CN-121971071-A - Intelligent mirror multidimensional health assessment system

CN121971071ACN 121971071 ACN121971071 ACN 121971071ACN-121971071-A

Abstract

The invention provides an intelligent mirror multi-dimensional health assessment system, which relates to the technical field of health assessment and comprises an acquisition module, a pointing module and a control module, wherein the acquisition module is used for carrying out human body skeletonizing processing on an acquired continuous three-dimensional image sequence of user movement and position information of user limbs in space, extracting three-dimensional space coordinate data of key joints of the user body to obtain a real-time skeleton model of the user movement, and the pointing module is used for establishing a first reference vector pointing to the center points of left and right shoulder joints by using the space base points and a second reference vector pointing to the center points of left and right hip joints based on the real-time skeleton model so as to determine a main body reference plane of the user movement. According to the invention, by extracting the motion skeleton data, constructing the reference and dynamic characteristic plane, calculating the key index and evaluating, generating the motion posture correction parameter and the visual calibration guide, the evaluation and real-time optimization of the motion quality are realized, the motion effect is improved, and the damage risk is reduced.

Inventors

ZHENG SHUYAN
ZHANG ZIHUAN

Assignees

浙江苏必利尔科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260126

Claims (10)

1. A smart mirror multi-dimensional health assessment system, comprising: the acquisition module is used for carrying out human body skeletonization on the acquired continuous three-dimensional image sequence of the user movement and the position information of the user limbs in the space, extracting three-dimensional space coordinate data of key joints of the user body and obtaining a real-time skeleton model of the user movement; the pointing module is used for taking the central point of the trunk of the user as a space base point based on the real-time skeleton model, and establishing a first reference vector pointing to the central points of left and right shoulder joints by the space base point and a second reference vector pointing to the central points of left and right hip joints so as to determine a main body reference plane of the movement of the user; The construction module is used for selecting four characteristic points of the left shoulder joint, the right shoulder joint, the left hip joint and the right hip joint on the basis of the main body reference plane based on the real-time skeleton model so as to construct a dynamic characteristic plane representing the motion state of limbs; The calculation module is used for obtaining a space attitude offset by calculating a normal vector included angle between the dynamic characteristic plane and the main body reference plane, and constructing a minimum bounding rectangle based on the space positions of the four characteristic points so as to obtain a motion stability index; the evaluation module is used for carrying out health evaluation on the motion quality of the user through a pre-trained evaluation model based on the space attitude offset and the motion stability index and combining the three-dimensional space coordinate data of key joints of the body of the user, and generating a comprehensive evaluation result containing motion attitude correction parameters; And the display module is used for generating visual contents comprising the spatial attitude calibration guidance and the motion stability optimization suggestion according to the comprehensive evaluation result and displaying the visual contents on the intelligent mirror surface in real time.
2. The smart mirror multi-dimensional health assessment system according to claim 1, wherein the human body skeletonizing process is performed on the obtained continuous three-dimensional image sequence of the user motion and the position information of the user limb in the space, and three-dimensional space coordinate data of key joints of the user body are extracted, so as to obtain a real-time skeleton model of the user motion, and the method comprises the following steps: Capturing position information of limbs of a user in space through an infrared sensor, wherein the position information comprises distance and azimuth data of each limb part to the sensor; identifying a plurality of key nodes of the user's body based on the sequence of continuous three-dimensional images and the positional information; and calculating coordinate data of each node in a three-dimensional space according to the plurality of key nodes, and constructing a real-time skeleton model based on the three-dimensional space coordinate data of each node and the human skeleton connection relation.
3. The smart mirror multi-dimensional health assessment system of claim 2, wherein establishing a first reference vector pointing to left and right shoulder joint center points with spatial base points and a second reference vector pointing to left and right hip joint center points to determine a subject reference plane of user motion based on the real-time skeletal model using the user torso center point as the spatial base point comprises: Calculating the space coordinates of the central point of the trunk of the user based on the three-dimensional space coordinate data of the left hip joint and the right hip joint in the real-time skeleton model; calculating the space coordinate midpoints of the left shoulder joint center point and the right shoulder joint center point to be used as shoulder center points; Calculating to obtain a first reference vector according to the space coordinates of the trunk center point and the shoulder center point; And determining a normal vector of the main body reference plane through vector cross product operation based on the first reference vector and the second reference vector, and establishing the main body reference plane of the user motion.
4. The smart mirror multi-dimensional health assessment system according to claim 3, wherein selecting four feature points of left and right shoulder joints, left and right hip joints based on a real-time skeletal model on the basis of a subject reference plane to construct a dynamic feature plane characterizing a limb movement state comprises: Based on a real-time skeleton model, acquiring real-time three-dimensional space coordinates of four characteristic points, including real-time three-dimensional space coordinates of a left shoulder joint center point, a right shoulder joint center point, a left hip joint center point and a right hip joint center point; Establishing a first reference triangle plane through the left shoulder joint center point, the right shoulder joint center point and the left hip joint center point; Based on the normal vector of the first reference triangle plane and the normal vector of the second reference triangle plane, calculating to obtain the synthetic method vector of the dynamic characteristic plane; And constructing a complete dynamic characteristic plane for representing the limb movement state according to the comprehensive method vector and the spatial distribution characteristics of the four characteristic points.
5. The smart mirror multi-dimensional health assessment system according to claim 4, wherein the obtaining of the spatial attitude offset by calculating the normal vector angle between the dynamic feature plane and the main body reference plane, and the constructing of the minimum bounding rectangle based on the spatial positions of the four feature points, to obtain the motion stability index, comprises: Calculating a space included angle between the combined method vector of the dynamic characteristic plane and the normal vector of the main body reference plane to obtain a space attitude offset; Determining a minimum circumscribed rectangle containing the spatial positions of all feature points based on three-dimensional space coordinates of a left shoulder joint center point, a right shoulder joint center point, a left hip joint center point and a right hip joint center point; Calculating the length-width ratio of the minimum circumscribed rectangle to obtain the limb movement symmetry index, and calculating the inclination angle of the minimum circumscribed rectangle relative to the main body reference plane to obtain the limb movement stability index.
6. The smart mirror multi-dimensional health assessment system according to claim 5, wherein based on the spatial pose offset and the motion stability index, in combination with three-dimensional spatial coordinate data of key joints of the user's body, performing health assessment of the user's motion quality through a pre-trained assessment model, generating a comprehensive assessment result including motion pose correction parameters, comprising: Inputting the space attitude offset, the limb movement symmetry index, the limb movement stability index and the three-dimensional space coordinate data of the key joints into a pre-trained evaluation model for analysis and processing to obtain preliminary evaluation parameters including a shoulder joint angle offset value, a hip joint angle offset value and a trunk stability coefficient; Generating motion pose correction parameters for the shoulder joint motion angle, the hip joint motion angle and the trunk stability of the user based on the preliminary evaluation parameters; and calculating to obtain a comprehensive health score containing a plurality of dimension scores according to the degree of difference between the motion posture correction parameters and the current motion state of the user.
7. The smart mirror multi-dimensional health assessment system according to claim 6, wherein the input of the three-dimensional spatial coordinate data of the spatial pose offset, the limb movement symmetry index, the limb movement stability index, and the key joint to the pre-trained assessment model for analysis and processing, to obtain the preliminary assessment parameters including the shoulder joint angle offset value, the hip joint angle offset value, and the torso stability coefficient, comprises: Calculating real-time motion angles of the left shoulder joint and the right shoulder joint and real-time motion angles of the left hip joint and the right hip joint based on three-dimensional space coordinate data of the key joints; comparing the real-time movement angle with a pre-stored standard movement angle to generate initial angle deviation data of the shoulder joint and the hip joint; feature level fusion is carried out on initial angle deviation data, space attitude offset, limb movement symmetry indexes and limb movement stability indexes, a multi-dimensional feature vector is constructed, the multi-dimensional feature vector is input into a pre-trained evaluation model, and feature extraction layers in the evaluation model carry out feature mining on the multi-dimensional feature vector to obtain high-dimensional abstract features; and carrying out nonlinear transformation and weight distribution on the high-dimensional abstract features through a full connection layer in the evaluation model to obtain preliminary evaluation parameters, wherein the preliminary evaluation parameters comprise shoulder joint angle deviation values for quantifying the normalization of shoulder motions, hip joint angle deviation values for quantifying the normalization of hip motions and trunk stability coefficients for representing the stability degree of the trunk in motion.
8. The smart mirror multi-dimensional health assessment system according to claim 7, wherein generating visual content comprising spatial pose calibration guidance and motion stability optimization advice based on the comprehensive assessment results and displaying the visual content on the smart mirror in real time, comprises: Generating shoulder joint angle calibration guide information based on correction content for the shoulder joint angle in the motion gesture correction parameters; generating hip joint angle calibration guide information based on correction content aiming at the hip joint angle in the motion gesture correction parameters; Integrating the shoulder joint angle calibration guidance, the hip joint angle calibration guidance and the gravity center distribution adjustment suggestion to form complete visual guidance content; And identifying the difference between the current motion state of the user and the target motion state through different colors, and displaying the visual guidance content on the user image of the intelligent mirror in a real-time superposition manner.
9. A computing device, comprising: One or more processors; storage means for storing one or more programs that when executed by the one or more processors cause the one or more processors to perform the system of any of claims 1-8.
10. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a program which, when executed by a processor, performs the system according to any of claims 1 to 8.

Description

Intelligent mirror multidimensional health assessment system Technical Field The invention relates to the technical field of health assessment, in particular to an intelligent mirror multidimensional health assessment system. Background Smart mirrors with health monitoring capabilities typically capture a user's motion image via an integrated camera and perform a preliminary gesture analysis based thereon, however, such systems may have limitations in terms of their assessment dimensions in analyzing some movements involving body core swing, such as bending, swivelling, etc. In particular, some systems may rely primarily on changes in the position of the articulation point within the two-dimensional image plane to analyze the pose, such as by monitoring the pixel coordinates of the articulation point in the picture to estimate the magnitude of the motion, which sometimes makes accurate quantification of spatial displacement and rotation of the limb in a direction perpendicular to the mirror plane difficult. For example, when a user bends sideways, the inclination angle of the trunk and the stability of the pelvis are key to evaluating the quality of the motion, but if the user views from only two-dimensional angles, the actual displacement of the body in three-dimensional space may not be sufficiently captured, resulting in insufficient evaluation information. Disclosure of Invention The technical problem to be solved by the invention is to provide an intelligent mirror multidimensional health assessment system which provides scientific guidance for motion gesture optimization. In order to solve the technical problems, the technical scheme of the invention is as follows: in a first aspect, a smart mirror multi-dimensional health assessment system includes: the acquisition module is used for carrying out human body skeletonization on the acquired continuous three-dimensional image sequence of the user movement and the position information of the user limbs in the space, extracting three-dimensional space coordinate data of key joints of the user body and obtaining a real-time skeleton model of the user movement; the pointing module is used for taking the central point of the trunk of the user as a space base point based on the real-time skeleton model, and establishing a first reference vector pointing to the central points of left and right shoulder joints by the space base point and a second reference vector pointing to the central points of left and right hip joints so as to determine a main body reference plane of the movement of the user; The construction module is used for selecting four characteristic points of the left shoulder joint, the right shoulder joint, the left hip joint and the right hip joint on the basis of the main body reference plane based on the real-time skeleton model so as to construct a dynamic characteristic plane representing the motion state of limbs; The calculation module is used for obtaining a space attitude offset by calculating a normal vector included angle between the dynamic characteristic plane and the main body reference plane, and constructing a minimum bounding rectangle based on the space positions of the four characteristic points so as to obtain a motion stability index; the evaluation module is used for carrying out health evaluation on the motion quality of the user through a pre-trained evaluation model based on the space attitude offset and the motion stability index and combining the three-dimensional space coordinate data of key joints of the body of the user, and generating a comprehensive evaluation result containing motion attitude correction parameters; And the display module is used for generating visual contents comprising the spatial attitude calibration guidance and the motion stability optimization suggestion according to the comprehensive evaluation result and displaying the visual contents on the intelligent mirror surface in real time. In a second aspect, a computing device includes: One or more processors; And a storage device for storing one or more programs that, when executed by the one or more processors, cause the one or more processors to implement the system. In a third aspect, a computer readable storage medium has a program stored therein, which when executed by a processor, implements the system. The scheme of the invention at least comprises the following beneficial effects: the three-dimensional space information of the user motion is captured, a real-time skeleton model is built, the motion scene of the limb, which relates to the swinging of a core part, in the direction perpendicular to a mirror surface is captured, the problem that the system is difficult to quantify the three-dimensional motion characteristics is solved, the health assessment is more comprehensive, the posture normalization and the limb stability in the user motion process can be objectively reflected based on the construction of a main body reference plane and