Human balance assessment device, method and apparatus

Abstract

The present application relates to the technical field of rehabilitation medical equipment, and more particularly to a human body balance assessment device, a human body balance assessment method and device, which starts a projection device and a three-dimensional force platform in the human body balance assessment device; obtains pressure information on a standing platform through a three-dimensional force sensor, and calculates human body pressure data based on the pressure information; collects a human body contour image and a human eye image through a pose collection device, and collects a foot bottom image through a foot bottom image collection device; and performs human body balance state assessment according to at least one of the human body pressure data, the foot bottom image, the human body contour image and the human eye image, analyzes the human body balance state through multi-dimensional balance assessment, realizes comprehensive dynamic balance assessment, has higher precision, and flexibly combines mechanical parts and processing parts, and has wider applicability.

Description

Technical Field

[0001] This invention relates to the field of rehabilitation medical equipment technology, and in particular to human balance assessment equipment, human balance assessment methods and devices. Background Technology

[0002] Maintaining human balance depends on the coordinated function of the vestibular, visual, and proprioceptive systems. These sensory systems may become dysfunctional due to degenerative changes, infection, or trauma, thereby increasing the risk of falls and potentially leading to serious consequences such as hip fractures, back injuries, or even life-threatening situations. To reduce the occurrence of these situations, balance function assessments can be conducted on users in advance to identify fall risks as early as possible and intervene in a timely manner, providing scientific guidance for subsequent targeted examinations and interventions.

[0003] However, in traditional techniques, the equipment used for balance function assessment, both domestically and internationally, is large in size and fixed in position. When measuring human balance, it is less effective for dynamic balance testing, and the measurement of human balance is generally not accurate enough to provide high-precision human balance data and rehabilitation diagnosis and treatment assistance.

[0004] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this invention is to provide a human balance assessment device, a human balance assessment method, and an apparatus, aiming to solve the technical problems of poor assessment results and low accuracy when performing balance tests using existing human balance assessment devices.

[0006] To achieve the above objectives, the present invention provides a human balance assessment device, comprising: a main frame, a three-dimensional force table, a posture acquisition device, and a control unit;

[0007] The main frame is set on the ground and is used to install the three-dimensional force table and the pose acquisition device;

[0008] The three-dimensional force table includes a base and a balancing force table. Two guide rails are arranged in parallel on the base. The balancing force table is slidably connected to the guide rails. The balancing force table includes at least a light source, a three-dimensional force sensor, a motor, an AF-N actuator, a foot image acquisition device, and a standing platform.

[0009] The three-dimensional force platform adjusts the balance of the user standing on it by moving and / or adjusting the horizontal state of the standing platform on the guide rail.

[0010] The three-dimensional force table is also used to collect images of the soles of the feet and human pressure data of the user standing on the three-dimensional force table;

[0011] The pose acquisition device is used to acquire images of the user's eyes and body contours while standing on the three-dimensional force platform.

[0012] The control unit is used to assess the human body's balance state based on the human body pressure data, the foot image, the human body contour image, and the human eye image.

[0013] Optionally, the human balance assessment device further includes a projection device;

[0014] The main frame is equipped with movable mounting crossbars;

[0015] The projection device is mounted on the mounting crossbar;

[0016] The pose acquisition device is used to acquire the human eye image when the user views the target image while the projection device projects the target image.

[0017] Optionally, the human balance assessment device further includes:

[0018] A human safety protection device is suspended on the mounting crossbar to protect the user's balance when the three-dimensional force table is activated.

[0019] The present invention also provides a method for assessing human balance, the method comprising the following steps:

[0020] Activate the projection device and three-dimensional force table in the human balance assessment equipment;

[0021] Pressure information on a standing platform is acquired using a three-dimensional force sensor, and human body pressure data is calculated based on this pressure information.

[0022] Human contour images and human eye images are acquired using a pose acquisition device, and foot images are acquired using a foot image acquisition device.

[0023] The human balance status is assessed based on at least one of the human pressure data, the foot image, the human contour image, and the human eye image.

[0024] Optionally, the step of assessing human balance based on at least one of the human pressure data, the plantar image, the human contour image, and the human eye image further includes:

[0025] Acquire the data timestamps of the human body pressure data, the foot image, the human body contour image, and the human eye image;

[0026] The human pressure data, the foot image, the human contour image, and the human eye image are time-aligned based on the data timestamp;

[0027] The human balance status is assessed based on at least one of the aligned human pressure data, the foot image, the human contour image, and the human eye image.

[0028] Optionally, the step of assessing human balance based on at least one of the aligned human pressure data, the plantar image, the human contour image, and the human eye image includes:

[0029] The sole image is preprocessed to obtain a footprint image;

[0030] Obtain the pixel value of each pixel in the footprint image and the image weighted grayscale centroid;

[0031] The plantar pressure distribution data is calculated based on the pixel values ​​of each pixel and the image weighted gray-level centroid.

[0032] Calculate the arch index, plantar support area, and pressure center offset based on the plantar pressure distribution data;

[0033] The first assessment result is obtained by evaluating the human body's balance state based on the arch index, plantar support area, and center of pressure offset.

[0034] Optionally, the step of calculating the arch index, plantar support area, and pressure center offset based on the plantar pressure distribution data includes:

[0035] The footprint image is divided into the forefoot region, the middlefoot region, and the hindfoot region;

[0036] Calculate the forefoot support area corresponding to the forefoot region, the midfoot support area corresponding to the midfoot region, and the heel support area corresponding to the heel region based on the plantar pressure distribution data.

[0037] The plantar support area is determined based on the sum of the forefoot support area, midfoot support area, and heel support area.

[0038] The arch index is obtained by calculating the area ratio between the midfoot support area and the sole support area.

[0039] The pressure center offset is calculated based on the plantar pressure distribution data and historical plantar pressure distribution data.

[0040] Optionally, the step of assessing human balance based on at least one of the aligned human pressure data, the plantar image, the human contour image, and the human eye image further includes:

[0041] Identify skeletal key points in each human body contour image and label them according to the type of each skeletal key point;

[0042] One-dimensional vector concatenation is performed on each marked skeletal key point to generate multiple skeletal RGB images, where the coordinates of each skeletal key point in the RGB images correspond to RGB pixel values.

[0043] The pixels of the multiple skeletal RGB images are normalized to obtain multiple target skeletal RGB images;

[0044] The target skeleton RGB image is input into a trained neural network model for analysis to obtain a second evaluation result. The second evaluation result includes at least one of the following: skeleton point trajectory, joint angle, and trunk-limb coordination.

[0045] Optionally, the step of assessing human balance based on at least one of the aligned human pressure data, the plantar image, the human contour image, and the human eye image further includes:

[0046] Obtain the set of pressure center coordinates corresponding to the human body pressure dataset;

[0047] The pressure center is marked and preprocessed to obtain the target pressure center coordinate set;

[0048] Based on the target pressure center coordinate set, time-domain and frequency-domain measurement indicators are calculated. The time-domain measurement indicators include at least: time-domain average distance, time-domain root mean square distance, time-domain swing path, time-domain average velocity, and human body tilt angle. The frequency-domain measurement indicators include at least: power spectral density and power spectral moment.

[0049] The time-domain fluctuation and frequency-domain offset of human pressure data are determined based on the time-domain and frequency-domain measurement indicators.

[0050] The third evaluation result is obtained based on the time-domain fluctuation and frequency-domain offset.

[0051] Furthermore, to achieve the above objectives, the present invention also proposes a human balance assessment device, the human balance assessment device comprising:

[0052] The startup module is used to activate the projection device and the three-dimensional force table in the human balance assessment device.

[0053] The calculation module is used to acquire pressure information on the standing platform through a three-dimensional force sensor and calculate human body pressure data based on the pressure information.

[0054] The acquisition module is used to acquire human contour images and human eye images through a pose acquisition device, and to acquire foot images through a foot image acquisition device.

[0055] The assessment module is used to assess the human balance state based on at least one of the human pressure data, the foot image, the human contour image, and the human eye image.

[0056] Furthermore, to achieve the above objectives, the present invention also proposes a human balance assessment device, which includes: a memory, a processor, and a human balance assessment program stored in the memory and executable on the processor, wherein the human balance assessment program is configured to implement the steps of the human balance assessment method described above.

[0057] In addition, to achieve the above objectives, the present invention also proposes a storage medium storing a human balance assessment program, which, when executed by a processor, implements the steps of the human balance assessment method as described above.

[0058] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the human balance assessment method described above.

[0059] This invention activates the projection device and three-dimensional force stage in the human balance assessment device; acquires pressure information on the standing platform through a three-dimensional force sensor, and calculates human pressure data based on the pressure information; acquires human contour images and human eye images through a pose acquisition device, and acquires foot images through a foot image acquisition device; and assesses human balance status based on at least one of the human pressure data, the foot images, the human contour images, and the human eye images. Through multi-dimensional balance assessment and analysis, it achieves comprehensive dynamic balance assessment with higher accuracy. Furthermore, the flexible combination of mechanical and processing components broadens its applicability and avoids the technical problems of poor assessment results and low accuracy in existing human balance assessment devices during balance testing. Attached Figure Description

[0060] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0061] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a schematic diagram of the human balance assessment device of the present invention;

[0063] Figure 2 This is a schematic diagram of the three-dimensional force table in the human balance assessment device of the present invention;

[0064] Figure 3 This is a flowchart illustrating the first embodiment of the human balance assessment method of the present invention;

[0065] Figure 4 This is a schematic diagram of the user's form during the operation of a three-dimensional force table according to an embodiment of the human body balance assessment method of the present invention.

[0066] Figure 5 This is a schematic diagram illustrating the method of segmenting footprint images using the minimum bounding rectangle in an embodiment of the human balance assessment method of the present invention.

[0067] Figure 6 This is a schematic diagram showing the position of the camera device on the mounting crossbar in an embodiment of the human balance assessment method of the present invention;

[0068] Figure 7 This is a schematic diagram of the motion modeling stage of an embodiment of the human balance assessment method of the present invention;

[0069] Figure 8 This is a schematic diagram of the coordinate axes of the pressure center on a three-dimensional force table according to an embodiment of the human balance assessment method of the present invention;

[0070] Figure 9 This is a structural block diagram of the first embodiment of the human balance assessment device of the present invention;

[0071] Figure 10 This is a schematic diagram of the structure of the human balance assessment device in the hardware operating environment involved in the embodiments of the present invention.

[0072] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0073] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0074] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0075] This invention provides a human balance assessment device, for reference. Figure 1 and Figure 2 , Figure 1 This is a schematic diagram of the human balance assessment device in this embodiment. Figure 2 This is a schematic diagram of the structure of the three-dimensional force table in this embodiment. The human body balance assessment device includes: a main frame 1, a three-dimensional force table 5, a pose acquisition device 4, and a control unit 7.

[0076] The main frame 1 is set on the ground and is used to install the three-dimensional force table 5 and the pose acquisition device 4.

[0077] The three-dimensional force table 5 includes a base 56 and a balancing force table. Two guide rails 55 are arranged in parallel on the base. The balancing force table is slidably connected to the guide rails 55. The balancing force table includes at least a light source 51, a three-dimensional force sensor 52, a motor 53, an AF-N actuator 54, a foot image acquisition device 57, and a standing platform.

[0078] A three-dimensional force table, by moving and / or adjusting the horizontal state of the standing platform on the guide rail, adjusts the balance of the user standing on the three-dimensional force table;

[0079] The three-dimensional force table is also used to acquire images of the soles of the feet and human pressure data of the user standing on the three-dimensional force table;

[0080] The pose acquisition device is used to acquire images of the user's eyes and body contours while standing on the three-dimensional force platform.

[0081] The control unit is used to assess the human body's balance state based on the human body pressure data, the foot image, the human body contour image, and the human eye image.

[0082] It should be noted that in recent years, due to the widespread adoption of intelligent surveillance, some research has begun to explore the use of image analysis techniques (such as deep learning algorithms and pose estimation) to detect human posture and movement, especially in areas such as gait analysis and fall risk assessment. These studies utilize cameras and video analysis to monitor posture and balance. However, relying solely on image analysis may not achieve the same accuracy as traditional sensors (such as force platforms), especially when detecting minute posture adjustments and subtle shifts in the center of gravity. Image analysis is easily affected by factors such as ambient light, viewing angle, and resolution, leading to lower reliability of the measurement data. Planar image analysis calculates center of gravity information based on the outline of the patient captured in video. If the patient has a limb disability and is obscured by clothing, the error in calculating the center of gravity will be significant. Therefore, two-dimensional images lack accurate capture of depth information, making image analysis potentially less effective than force platforms or inertial sensors, thus affecting the accuracy of balance assessment.

[0083] Compared to traditional balance assessment devices, in this embodiment, the human balance assessment device includes: a main frame, a three-dimensional force table, a posture acquisition device, and a control unit. The three-dimensional force table base and the balance force table are provided with two guide rails on the base. The balance force table is slidably connected to the base through the guide rails. When the user stands on the balance force table, the balance force table can slide on the guide rails to adjust the user's balance state, thereby observing the user's limb control ability when in an unstable state and realizing the dynamic balance test of the human body.

[0084] Meanwhile, the balance force table includes at least a light source, a three-dimensional force sensor, a motor, an AF-N actuator, a plantar image acquisition device, and a standing platform. The motor and AF-N actuator can control the horizontal state of the platform surface of the balance force table to tilt, so as to adjust the user's balance state. It can also collect the pressure data of the user on the standing platform during the balance test through the three-dimensional force sensor, and combine it with the plantar image acquired by the plantar image acquisition device to realize the dynamic balance test when the user is dealing with an unstable state.

[0085] In one feasible embodiment, the human balance assessment device further includes a projection device;

[0086] The main frame is equipped with movable mounting crossbars;

[0087] The projection device is mounted on the mounting crossbar;

[0088] The pose acquisition device is used to acquire the human eye image when the user views the target image while the projection device projects the target image.

[0089] It should be noted that the projection device can be a projector, mobile terminal or other device with image mapping function. This embodiment does not impose any specific restrictions on this. In addition, in order to facilitate the installation and movement of the device, this embodiment can provide a movable mounting crossbar on the main frame to avoid drilling or other operations during the installation process. Both the projection device and the pose acquisition device can be mounted on the mounting crossbar.

[0090] The pose acquisition device can be a matrix of multiple image acquisition devices, such as a camera matrix. By acquiring user images from multiple angles, it can achieve human pose detection. At the same time, it can also acquire images of the human eye to combine with the orientation of the mapped image from the projection device for balance testing.

[0091] In one feasible embodiment, the human balance assessment device further includes:

[0092] A human safety protection device is suspended on the mounting crossbar to protect the user's balance when the three-dimensional force table is activated.

[0093] In its implementation, the human balance assessment device has a height of 2 meters and a length and width of 1 meter. Hooks are installed on the crossbar beams for suspending protective clothing (safety garments) to ensure the safety of the subjects. The connection mechanism between the crossbar and the fixed frame involves spring-loaded latches at both ends of the crossbar, with built-in spring buffer structures to reduce vibration transmission during device operation. Corresponding grooves and locking holes are designed on the frame for quick installation and removal of the crossbar from the fixed frame. Neodymium iron boron magnets (with a magnetic force ≥ 5 kg) are embedded in the contact surfaces between the crossbar and the frame to aid in quick alignment. Silicone shock-absorbing pads (damping coefficient 0.3) are installed between the crossbar and the frame to achieve vibration damping. The projector is connected to the crossbar via a sliding rail at the top, with the projector connected via a universal ball joint. The adjustment range is ±30° for pitch and 360° for horizontal rotation. Furthermore, the detachable, drill-free crossbar can be removed from the fixed frame and used with the projector and camera for independent rehabilitation training at the patient's home.

[0094] Based on this, embodiments of the present invention provide a method for assessing human balance, referring to... Figure 3 , Figure 3 This is a flowchart illustrating the first embodiment of a human balance assessment method according to the present invention.

[0095] In this embodiment, the human balance assessment method includes:

[0096] Step S10: Activate the projection device and three-dimensional force table in the human balance assessment device.

[0097] Step S20: Obtain pressure information on the standing platform using a three-dimensional force sensor, and calculate human body pressure data based on the pressure information.

[0098] Step S30: Acquire human body contour images and human eye images through a pose acquisition device, and acquire foot images through a foot image acquisition device.

[0099] Step S40: Assess the human balance state based on at least one of the human pressure data, the foot image, the human contour image, and the human eye image.

[0100] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device, control chip, or control computer capable of performing the above functions. The following description uses a control computer as an example to illustrate this embodiment and the subsequent embodiments.

[0101] refer to Figure 4 , Figure 4This is a schematic diagram of the user's form during the operation of the three-dimensional force table in this embodiment. Activating the projection device and the three-dimensional force table in the human body balance assessment device means controlling the projection device to turn on the projection to project a specific image onto a preset plane, and controlling the movement of the balance force table on the guide rail and / or controlling the tilt of the balance force table to control the tilt of the human body, thereby collecting dynamic posture and other data that affect the human body balance during the tilting process, thereby realizing the assessment of human body balance posture and providing auxiliary opinions for subsequent disease diagnosis or rehabilitation.

[0102] Understandably, the three-dimensional force sensors on the balancing platform can be evenly distributed below the standing platform, for example... Figure 2 The four corners of the body are not specifically limited in this embodiment. Human body pressure data refers to the time-domain measurement index and frequency-domain measurement index of the human body pressure center. The time-domain measurement index includes at least: time-domain average distance, time-domain root mean square distance, time-domain swing path, time-domain average velocity, and human body tilt angle. The frequency-domain measurement index includes at least: power spectral density and power spectral moment. This embodiment does not specifically limit this.

[0103] Specifically, due to the activation of the projection device and the three-dimensional force table, the healthy user is in an unbalanced state. The body tilt angle, dynamic posture control, foot pressure distribution, and overall coordination of the human limbs are all different from those of patients with otogenic vertigo. Therefore, in this embodiment, the presence of otogenic vertigo symptoms can be determined by comparing the human pressure data, the foot image, the human contour image, and the human eye image.

[0104] In specific implementation, since the data collected in this embodiment comes from different devices, it may be necessary to combine multi-dimensional data for comprehensive judgment in the subsequent judgment process. After the data collection is completed, the human pressure data, the foot image, the human contour image and the human eye image can be time-aligned to improve the credibility of the data and the reliability of the subsequent judgment.

[0105] In a feasible embodiment, the assessment of human balance based on at least one of the human pressure data, the plantar image, the human contour image, and the human eye image further includes:

[0106] Acquire the data timestamps of the human body pressure data, the foot image, the human body contour image, and the human eye image;

[0107] The human pressure data, the foot image, the human contour image, and the human eye image are time-aligned based on the data timestamp;

[0108] The human balance status is assessed based on at least one of the aligned human pressure data, the foot image, the human contour image, and the human eye image.

[0109] Specifically, due to the different tilting or moving speeds of the three-dimensional force table, it is necessary to collect a continuous and large number of image frames or pressure datasets when acquiring data such as human pressure data, foot images, human contour images, and human eye images. Pressure data or images at the same moment represent the user's body balance state at that moment. Therefore, in order to improve the reliability of the user's balance state, time alignment is performed by the timestamps of each data or image. That is, pressure data or images with the same timestamp are treated as a group of data and participate in subsequent image processing or data feature calculation.

[0110] In one feasible embodiment, the assessment of human balance based on at least one of the aligned human pressure data, the plantar image, the human contour image, and the human eye image includes:

[0111] The sole image is preprocessed to obtain a footprint image;

[0112] Obtain the pixel value of each pixel in the footprint image and the image weighted grayscale centroid;

[0113] The plantar pressure distribution data is calculated based on the pixel values ​​of each pixel and the image weighted gray-level centroid.

[0114] Calculate the arch index, plantar support area, and pressure center offset based on the plantar pressure distribution data;

[0115] The first assessment result is obtained by evaluating the human body's balance state based on the arch index, plantar support area, and center of pressure offset.

[0116] In this embodiment, different evaluation methods are used for data of different dimensions, and data of different dimensions can be combined to improve the reliability of evaluation results. Taking plantar images as an example, since plantar images can be used to calculate or characterize the user's gait parameters or whether the body is disordered, for example, patients with otogenic vertigo may have abnormal foot pressure distribution due to vestibular-proprioceptive integration disorder. Therefore, this embodiment can calculate parameters such as arch index, plantar support area and pressure center offset based on plantar images to determine whether the user has symptoms related to otogenic vertigo.

[0117] In the specific implementation, the foot image is acquired, then converted from the BGR color space to the HSV color space, and a threshold range for green is set. Erosion and dilation operations are then performed on the image to improve image quality, remove noise, and make the edges of the footprint clearer, which facilitates subsequent processing and accurate extraction of the footprint image.

[0118] To analyze the pressure distribution on the sole of the foot in detail, the image was converted to a grayscale image, and the pressure value information in the image was mapped using the color range of Colormap Jet. The sum of all pixel values ​​in the grayscale image and the image weighted grayscale centroid were calculated to identify areas of concentrated pressure, so as to further calculate the foot length FL, maximum foot width MFW, big toe distance BTD, interankular distance IMD, effective foot length EFL, support area BoS, and foot angle α.

[0119] By analyzing the deviation between the above foot data and the foot data of normal users on a three-dimensional force platform, it is possible to determine whether the current user has deformities, limb incoordination, or other problems.

[0120] In one feasible embodiment, the step of calculating the arch index, plantar support area, and pressure center offset based on the plantar pressure distribution data includes:

[0121] The footprint image is divided into the forefoot region, the middlefoot region, and the hindfoot region;

[0122] Calculate the forefoot support area corresponding to the forefoot region, the midfoot support area corresponding to the midfoot region, and the heel support area corresponding to the heel region based on the plantar pressure distribution data.

[0123] The plantar support area is determined based on the sum of the forefoot support area, midfoot support area, and heel support area.

[0124] The arch index is obtained by calculating the area ratio between the midfoot support area and the sole support area.

[0125] The pressure center offset is calculated based on the plantar pressure distribution data and historical plantar pressure distribution data.

[0126] It should be understood that, in order to perform a more detailed analysis of different areas of the sole, this embodiment uses the minimum bounding rectangle method to segment the footprint image, referring to... Figure 5 , Figure 5This is a schematic diagram of the method for segmenting the footprint image using the minimum bounding rectangle in this embodiment. The foot area is divided into three equal parts along the height of the rectangle: front (A), middle (B), and back (C). The outlines of each part of the foot can be extracted, and the areas of these areas can be calculated. Finally, by comparing the ratio of the area of ​​the middle foot area (B) to the area of ​​the entire foot, the arch index AI can be calculated. The arch index AI is an important numerical indicator for judging foot abnormalities such as flat feet or high arches. Generally, a low arch index indicates flat feet, and a high arch index indicates high arches.

[0127] In addition, the arch index and center of pressure offset can be combined for joint analysis to identify whether the current user has a compensatory gait, so as to provide auxiliary opinions for subsequent patient rehabilitation and treatment. Compensatory gait refers to the abnormal gait pattern formed by the body replacing and compensating for the normal gait by other parts or methods when the human body is injured or has functional impairment. Examples include hemiplegic gait and abnormal gait caused by weakness of the hip abductor muscles. This embodiment does not make specific limitations on this.

[0128] In one feasible embodiment, the step of assessing human balance based on at least one of the aligned human pressure data, the plantar image, the human contour image, and the human eye image further includes:

[0129] Identify skeletal key points in each human body contour image and label them according to the type of each skeletal key point;

[0130] One-dimensional vector concatenation is performed on each marked skeletal key point to generate multiple skeletal RGB images, where the coordinates of each skeletal key point in the RGB images correspond to RGB pixel values.

[0131] The pixels of the multiple skeletal RGB images are normalized to obtain multiple target skeletal RGB images;

[0132] The target skeleton RGB image is input into a trained neural network model for analysis to obtain a second evaluation result. The second evaluation result includes at least one of the following: skeleton point trajectory, joint angle, and trunk-limb coordination.

[0133] It should be noted that in this embodiment, the human body contour image is acquired synchronously by four image acquisition devices mounted on the crossbar, as shown in the reference. Figure 6 , Figure 6This is a schematic diagram of the location of the camera device installed on the crossbar in this embodiment. There are four standard interfaces (200mm apart) on the crossbar. The cameras are fixed by threaded interfaces. The four high-definition cameras are arranged one in front of, one in front of, one in the left and one in the right of the subject. The left and right cameras are tilted at 15°, and the front and back cameras are vertically downward to ensure no blind spots and provide multi-angle information on the patient's posture and gait.

[0134] In the specific implementation, refer to Figure 7 , Figure 7 This is a schematic diagram of the motion modeling stages in this embodiment. Multi-angle cameras capture video of the patient and convert it into frame-by-frame human image data. Then, contour recognition is performed on the image data to obtain model data of the main limbs of the human body. Finally, after processing by the CNN algorithm, the motion model parameters of the human body are obtained. Specifically, the key points of the human skeleton are regarded as a dynamic trajectory sequence in the time domain. The joint position histogram distribution of the key points of the skeleton in the time domain is obtained by solving the model through Hidden Markov Model (HMM), Conditional Random Fields (CRFs), Temporal Pyramid, etc., and the rotation and displacement features of the 3D position of the key points are obtained.

[0135] The skeletal points in each image are divided into five parts: left and right arm skeletal points, left and right leg skeletal points, and torso skeletal points, and are distinguished by different colors. According to the order of presentation, the skeletal points of each frame are stretched into a one-dimensional vector. Then, the vectors of each frame in a complete video pose sequence are concatenated to form an RGB image. The (R, G, B) channels correspond to the (x, y, z) coordinates of each skeletal point, thus completing the mapping of human video pose key points to an image.

[0136] Specifically, the row definition of the human pose image is: Ri = [xi1; xi2; :::; xiN ], Gi = [yi1; yi2; :::; yiN ], Bi = [zi1; zi2; :::; ziN ], where i is the index of the key point and N is the total number of frames in the video sequence. Then, by concatenating the coordinate system pixel correspondences of the vectors in each frame, the final pose image is obtained, expressed as N×M×3, where M is the number of bone points in each frame (which remains constant).

[0137] Then, the pixels of each image are normalized:

[0138]

[0139] Where Cmax and Cmin are the maximum and minimum coordinate values ​​in the training dataset. 255 is the value that is normalized to the (0~255) pixel range of the image representation.

[0140] In one feasible embodiment, the step of assessing human balance based on at least one of the aligned human pressure data, the plantar image, the human contour image, and the human eye image further includes:

[0141] Obtain the set of pressure center coordinates corresponding to the human body pressure dataset;

[0142] The pressure center is marked and preprocessed to obtain the target pressure center coordinate set;

[0143] Based on the target pressure center coordinate set, time-domain and frequency-domain measurement indicators are calculated. The time-domain measurement indicators include at least: time-domain average distance, time-domain root mean square distance, time-domain swing path, time-domain average velocity, and human body tilt angle. The frequency-domain measurement indicators include at least: power spectral density and power spectral moment.

[0144] The time-domain fluctuation and frequency-domain offset of human pressure data are determined based on the time-domain and frequency-domain measurement indicators.

[0145] The third evaluation result is obtained based on the time-domain fluctuation and frequency-domain offset.

[0146] It should be noted that the center of pressure refers to the theoretical center point of a human body standing on a three-dimensional force platform. Specifically, refer to... Figure 8 , Figure 8 This is a schematic diagram of the coordinate axes for calculating the center of pressure on the three-dimensional force platform in this embodiment. By processing the sensor data, the center of pressure (CoP) of the human body is calculated based on the principle of torque conservation. The center of the force platform is set as the origin, the distance between two adjacent sensors is 520mm, and the planar positions of the four sensors are A(…). , B () , ), C ( , ), D ( , ),in, = = = =L=260, = = = =L=260. The Z-axis outputs of the four sensors are as follows: , , , The projection of the human body's center of gravity onto the force platform is point G, with coordinates (x, y). From the formula:

[0147]

[0148]

[0149] Substituting the coordinate values ​​into the above formula, we get:

[0150]

[0151]

[0152] This yields the coordinates (x, y) of the center of pressure (CoP) of the human body, where x represents the left-right (ML) direction and y represents the front-back (AP) direction. Based on CoP, time-domain measurement indicators can be calculated, including: mean distance to CoP (MDIST), root mean square distance (RDIST), total swing path (LNG), mean velocity (MVEL), body tilt angle, and 95% power frequency.

[0153] Coordinates of the average CoP ( , The calculation formula is as follows:

[0154]

[0155]

[0156] In the formula: N is the total number of CoP sampling points, that is, the number of CoP location data collected within the test time, T is the sampling time (unit: seconds), and f is the sampling frequency (unit: Hz), satisfying N=T×f.

[0157] Subtracting the average CoP from the initial CoP eliminates bias or noise in the data. The calculation formula is as follows:

[0158]

[0159] Time-domain measurement metrics include: Mean Distance to CoP (MDIST), Root Mean Square Distance (RDIST), Total Swing Path (LNG), Mean Velocity (MVEL), and Body Tilt Angle. Specifically, the Mean Distance (MDIST) is the average distance to the mean CoP, and its calculation formula is as follows:

[0160]

[0161] The formula for calculating the root mean square distance (RDIST) is:

[0162]

[0163] The formula for calculating the total LNG swing path is as follows:

[0164]

[0165] Average velocity (MVEL) refers to the total swing path divided by the sampling time, and its calculation formula is:

[0166]

[0167] The height of the center of gravity, Gh, was estimated using biomechanical methods, where Gh = 0.55h, and h is the subject's height. GL_ap_plat is the projection of the center of gravity onto the sagittal plane of the platform, approximated by the displacement of CoP. Accordingly, the formula for calculating the tilt angle of the user in a standing posture is:

[0168]

[0169] The power spectral density given in the frequency domain measurement specifications is calculated based on the Fourier transform, and the specific formula is as follows:

[0170]

[0171] Where Y(f) is the Fourier transform of the signal.

[0172] The formula for calculating the spectral moment is:

[0173]

[0174] in It is the frequency resolution, the reciprocal of the sampling duration, i.e. =1 / T Hz. The calculation frequency range of the power spectral density is from 1 / T to 5 Hz. Due to the influence of the DC component and the fundamental frequency component, the first two frequency points are excluded. In the above formula, the value of i ranges from 3 to 100, and m is the number of the frequency point.

[0175] 95% power frequency is v Where v is the smallest integer obtained by the following formula:

[0176]

[0177] Where u0 is the zero-order spectral moment, which is the sum of the power spectral densities at all frequency points.

[0178] In one embodiment, to accurately capture human eye images, a face image can be captured first, and then the eyes can be separated from the image, which can be done in three stages: image preprocessing, face detection, and eye localization. Specifically, image preprocessing refers to using median filtering and low-pass Gaussian filtering to remove noise from the image. This operation can also blur image edges; using a high-quality camera can reduce image noise from the source.

[0179] Face detection refers to the process based on skin color detection technology. First, skin color detection technology is used to initially extract the face region from the image. Then, morphological processing is used to fill in holes in the face region and smooth the edges. Finally, the contour of the face region is extracted, and the bounding rectangle of the contour is used as the face's location. After detecting the face region, to further determine the face's position, the facial contour is extracted, and the face's location is determined by the bounding rectangle of the contour. When extracting the facial contour from the region obtained through skin color detection, many contours are often obtained. Therefore, this embodiment also sets two constraints: 1. The extracted contour must be the outer contour of the contour to remove the inner contours of the eyebrow and eye areas; 2. The area enclosed by the contour should be greater than 20% of the image area, which can effectively remove objects in the background.

[0180] Human eye localization refers to extracting images of the human eye, i.e., determining the size of the eye and the coordinates of its center. This is achieved using an integral projection algorithm. The integral projection algorithm first utilizes prior knowledge of facial region distribution, and then uses the facial grayscale integral projection curve to further determine the position of the human eye, removing interference from the eyebrows. The essence of grayscale integration is to obtain grayscale distribution information, i.e., grayscale peaks and troughs, by projecting the image in both vertical and horizontal directions, thereby achieving the purpose of target detection. The integral projection results of the facial image in both horizontal and vertical directions are shown in the figure. The vertical coordinate interval [60, 240] represents the prior knowledge and experience of the human eye. Within this range, the positions of the eyebrows and eyes are relatively dark, so there are obvious troughs in the integral projection function, i.e., local extrema. Therefore, the features can be used to complete the localization of the human eye and eyebrows in the horizontal direction.

[0181] After obtaining the eye coordinates using the above positioning method, and combining this with the eye size calculated from the facial region, the eye image can be segmented. The segmentation result is shown in the figure; the black area of ​​the eye is marked in green, the center of the eye is marked with a red dot, and the eye area is selected with a green box, thus obtaining the human eye image.

[0182] This embodiment activates the projection device and three-dimensional force stage in the human balance assessment device; acquires pressure information on the standing platform through a three-dimensional force sensor, and calculates human pressure data based on the pressure information; acquires human contour images and human eye images through a pose acquisition device, and acquires foot images through a foot image acquisition device; assesses human balance state based on at least one of the human pressure data, the foot images, the human contour images, and the human eye images. By analyzing human balance state through multi-dimensional balance assessment, a comprehensive dynamic balance assessment is achieved with higher accuracy. At the same time, the mechanical and processing parts are flexibly combined, making it more applicable.

[0183] This application also provides a human balance assessment device; please refer to... Figure 9The human balance assessment device includes:

[0184] The startup module 10 is used to start the projection device and the three-dimensional force table in the human balance assessment device.

[0185] The calculation module 20 is used to acquire pressure information on the standing platform through a three-dimensional force sensor and calculate human body pressure data based on the pressure information.

[0186] The acquisition module 30 is used to acquire human contour images and human eye images through a pose acquisition device, and to acquire foot images through a foot image acquisition device.

[0187] The assessment module 40 is used to assess the human balance state based on at least one of the human pressure data, the foot image, the human contour image, and the human eye image.

[0188] This embodiment activates the projection device and three-dimensional force stage in the human balance assessment device; acquires pressure information on the standing platform through a three-dimensional force sensor, and calculates human pressure data based on the pressure information; acquires human contour images and human eye images through a pose acquisition device, and acquires foot images through a foot image acquisition device; assesses human balance state based on at least one of the human pressure data, the foot images, the human contour images, and the human eye images. By analyzing human balance state through multi-dimensional balance assessment, a comprehensive dynamic balance assessment is achieved with higher accuracy. At the same time, the mechanical and processing parts are flexibly combined, making it more applicable.

[0189] In one embodiment, the evaluation module 40 is further configured to acquire data timestamps of the human body pressure data, the plantar image, the human body contour image, and the human eye image; perform time alignment of the human body pressure data, the plantar image, the human body contour image, and the human eye image according to the data timestamps; and evaluate the human body balance state based on at least one of the aligned human body pressure data, the plantar image, the human body contour image, and the human eye image.

[0190] In one embodiment, the evaluation module 40 is further configured to preprocess the plantar image to obtain a footprint image; acquire the pixel value and weighted grayscale centroid of each pixel in the footprint image; calculate plantar pressure distribution data based on the pixel value and weighted grayscale centroid of each pixel; calculate the arch index, plantar support area, and pressure center offset based on the plantar pressure distribution data; and evaluate the human balance state based on the arch index, plantar support area, and pressure center offset to obtain a first evaluation result.

[0191] In one embodiment, the evaluation module 40 is further configured to divide the footprint image to obtain a forefoot region, a midfoot region, and a heel region; calculate the forefoot support area corresponding to the forefoot region, the midfoot support area corresponding to the midfoot region, and the heel support area corresponding to the heel region based on the plantar pressure distribution data; determine the plantar support area based on the sum of the forefoot support area, the midfoot support area, and the heel support area; calculate the area ratio between the midfoot support area and the plantar support area to obtain the arch index; and calculate the pressure center offset based on the plantar pressure distribution data and historical plantar pressure distribution data.

[0192] In one embodiment, the evaluation module 40 is further configured to identify skeletal key points in each human contour image and mark them according to the type of each skeletal key point; perform one-dimensional vector concatenation based on each marked skeletal key point to generate multiple skeletal RGB images, wherein the coordinates of each skeletal key point in the skeletal RGB images correspond to RGB pixel values; normalize the pixels of the multiple skeletal RGB images to obtain multiple target skeletal RGB images; input the target skeletal RGB images into a trained neural network model for analysis to obtain a second evaluation result, wherein the second evaluation result includes at least one of skeletal point trajectory, joint angle, and trunk-limb coordination.

[0193] In one embodiment, the evaluation module 40 is further configured to: acquire the pressure center coordinate set corresponding to the human body pressure dataset; preprocess the pressure centers by marking them to obtain the target pressure center coordinate set; calculate time-domain and frequency-domain measurement indicators based on the target pressure center coordinate set, wherein the time-domain measurement indicators include at least: time-domain average distance, time-domain root mean square distance, time-domain swing path, time-domain average velocity, and human body tilt angle; and the frequency-domain measurement indicators include at least: power spectral density and power spectral moment; determine the time-domain fluctuation and frequency-domain offset of the human body pressure data based on the time-domain and frequency-domain measurement indicators; and obtain a third evaluation result based on the time-domain fluctuation and frequency-domain offset.

[0194] This application provides a human balance assessment device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the human balance assessment method in the first embodiment described above.

[0195] The following is for reference. Figure 10The diagram illustrates a structural schematic suitable for implementing the human balance assessment device in the embodiments of this application. The human balance assessment device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 10 The human balance assessment device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0196] like Figure 10 As shown, the human balance assessment device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the human balance assessment device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. The communication device 1009 allows the human balance assessment device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows human balance assessment devices with various systems, it should be understood that implementing or having all of the systems shown is not required. More or fewer systems may be implemented alternatively.

[0197] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0198] The human balance assessment device provided in this application, employing the human balance assessment method described in the above embodiments, can solve the technical problems of human balance assessment. Compared with the prior art, the beneficial effects of the human balance assessment device provided in this application are the same as those of the human balance assessment method provided in the above embodiments, and other technical features of this human balance assessment device are the same as those disclosed in the method of the previous embodiment, and will not be repeated here.

[0199] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0200] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0201] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to perform the human balance assessment method in the above embodiments.

[0202] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, 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 devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0203] The aforementioned computer-readable storage medium may be included in the human balance assessment device; or it may exist independently and not be assembled into the human balance assessment device.

[0204] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the human balance assessment device, cause the human balance assessment device to perform a human balance assessment.

[0205] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0206] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0207] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0208] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described human balance assessment method, thereby solving the technical problem of human balance assessment. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the human balance assessment method provided in the above embodiments, and will not be repeated here.

[0209] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the human balance assessment method described above.

[0210] The computer program product provided in this application can solve the technical problem of human balance assessment. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the human balance assessment method provided in the above embodiments, and will not be repeated here.

[0211] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A human balance assessment device, characterized by, The human body balance evaluation device comprises a main frame, a three-dimensional force platform, a posture acquisition device and a control unit; The main frame is arranged on the ground and is used for mounting the three-dimensional force platform and the posture acquisition device; The three-dimensional force platform comprises a base and a balance force platform, two guide rails are arranged in parallel on the base, and the balance force platform is in sliding connection with the guide rails, wherein the balance force platform at least comprises a light source, a three-dimensional force sensor, a motor, an AF-N actuator, a plantar image acquisition device and a standing platform; The three-dimensional force platform adjusts the balance state of the user standing on the three-dimensional force platform by moving on the guide rails and / or adjusting the horizontal state of the standing platform; The three-dimensional force platform is also used for collecting the plantar image and human body pressure data of the user standing on the three-dimensional force platform; The posture acquisition device is used for collecting the human eye image and human body contour image of the user standing on the three-dimensional force platform; The control unit is used for evaluating the human body balance state according to the human body pressure data, the plantar image, the human body contour image and the human eye image with the same time stamp as the pressure data or the image as a group of data; specifically comprising the following steps, acquiring the data time stamp of the human body pressure data, the plantar image, the human body contour image and the human eye image; aligning the human body pressure data, the plantar image, the human body contour image and the human eye image according to the data time stamp; evaluating the human body balance state according to at least one of the aligned human body pressure data, the plantar image, the human body contour image and the human eye image; The evaluation of the human body balance state according to at least one of the aligned human body pressure data, the plantar image, the human body contour image and the human eye image further comprises: acquiring a set of pressure center coordinates corresponding to the human body pressure data set; preprocessing the set of pressure center coordinates to obtain a set of target pressure center coordinates; calculating time domain measurement indexes and frequency domain measurement indexes according to the set of target pressure center coordinates, wherein the time domain measurement indexes at least include time domain average distance, time domain root mean square distance, time domain swing path, time domain average speed and human body inclination angle, and the frequency domain measurement indexes at least include power spectral density and power spectral moment; determining the time domain fluctuation and the frequency domain offset of the human body pressure data according to the time domain measurement indexes and the frequency domain measurement indexes; obtaining a third evaluation result according to the time domain fluctuation and the frequency domain offset.

2. The human balance assessment device of claim 1, wherein, The human body balance evaluation device further comprises a projection device; The main frame is provided with a movable mounting horizontal rod; The projection device is arranged on the mounting horizontal rod; The posture acquisition device is used for collecting the human eye image when the user views the target image projected by the projection device.

3. The human balance assessment device of claim 1, wherein, The human body balance evaluation device further comprises: A human body protection device is suspended on the mounting horizontal rod and is used for protecting the balance state of the user when the three-dimensional force platform is started.

4. A method of human balance assessment, characterised in that, The human body balance evaluation method is applied to the human body balance evaluation device as claimed in any one of claims 1 to 3; The human body balance evaluation method comprises: starting a projection device and a three-dimensional force platform in the human body balance evaluation device; acquiring pressure information on the standing platform through the three-dimensional force sensor, and calculating human body pressure data based on the pressure information; acquiring human body contour images and human eye images through the posture acquisition device, and acquiring plantar images through the plantar image acquisition device; performing human body balance state evaluation on the human body pressure data, the plantar images, the human body contour images, and the human eye images with the same time stamp as a group of data; specifically including the following steps, acquiring data time stamps of the human body pressure data, the plantar images, the human body contour images, and the human eye images; time aligning the human body pressure data, the plantar images, the human body contour images, and the human eye images according to the data time stamps; performing human body balance state evaluation on at least one of the aligned human body pressure data, the plantar images, the human body contour images, and the human eye images; the human body balance state evaluation on at least one of the aligned human body pressure data, the plantar images, the human body contour images, and the human eye images further includes: acquiring a set of pressure center coordinates corresponding to the human body pressure data set; preprocessing the set of pressure center coordinates to obtain a target set of pressure center coordinates; calculating time domain measurement indicators and frequency domain measurement indicators according to the target set of pressure center coordinates, the time domain measurement indicators at least including time domain average distance, time domain root mean square distance, time domain swing path, time domain average speed, and human body inclination angle, and the frequency domain measurement indicators at least including power spectral density and power spectral moment; determining time domain fluctuations and frequency domain offsets of the human body pressure data according to the time domain measurement indicators and the frequency domain measurement indicators; obtaining a third evaluation result according to the time domain fluctuations and the frequency domain offsets.

5. The method of balance assessment of a human body according to claim 4, characterized in that, the human body balance state evaluation on at least one of the aligned human body pressure data, the plantar images, the human body contour images, and the human eye images includes: preprocessing the plantar images to obtain footprint images; acquiring pixel values and image weighted gray gravity centers of each pixel point in the footprint images; calculating plantar pressure distribution data based on the pixel values and the image weighted gray gravity centers of each pixel point; calculating arch index, plantar support area, and pressure center offset according to the plantar pressure distribution data; performing human body balance state evaluation according to the arch index, the plantar support area, and the pressure center offset to obtain a first evaluation result.

6. The method of balance assessment of a human body according to claim 5, wherein, the calculation of the arch index, the plantar support area, and the pressure center offset according to the plantar pressure distribution data includes: dividing the footprint images to obtain forefoot region, midfoot region, and hindfoot region; calculating forefoot support area corresponding to the forefoot region, midfoot support area corresponding to the midfoot region, and hindfoot support area corresponding to the hindfoot region according to the plantar pressure distribution data; determining the plantar support area according to the sum of the forefoot support area, the midfoot support area, and the hindfoot support area; Calculate the area ratio between the foot arch support area and the foot bottom support area to obtain the foot arch index; Calculate the pressure center offset according to the foot pressure distribution data and historical foot pressure distribution data.

7. The method of balance assessment of a human body according to claim 6, wherein, The human body balance state evaluation according to at least one of the aligned human body pressure data, the foot bottom image, the human body contour image, and the human eye image further includes: Identify the bone key points in each human body contour image, and mark according to the type of each bone key point; According to each marked bone key point, a one-dimensional vector is concatenated to generate a plurality of bone RGB images, and the coordinates of each bone key point in the bone RGB image correspond to the RGB pixel value; Pixel normalization is performed on the plurality of bone RGB images to obtain a plurality of target bone RGB images; The plurality of target bone RGB images are input into the trained neural network model for analysis to obtain a second evaluation result, which at least includes at least one of bone point trajectory, joint angle, and trunk-limb coordination.

8. A human balance assessment device, characterised in that, It includes: The starting module is used to start the projection device and the three-dimensional force platform in the human body balance evaluation equipment; The calculation module is used to obtain the pressure information on the standing platform through the three-dimensional force sensor, and calculate the human body pressure data based on the pressure information; The acquisition module is used to acquire human body contour images and human eye images through the pose acquisition device, and to acquire foot bottom images through the foot bottom image acquisition device; The evaluation module is used to evaluate the human body balance state according to the human body pressure data, the foot bottom image, the human body contour image, and the human eye image, that is, the pressure data or image with the same time stamp as a group of data; Specifically It includes the following steps, Obtain the data time stamp of the human body pressure data, the foot bottom image, the human body contour image, and the human eye image; According to the data time stamp, the human body pressure data, the foot bottom image, the human body contour image, and the human eye image are time-aligned; According to at least one of the aligned human body pressure data, the foot bottom image, the human body contour image, and the human eye image, the human body balance state evaluation is performed; The human body balance state evaluation according to at least one of the aligned human body pressure data, the foot bottom image, the human body contour image, and the human eye image further includes: Obtain the pressure center coordinate set corresponding to the human body pressure data set; Preprocess the pressure center coordinate set to obtain a target pressure center coordinate set; According to the target pressure center coordinate set, calculate the time domain measurement index and the frequency domain measurement index, the time domain measurement index at least includes: time domain average distance, time domain root mean square distance, time domain swing path, time domain average speed and body inclination, the frequency domain measurement index at least includes: power spectral density and power spectrum moment; According to the time domain fluctuation and the frequency domain offset of the human body pressure data, the time domain fluctuation and the frequency domain offset are determined according to the time domain measurement index and the frequency domain measurement index; According to the time domain fluctuation and the frequency domain offset, a third evaluation result is obtained.

Images