclosure does not impose specific limitations on this.

[0092] In step 203, periodic micro-motion signals are filtered out from the static echo signal, and phase information of consecutive frames is extracted to obtain the human body micro-motion signals corresponding to the static echo signal.

[0093] In this embodiment, the static echo signal includes the subject's static balance characteristics, which can be analyzed by extracting phase information from consecutive frames. This can be achieved by performing a Fast Fourier Transform (FFT) on the static echo signal to convert the time-domain signal into a frequency-domain signal, and obtaining the human body micro-motion signal based on the conversion relationship between path difference and phase difference. When the static echo signal includes both global and local static echo signals, the global micro-motion signal can be obtained from the global static echo signal, and the local micro-motion signal corresponding to the human body location can be obtained from the local static echo signal.

[0094] In this embodiment of the disclosure, to avoid interference from periodic fluctuation signals caused by breathing or other factors unrelated to body balance, periodic micro-motion signals can also be filtered out, such as by using independent component analysis, low-pass filtering, etc., to suppress respiratory harmonics and retain non-periodic micro-motion signals related to body balance. For example, as shown in the following formulas (2) and (3).

[0095]

[0096]

[0097] Formula (2) calculates the transmission frequency of the frequency-modulated continuous wave as f. c The wavelength λ at time t is converted into phase information by formula (3). Here, Δd is the micro-displacement, which includes periodic micro-displacement signals and non-periodic micro-displacement signals related to equilibrium. Periodic micro-displacement signals can be filtered out by referring to the above disclosure.

[0098] In step 204, static stability parameters of the subject when the human body posture is static are extracted based on human micro-motion signals.

[0099] In this embodiment of the disclosure, static stability parameters of the subject when the human body posture is static can be extracted from the human body micro-motion signal based on peak and trough detection. For example, the maximum value Max can be extracted. enve Minimum value Minenve The range of the maximum and minimum values, Max enve -Min enve It can reflect the amplitude of human body swaying, i.e., static micromomentum; Sample entropy (Samp) enve It can reflect the regularity of the human body; the number of peaks N peak Number of troughs N vally It can reflect the frequency of human body swaying. The static stability parameters in the examples above can reflect the subject's ability to maintain body balance in a static state, so as to facilitate subsequent static balance ability analysis. Those skilled in the art can select or expand other static stability parameters according to actual measurement conditions and needs, and the embodiments disclosed herein do not impose specific limitations in this regard.

[0100] In step 205, the subject's static balance ability is analyzed based on static stability parameters.

[0101] In this embodiment of the disclosure, a standardized assessment can be performed based on static stability parameters. For example, a standardized feature library of static stability parameters for different populations can be established, and the population type of the subject can be determined by parameter matching, thereby determining the subject's static balance ability. Alternatively, the static stability parameters can be used to analyze the changing trend of the subject's static maintenance ability, and in the long-term continuous static detection, the potential risk of imbalance can be predicted and detected in a timely manner.

[0102] In an exemplary embodiment of this disclosure, step 205 includes steps C1 to C2.

[0103] In step C1, the numerical range to which the static stability parameter belongs is determined.

[0104] In step C2, the subject's static balance ability is determined based on the static balance level corresponding to the numerical range.

[0105] In this embodiment, a standardized static balance grade can be determined for each subject based on their personal information, such as gender, age, height, body type, and medical history. In the static balance grade, each static balance level corresponds to at least one numerical range of a static stability parameter. When a static stability parameter falls within that range, it indicates that the subject's performance on that static stability parameter conforms to the corresponding static balance grade. When there are more than one static stability parameter, the static balance grades corresponding to different static stability parameters may be the same or different. When different static stability parameters conform to different static balance grades, the highest static balance grade, the lowest static balance grade, or an average or weighted average calculation can be performed to determine the subject's static balance ability.

[0106] For example, static stability parameters may include range, sample entropy, and oscillation frequency, with values ​​ranging from low to high corresponding to their respective low, moderate, and high value ranges. Static equilibrium levels can be divided into four grades: excellent, good, average, and poor.

[0107] When the range, sample entropy, and oscillation frequency are all in the low range, the subject's static balance level can be considered excellent.

[0108] When any one of the range, sample entropy, and oscillation frequency falls within the corresponding moderate range, the subject's static balance level can be considered good.

[0109] When any two of the range, sample entropy, and oscillation frequency fall within the corresponding moderate value range, the subject's static balance level can be considered average.

[0110] When any one of the range, sample entropy, and oscillation frequency falls within the corresponding high value range, the subject's static balance level can be relatively poor.

[0111] In this embodiment of the disclosure, the division of the numerical ranges and the determination of the static balance level are merely illustrative. Those skilled in the art can adjust the number and length of the numerical ranges, as well as the correspondence and judgment logic of the static balance levels, according to actual needs. For example, the numerical ranges can also be divided into low numerical ranges, moderate numerical ranges, high numerical ranges, and extremely high numerical ranges. The judgment logic can also be that all static stability parameters are excellent when they are in the low numerical range, good when they are in the moderate numerical range, average when they are in the high numerical range, and poor when they are in the extremely high numerical range. This embodiment of the disclosure does not impose specific limitations in this regard.

[0112] In step 206, when the human body is in a dynamic posture, the dynamic echo signal of the subject under the preset dynamic posture is collected by millimeter-wave radar, and point cloud is generated based on the dynamic echo signal to obtain point cloud information.

[0113] In this embodiment of the disclosure, step 206 can be referred to the relevant description of step 104 above. To avoid repetition, it will not be repeated here.

[0114] In an exemplary embodiment of this disclosure, in step 206 above, when the human body posture is dynamic, the dynamic echo signal of the subject under the preset dynamic posture is collected by millimeter-wave radar, including the following steps D1 to D2.

[0115] In step D1, when the human body is in a dynamic posture, the subject is prompted with a preset dynamic posture.

[0116] In step D2, while the subject performs a preset dynamic posture, the dynamic echo signal of the subject under the preset dynamic posture is acquired by millimeter-wave radar.

[0117] In this embodiment, when the human body is in a dynamic posture, the measurement can be proactively initiated to the subject, prompting them with a preset dynamic posture. This can be achieved through voice prompts, display screens, or other means to explain and demonstrate the preset dynamic posture, guiding the subject to complete the corresponding posture. The preset dynamic postures can cover weight transfer, such as standing up and squatting, as well as support transfer, such as leg raising. Prior to this, voice prompts, such as "Please begin preparing for the test," or interactive display screens, such as providing a start button for dynamic balance testing, can guide the subject into the testing state.

[0118] During the process of the subject performing a preset dynamic posture based on the prompts, the dynamic echo signal of the subject can be collected by millimeter-wave radar. The relevant description of step 104 above can be referred to, and will not be repeated here to avoid repetition.

[0119] In an exemplary embodiment of this disclosure, step 206 above, generating point cloud information for the subject based on the dynamic echo signal, may include the following steps E1 to E2.

[0120] Step E1: Extract radial distance, radial velocity, point azimuth and point pitch angle from the dynamic echo signal to generate the initial point cloud information corresponding to the subject.

[0121] In this embodiment, based on the obtained dynamic echo signal, a human body point cloud of the subject can be generated and coordinate transformed to obtain point cloud information in a Cartesian coordinate system in three-dimensional space. After sampling the dynamic echo signal of the subject through the ADC of the millimeter-wave radar, the radial distance is first obtained through distance-dimensional FFT, and then the radial velocity is analyzed through Doppler-dimensional FFT. On this basis, the constant false alarm rate (CFAR) algorithm is used to filter out background noise and detect effective peak points. Furthermore, the azimuth angle of each point is calculated through angle-dimensional FFT. and pitch angle θ i This generates initial point cloud information (R). i V i A i ), where R i V is the radial distance. i Let A be the radial velocity. i Depend on Confirmed. At this point, the initial point cloud information is represented in polar coordinates.

[0122] Step E2: Transform the initial point cloud information from the polar coordinate system to the rectangular coordinate system to obtain the point cloud information.

[0123] In this embodiment of the disclosure, the initial point cloud information can be transformed from polar coordinates to a rectangular coordinate system in three-dimensional space. The transformation formula can be shown in formulas (4) to (6) below, which correspond to the transformation relationships of the X-axis, Y-axis and Z-axis respectively.

[0124]

[0125] z i =R i cosθ i (6)

[0126] Therefore, the point cloud information cp in the Cartesian coordinate system can be obtained. i =[x i ,y i ,z i V i ,I i ], where I i Indicates the intensity of reflection.

[0127] In step 207, the centroid of the point cloud corresponding to the subject is obtained based on the point cloud information.

[0128] In this embodiment, the centroid of a point cloud refers to the geometric center of all points in a point cloud cluster after clustering the point cloud information. Based on the obtained point cloud information, the centroid of the point cloud can be obtained by performing multi-frame clustering on the point cloud information and removing noise. The centroid of the point cloud, Cen, can be calculated as shown in the following formula (7):

[0129]

[0130] Where N is the number of point clouds.

[0131] In step 208, dynamic adaptive parameters of the subject under a preset dynamic posture are extracted based on the centroid of the point cloud.

[0132] In this embodiment, the centroid of the point cloud can characterize the motion state of the subject. Dynamic adaptive parameters such as centroid sway amplitude, centroid oscillation symmetry, displacement, movement speed, and action completion time can be extracted from the centroid of the point cloud to characterize the subject's adaptability in maintaining stability under a preset dynamic posture. Among them, the centroid sway amplitude Amp can be calculated by the following formula (8):

[0133] Amp = max(Cen) ts -Cen init ),Cen ts ={Cen j |j=1,2,…,m} (8)

[0134] In the aforementioned formula (8), Cen ts To preset the position of the cloud centroid at each time point during the dynamic pose process, Cen init The position of the cloud centroid before the preset dynamic pose begins.

[0135] The symmetry of the center of mass swing indicates the symmetry of the left-right swing of the subject's center of mass during the execution of a pre-set dynamic posture. The symmetry of the center of mass swing (Sym) can be represented by comparing the number of deviations to one side, as calculated by the following formula (9):

[0136]

[0137] In the aforementioned formula (9), Count(Cen) ts -Cen init >0) is the count of the side that deviates from the initial position of the point cloud centroid.

[0138] Movement speed v com The ratio of the point cloud centroid movement distance ΔCen to the movement time Δt can be calculated as shown in the following formula (10):

[0139]

[0140] Where, x cen (t) and x cen (t+Δt) represents the x-coordinate of the centroid of the point cloud at the start of the preset dynamic pose and Δt time, respectively; the y-coordinate and z-coordinate are deduced similarly.

[0141] In step 209, the subject's dynamic balance ability is analyzed based on dynamic adaptive parameters.

[0142] In this embodiment of the disclosure, standardized evaluation can be performed based on dynamic adaptive parameters. Standardized grading or scoring can be performed in the time and spatial domains. Quantitative results of the point cloud centroid can also be provided to characterize the subject's dynamic balance ability. The quantitative results may include parameters representing the quality of action completion, such as the amplitude of centroid sway and the symmetry of centroid oscillation, or parameters representing the efficiency of action completion, such as movement speed and completion time.

[0143] In an exemplary embodiment of this disclosure, step 209 may include step F1.

[0144] In step F1, the dynamic adaptability parameters are input into the pre-trained model to obtain the dynamic balance level output by the pre-trained model, thereby determining the subject's dynamic balance ability.

[0145] In this embodiment, the aforementioned dynamic adaptive parameters can be input into a pre-trained model. The pre-trained model can be a classification or scoring model built using basic models such as SVM or random forest classifiers. The pre-trained model can be trained based on the dynamic balance level of a population and its corresponding statistical dynamic adaptive parameters. It can be a single pre-trained model or different pre-trained models trained for different populations. During dynamic balance level prediction, the corresponding pre-trained model is retrieved based on the subject's identity information. In the prediction process, one pre-trained model or multiple identical or different pre-trained models can be used for prediction. The dynamic balance levels output by one or more pre-trained models can be selected or fused to represent the subject's dynamic balance ability based on the dynamic balance level.

[0146] In step 210, the subject's balance ability is analyzed based on static balance ability and dynamic balance ability.

[0147] In this embodiment of the disclosure, step 210 can be referred to the relevant description of step 106 above. To avoid repetition, it will not be repeated here.

[0148] In an exemplary embodiment of this disclosure, after step 210, the following step J is also included.

[0149] In step J, when the human body's balance ability meets the risk conditions, a risk warning corresponding to the risk conditions is triggered.

[0150] In this embodiment of the disclosure, based on the detection and assessment of human balance ability, further risk alerts and interventions can be provided to form a detection, tracking, and intervention process. Risk conditions can be fall threshold conditions constructed based on human balance ability, or fall threshold conditions for core and lower limb strength decline, etc., thereby enabling timely alerts for the detection and intervention of similar risks. Specifically, based on human balance ability detection, risk alerts can include single fall risk, training suggestions for physical decline, and medical examination and treatment suggestions for disease risks, etc., providing corresponding risk alerts according to the triggered risk conditions and the detection results of human balance ability.

[0151] In an exemplary embodiment of this disclosure, the risk conditions include: in human balance ability, the static balance ability measured in a single measurement does not meet the first static balance index, and the dynamic balance ability does not meet the first dynamic balance index.

[0152] In this embodiment, the first static balance index can be set according to the representation of static balance ability. For example, if static balance ability is divided into three levels from high to low, the first static balance index can be a static balance ability of level one or higher than level two; or if it is divided into excellent, good, average, and poor, the first static balance index can be a static balance ability of excellent or higher than good. The first dynamic balance index can be determined similarly. If the static balance ability in a single measurement does not meet the first static balance index, and the dynamic balance ability does not meet the first dynamic balance index, it indicates that the subject's static and dynamic balance abilities both exceed the fall threshold, thereby triggering the fall risk warning corresponding to the risk condition.

[0153] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, at least two consecutive measurements of static balance ability fail to meet a second static balance index.

[0154] In this embodiment of the disclosure, the second static balance index can represent the imbalance baseline when the human body is in a static posture. Static balance capability is defined as failing to meet the second static balance index for at least two consecutive measurements. For example, this could be manifested as the static micromomentum being higher than the imbalance baseline for at least two consecutive measurements, or other static balance characteristics. This embodiment of the disclosure does not impose specific limitations on this.

[0155] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, at least one measurement of static balance ability undergoes a sudden change, and the magnitude of the change does not conform to a third static balance index.

[0156] In this embodiment of the disclosure, the third static balance index can represent the allowable range of change in human posture when it is static. A sudden change in static balance capability can be caused by a large change in at least one static balance characteristic in adjacent measurements. When the magnitude of the change does not conform to the third static balance index, it can be considered that the change exceeds the safety boundary, thereby triggering a risk warning; when the magnitude of the change conforms to the third static balance index, it can be considered that the change is within the safety boundary, thereby not triggering a risk warning.

[0157] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, the slope of the static balance ability curve deviates from a fourth static balance index.

[0158] In this embodiment, the static balance capability curve can be a trend curve plotted based on long-term static balance capability test results, and its slope characterizes the long-term fluctuations in static balance capability. The fourth static balance index can represent the allowable fluctuation range of the slope of the static balance capability curve. When the slope of the static balance capability curve deviates from the fourth static balance index, it can be considered that the long-term static balance capability is at risk of degradation, thereby triggering a risk warning. The length of time referred to as "long-term" can be set according to actual needs, such as ten days, fifteen days, one month, one quarter, half a year, one year, etc., and this embodiment does not impose specific limitations on this.

[0159] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, at least two measurements of dynamic balance ability consecutively fail to meet a second dynamic balance index.

[0160] In this embodiment of the disclosure, the second dynamic balance index can represent the imbalance baseline when the human body is in a dynamic posture. Dynamic balance ability is defined as failing to meet the second dynamic balance index for at least two consecutive measurements. For example, this could be the amplitude of the center of mass swaying being higher than the imbalance baseline for at least two consecutive measurements, or other dynamic balance characteristics. This embodiment of the disclosure does not impose specific limitations on this.

[0161] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, at least one measurement of dynamic balance ability undergoes a sudden change, and the magnitude of the change does not conform to a third dynamic balance index.

[0162] In this embodiment of the disclosure, the third dynamic balance index can represent the allowable range of change in human posture when it is dynamic. A sudden change in dynamic balance capability can be caused by a large change in at least one dynamic balance characteristic in adjacent measurements. When the magnitude of the sudden change does not conform to the third dynamic balance index, it can be considered that the sudden change exceeds the safety boundary, thereby triggering a risk warning; when the magnitude of the sudden change conforms to the third dynamic balance index, it can be considered that the sudden change is within the safety boundary, thereby not triggering a risk warning.

[0163] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, the slope of the dynamic balance ability curve deviates from a fourth dynamic balance index.

[0164] In this embodiment, the dynamic balancing capability curve can be a trend curve plotted based on long-term dynamic balancing capability test results, and its slope characterizes the long-term fluctuations in dynamic balancing capability. The fourth dynamic balancing index can represent the allowable fluctuation range of the slope of the dynamic balancing capability curve. When the slope of the dynamic balancing capability curve deviates from the fourth dynamic balancing index, it can be considered that there is a risk of degradation in long-term dynamic balancing capability, thereby triggering a risk warning. The duration of "long-term" can be set according to actual needs, such as ten days, fifteen days, one month, one quarter, half a year, one year, etc., and this embodiment does not impose specific limitations on this.

[0165] In this embodiment of the disclosure, at least one of the above-mentioned risk conditions may be used. Those skilled in the art can also extend the risk conditions according to actual measurement conditions and needs. This embodiment of the disclosure does not impose specific limitations on this.

[0166] This disclosure provides a method for detecting human balance ability. The method determines the subject's posture through visual detection, then collects static echo signals from the subject using millimeter-wave radar when the posture is static, and analyzes the subject's static balance ability based on these signals. Conversely, when the subject's posture is dynamic, the method collects dynamic echo signals from the subject under a preset dynamic posture using millimeter-wave radar, generates point clouds based on these signals to obtain point cloud information, and analyzes the subject's dynamic balance ability based on this point cloud information. Finally, the method comprehensively analyzes the subject's human balance ability based on both static and dynamic balance capabilities. This non-contact detection method allows for flexible deployment, convenient detection, and sustainable processing. Based on visual detection, it distinguishes between static and dynamic balance analysis scenarios using millimeter-wave radar, enabling comprehensive and thorough balance detection of the subject. It eliminates reliance on human experience, standardizes the detection results, and integrates static and dynamic balance analysis results, making the results multidimensional and improving detection accuracy.

[0167] It should be noted that the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of this disclosure, and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Furthermore, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0168] Further reference Figure 3As shown, an exemplary embodiment of this disclosure provides a human balance ability detection device 300. This device may include a visual detection module 301 and a balance detection module 302. The balance detection module 302 includes a static balance detection submodule 3021, a dynamic balance detection submodule 3022, and a comprehensive balance detection submodule 3023. Specifically, the visual detection module 301 is used to determine the subject's human posture based on a human image of the subject; the static balance detection submodule 3021 is used to collect static echo signals from the subject using millimeter-wave radar when the human posture is static; and analyze the subject's static balance ability based on the static echo signals; the dynamic balance detection submodule 3022 is used to collect dynamic echo signals from the subject in a preset dynamic posture using millimeter-wave radar when the human posture is dynamic, and generate point clouds from the subject based on the dynamic echo signals to obtain point cloud information; and analyze the subject's dynamic balance ability based on the point cloud information; the comprehensive balance detection submodule 3023 is used to analyze the subject's corresponding human balance ability based on static and dynamic balance abilities.

[0169] In an exemplary embodiment of this disclosure, the integrated balance detection submodule 3023 is specifically used to filter out periodic micro-motion signals from the static echo signal and extract phase information of consecutive frames to obtain human micro-motion signals corresponding to the static echo signal; extract static stability parameters of the subject when the human body posture is static based on the human micro-motion signals; and analyze the subject's static balance ability based on the static stability parameters.

[0170] In an exemplary embodiment of this disclosure, the static balance detection submodule 3021 is specifically used to point the millimeter-wave radar at the corresponding human body position of the subject based on the human body joint information corresponding to the subject in the human body image, and obtain a local static echo signal.

[0171] In an exemplary embodiment of this disclosure, the static balance detection submodule 3021 is specifically used to determine the numerical range to which the static stability parameter belongs; and to determine the subject's static balance ability based on the static balance level corresponding to the numerical range.

[0172] In an exemplary embodiment of this disclosure, the dynamic balance detection submodule 3022 is specifically used to obtain the centroid of the point cloud corresponding to the subject based on point cloud information; extract the dynamic adaptability parameters of the subject under a preset dynamic posture based on the centroid of the point cloud; and analyze the dynamic balance ability of the subject based on the dynamic adaptability parameters.

[0173] In an exemplary embodiment of this disclosure, the dynamic balance detection submodule 3022 is specifically used to extract radial distance, radial velocity, point azimuth angle and point pitch angle from the dynamic echo signal to generate initial point cloud information corresponding to the subject; and to convert the initial point cloud information from the polar coordinate system to the rectangular coordinate system to obtain the point cloud information.

[0174] In an exemplary embodiment of this disclosure, the dynamic balance detection submodule 3022 is specifically used to input dynamic adaptability parameters into a pre-trained model, obtain the dynamic balance level output by the pre-trained model, and determine the subject's dynamic balance ability.

[0175] In an exemplary embodiment of this disclosure, the dynamic balance detection submodule 3022 is specifically used to prompt the subject with a preset dynamic posture when the human body posture is dynamic; and to collect the dynamic echo signal of the subject in the preset dynamic posture by millimeter-wave radar when the subject performs the preset dynamic posture.

[0176] In an exemplary embodiment of this disclosure, the visual detection module 301 is used to extract human joint information of the subject based on the human body image of the subject; determine the joint displacement of the subject based on the human joint information; and determine the human posture of the subject based on the joint displacement.

[0177] In an exemplary embodiment of this disclosure, after analyzing the subject's balance ability based on static and dynamic balance abilities, the method further includes: triggering a risk warning corresponding to the risk conditions when the subject's balance ability meets the risk conditions.

[0178] In an exemplary embodiment of this disclosure, the risk conditions include: in human balance ability, the static balance ability measured in a single measurement does not meet the first static balance index, and the dynamic balance ability does not meet the first dynamic balance index.

[0179] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, at least two consecutive measurements of static balance ability fail to meet a second static balance index.

[0180] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, at least one measurement of static balance ability undergoes a sudden change, and the magnitude of the change does not conform to a third static balance index.

[0181] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, the slope of the static balance ability curve deviates from a fourth static balance index.

[0182] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, at least two measurements of dynamic balance ability consecutively fail to meet a second dynamic balance index.

[0183] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, at least one measurement of dynamic balance ability undergoes a sudden change, and the magnitude of the change does not conform to a third dynamic balance index.

[0184] In an exemplary embodiment of this disclosure, the risk condition includes: in human balance ability, the slope of the dynamic balance ability curve deviates from a fourth dynamic balance index.

[0185] The human balance ability detection device provided in this embodiment determines the human posture of the subject through visual detection. When the human posture is static, it collects static echo signals from the subject using millimeter-wave radar and analyzes the subject's static balance ability based on these signals. When the human posture is dynamic, it collects dynamic echo signals from the subject in a preset dynamic posture using millimeter-wave radar, generates point clouds based on these signals to obtain point cloud information, and analyzes the subject's dynamic balance ability based on this point cloud information. Finally, it comprehensively analyzes the subject's human balance ability based on both static and dynamic balance abilities. This non-contact detection method allows for flexible deployment, convenient detection, and sustainable processing. Based on visual detection, it distinguishes between static and dynamic balance analysis scenarios using millimeter-wave radar, enabling comprehensive and thorough balance detection of the subject. It eliminates reliance on human experience, standardizes the detection results, and integrates static and dynamic balance analysis results, making the results more multidimensional and improving detection accuracy.

[0186] The specific details of each module in the above-mentioned human balance ability detection device have been described in detail in the method section of the implementation. For any undisclosed details, please refer to the implementation content of the method section. That is, the explanation and beneficial effects of the human balance ability detection method in the above-mentioned embodiments are also applicable to the human balance ability detection device 300 of this disclosure, and will not be elaborated here.

[0187] Those skilled in the art will understand that various aspects of this disclosure can be implemented as a system, method, or program product. Therefore, various aspects of this disclosure can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects, collectively referred to herein as a "circuit," "module," or "system."

[0188] The following reference Figure 4 To describe an electronic device 400 according to such an embodiment of the present disclosure. Figure 4 The electronic device 400 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0189] like Figure 4 As shown, the electronic device 400 is manifested in the form of a general-purpose computing device. The components of the electronic device 400 may include, but are not limited to: at least one processing unit 410, at least one storage unit 420, a bus 430 connecting different system components (including storage unit 420 and processing unit 410), and a display unit 440.

[0190] The storage unit stores program code that can be executed by the processing unit 410, causing the processing unit 410 to perform the steps described in the "Exemplary Methods" section above, according to various exemplary embodiments of this disclosure.

[0191] Storage unit 420 may include a readable medium in the form of a volatile storage unit, such as random access memory (RAM) 421 and / or cache memory 422, and may further include a read-only memory (ROM) 423.

[0192] Storage unit 420 may also include a program / utility 424 having a set (at least one) of program modules 425, such program modules 425 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0193] Bus 430 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0194] Electronic device 400 can also communicate with one or more external devices 500 (e.g., keyboard, pointing device, Bluetooth device, etc.), one or more devices that enable a user to interact with electronic device 400, and / or any device that enables electronic device 400 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 450. Furthermore, electronic device 400 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 460. As shown, network adapter 460 communicates with other modules of electronic device 400 via bus 430. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0195] Furthermore, exemplary embodiments of this disclosure also provide a computer-readable storage medium storing a program product capable of implementing the methods described above. In some possible embodiments, various aspects of this disclosure may also be implemented as a program product including program code that, when run on a terminal device, causes the terminal device to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure.

[0196] It should be noted that the computer-readable medium disclosed herein may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0197] In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can transmit, propagate, or transfer a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wireline, optical fiber, RF, etc., or any suitable combination thereof.

[0198] Furthermore, program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0199] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

Claims

1. A method for detecting human balance ability, characterized in that, include: Based on the subject's human body image, determine the subject's human posture; When the human body is in a static posture, the static echo signal of the subject is acquired by millimeter-wave radar; The subject's static balance ability is analyzed based on the static echo signal; When the human body posture is dynamic, the dynamic echo signal of the subject under the preset dynamic posture is collected by millimeter-wave radar, and point cloud is generated for the subject based on the dynamic echo signal to obtain point cloud information. The subject's dynamic balance ability is analyzed based on the point cloud information; The subject's human balance ability is analyzed based on the static balance ability and the dynamic balance ability.

2. The method according to claim 1, characterized in that, The analysis of the subject's static balance ability based on the static echo signal includes: Periodic micro-motion signals are filtered out from the static echo signal, and phase information of consecutive frames is extracted to obtain the human micro-motion signal corresponding to the static echo signal. Based on the human body micro-motion signals, the static stability parameters of the subject when the human body posture is static are extracted; The static balance ability of the subjects was analyzed based on the static stability parameters.

3. The method according to claim 1 or 2, characterized in that, The static echo signal includes a global static echo signal and a local static echo signal. The acquisition of the subject's static echo signal via millimeter-wave radar includes: Acquire the global static echo signal collected by the millimeter-wave radar from the subject in the forward-looking direction; Based on the human joint information corresponding to the subject in the human image, the millimeter-wave radar is directed to the corresponding human body position of the subject to obtain the local static echo signal.

4. The method according to claim 2, characterized in that, The analysis of the subject's static balance ability based on the static stability parameters includes: Determine the numerical range to which the static stability parameter belongs; The subject's static balance ability is determined based on the static balance level corresponding to the numerical range.

5. The method according to claim 1, characterized in that, The analysis of the subject's dynamic balance ability based on the point cloud information includes: Based on the point cloud information, obtain the point cloud centroid corresponding to the subject; Based on the centroid of the point cloud, the dynamic adaptive parameters of the subject under the preset dynamic posture are extracted; The dynamic balance ability of the subjects was analyzed based on the dynamic adaptive parameters.

6. The method according to claim 1 or 5, characterized in that, The step of generating a point cloud from the subject based on the dynamic echo signal to obtain point cloud information includes: Radial distance, radial velocity, point azimuth and point elevation angle are extracted from the dynamic echo signal to generate the initial point cloud information corresponding to the subject. The initial point cloud information is transformed from polar coordinates to Cartesian coordinates to obtain the point cloud information.

7. The method according to claim 5, characterized in that, The analysis of the subject's dynamic balance ability based on the dynamic adaptive parameters includes: The dynamic adaptability parameters are input into the pre-trained model to obtain the dynamic balance level output by the pre-trained model, thereby determining the subject's dynamic balance ability.

8. The method according to claim 1 or 5, characterized in that, The step of acquiring dynamic echo signals of the subject in a preset dynamic posture using millimeter-wave radar when the human body posture is dynamic includes: When the human body posture is dynamic, the subject is prompted with the preset dynamic posture; While the subject performs the preset dynamic posture, the dynamic echo signal of the subject under the preset dynamic posture is acquired by millimeter-wave radar.

9. The method according to any one of claims 1, 2, and 5, characterized in that, Determining the subject's human posture based on the subject's human image includes: Based on the subject's human body image, extract the subject's human body joint information; The joint displacement of the subject is determined based on the human joint information. The subject's human posture is determined based on the displacement of the joint points.

10. The method according to any one of claims 1, 2, and 5, characterized in that, After analyzing the subject's balance ability based on the static balance ability and the dynamic balance ability, the method further includes: When the human balance ability meets the risk conditions, a risk warning corresponding to the risk conditions is triggered; wherein, the risk conditions include at least one of the following: In the aforementioned human balance ability, the static balance ability measured in a single measurement does not meet the first static balance index, and the dynamic balance ability does not meet the first dynamic balance index. In the aforementioned human balance ability, the static balance ability measured at least twice consecutively fails to meet the second static balance index; In the human body balance ability, at least one measurement of the static balance ability shows a sudden change, and the magnitude of the change does not conform to the third static balance index; In the aforementioned human balance ability, the slope of the static balance ability curve deviates from the fourth static balance index. In the aforementioned human balance ability, the dynamic balance ability measured at least twice consecutively fails to meet the second dynamic balance index; In the human body balance ability, at least one measurement of the dynamic balance ability shows a sudden change, and the magnitude of the change does not conform to the third dynamic balance index. In the aforementioned human balance ability, the slope of the dynamic balance ability curve deviates from the fourth dynamic balance index.

11. A human balance ability detection device, characterized in that, The device includes a visual detection module and a balance detection module. The balance detection module includes a static balance detection submodule, a dynamic balance detection submodule, and a comprehensive balance detection submodule. The visual detection module is used to determine the subject's human posture based on the subject's human body image; The static balance detection submodule is used to acquire the static echo signal of the subject using millimeter-wave radar when the human body posture is static; and to analyze the subject's static balance ability based on the static echo signal. The dynamic balance detection submodule is used to collect the dynamic echo signal of the subject in a preset dynamic posture using millimeter-wave radar when the human body posture is dynamic, and generate point cloud information for the subject based on the dynamic echo signal; and analyze the subject's dynamic balance ability based on the point cloud information. The comprehensive balance detection submodule is used to analyze the subject's human balance ability based on the static balance ability and the dynamic balance ability.

12. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to perform the method of any one of claims 1 to 10 by executing the executable instructions.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 10.

Images