CN112966370A - Design method of human body lower limb muscle training system based on Kinect - Google Patents
Design method of human body lower limb muscle training system based on Kinect Download PDFInfo
- Publication number
- CN112966370A CN112966370A CN202110177730.9A CN202110177730A CN112966370A CN 112966370 A CN112966370 A CN 112966370A CN 202110177730 A CN202110177730 A CN 202110177730A CN 112966370 A CN112966370 A CN 112966370A
- Authority
- CN
- China
- Prior art keywords
- action
- lower limb
- standard
- training
- muscle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Processing Or Creating Images (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
A design method of a human body lower limb muscle training system based on Kinect is based on the following equipment: the computer is in signal connection with the Kinect somatosensory camera, and a Kinect program, action simulation software and a format conversion program matched with the Kinect and the action simulation software are loaded in the computer; the lower limb muscle exercising method comprises the following steps: the first step is as follows: establishing a standard action database; the second step is that: establishing a human-computer interaction interface; the third step: establishing an action training judgment system; the fourth step: establishing an action reproduction system; the fifth step: and establishing a training effect analysis system. The design can collect the action data of the lower limbs without wearing sensing equipment, meanwhile, the training effect is enhanced by utilizing a human-computer interaction interface, an evaluation system is set according to the age and the gender, and the reliability of a judgment result is high.
Description
Technical Field
The invention relates to a design method of a human body lower limb muscle training system based on Kinect, which is particularly suitable for realizing interactive training of lower limb muscles by utilizing a skeleton tracking technology.
Background
In physical fitness, many people neglect exercising the lower limbs because exercising the upper body muscles more easily demonstrates the aesthetic. It is unknown that lower limb exercise, particularly leg exercise, is more important for maintaining physical health. The traditional lower limb muscle body-building training is mainly accompanied and monitored by a body-building coach so as to meet the training requirement of movement.
With the development of scientific technology, the judgment of human body posture and motion mainly includes two types, namely contact type and non-contact type, the non-contact type mainly includes a vision-based image recognition technology, and the contact type commonly includes a wearable-based recognition technology. The image recognition based on computer vision can avoid the constraint brought by the wearing product to people.
Disclosure of Invention
The invention aims to solve the problems of inconvenience in wearing and lack of interaction in training in the prior art, and provides a design method of a Kinect-based human lower limb muscle training system which is free of wearing and convenient for interactive training.
In order to achieve the above purpose, the technical solution of the invention is as follows:
a design method of a human body lower limb muscle training system based on Kinect is based on the following equipment: the method comprises the following steps: the Kinect motion sensing camera is in signal connection with the computer, and a Kinect program, motion simulation software and a format conversion program matched with the Kinect and the motion simulation software are loaded in the computer;
the lower limb muscle exercising method comprises the following steps:
the first step is as follows: establishing a standard action database, selecting the crowd with no physical disabilities and normal lower limb functions, finishing the determination of the standard human lower limb training actions according to the gender and age grouping from the anatomical perspective, classifying the determined lower limb actions according to the lower limb expansion change conditions of different lower limb training actions, establishing the standard human lower limb action database, and storing the images of the standard human lower limb training actions into the standard human lower limb action database;
acquiring lower limb skeleton data flow information of each classified standard human lower limb training action by using a Kinect somatosensory camera, analyzing and processing the data flow information, and recording relevant angle characteristics corresponding to the standard human lower limb training action; extracting lower limb muscle joint coordinates and related angle characteristics, importing the lower limb muscle joint coordinates and the related angle characteristics into action simulation software, establishing a standard action human body lower limb skeleton muscle simulation model, extracting lower limb muscle joint coordinates and related angle characteristic data, converting the lower limb muscle joint coordinates and the related angle characteristic data into a data format which can be identified by the action simulation software, and then performing inverse dynamics simulation by utilizing the action simulation software to obtain standard muscle tendon length and standard muscle contraction quantity in an action process;
corresponding the obtained quasi muscle tendon length and the standard muscle contraction quantity of each muscle to corresponding muscles of a standard action human lower limb skeletal muscle simulation model to generate a reproduced image of a standard action;
storing the standard action related angle characteristics, the standard muscle tendon length, the standard muscle contraction quantity, the standard action human lower limb skeleton simulation model and the reproduced image of the standard action into a standard human lower limb action database as corresponding standard human lower limb training action database information;
the second step is that: establishing a human-computer interaction interface, establishing a visual human-computer interaction interface, and realizing human-computer interaction during human lower limb muscle training, wherein the human-computer interaction interface comprises: the system comprises a function selection interface, a training action guide interface, a data analysis and evaluation interface and an action reproduction comparison interface;
the third step: establishing an action training judgment system, establishing a human body lower limb muscle training judgment standard, and quantifying a training effect:
establishing standard action judgment standards, and setting standard range values of the relevant angle characteristic data of each standard action according to different age groups;
when the action training judging system carries out lower limb training, analyzing and processing lower limb skeleton data flow information acquired by the Kinect somatosensory camera, comparing the processed corresponding relevant angle characteristic data with a standard action standard database of a corresponding age by using an algorithm, judging that the action reaches the standard if a value obtained after comparison is within a standard range value, and otherwise, judging that the action does not reach the standard;
when training is finished, calculating the standard reaching percentage of the training actions, and outputting the standard reaching condition of each training action and the standard reaching percentage of the training actions to a data analysis and evaluation interface;
the fourth step: establishing an action reproduction system, establishing human body lower limb action reproduction simulation, and further judging an action training effect:
establishing an action reproduction system, reproducing each collected completed action process in action simulation software, and conveniently paying attention to the action details of a trainer;
when the action reproduction system receives a group of data stream information of training actions, analyzing and processing the lower limb skeleton data stream information acquired by the Kinect somatosensory camera, extracting the coordinates and angles of the muscle joints of the lower limbs, importing the coordinates and angles into action simulation software, modifying the parameters of a human lower limb skeleton muscle simulation model to obtain a lower limb action reproduction model, and reproducing the actions finished by a trainer;
the fifth step: establishing a training effect analysis system, analyzing the training effect of the human lower limb muscles, and quantitatively analyzing the effect:
establishing a lower limb muscle training effect evaluation standard, and setting different evaluation range values according to different age groups for standard muscle tendon length and standard muscle contraction quantity data of each standard action;
establishing a training effect analysis system, extracting lower limb muscle joint coordinates and related angle characteristic data, converting the lower limb muscle joint coordinates and the related angle characteristic data into data formats which can be identified by action simulation software, performing inverse dynamics simulation by utilizing the action simulation software, calculating the force and moment required by each muscle in the lower limb movement, further calculating the length of muscle tendons of the lower limb and muscle contraction quantity data, comparing the obtained length of the muscle tendons of the lower limb and the muscle contraction quantity data with the evaluation range value of the training effect evaluation standard of the corresponding age, obtaining a training effect evaluation result, and further quantifying the evaluation of the training effect;
and corresponding the obtained evaluation effect of each muscle to the corresponding muscle of the lower limb action reproduction model obtained in the fourth step to generate a reproduction image of the training action of the user, and outputting the reproduction image of the training action of the user and the reproduction image of the corresponding standard action to an action reproduction comparison interface for displaying.
The relevant angular features include: the left knee joint angle, the right knee joint angle, the hip opening angle, the included angle between the connecting line of the left hip joint and the right hip joint and the X axis, the included angle between the left thigh and the X axis, the included angle between the right thigh and the X axis, the included angle between the connecting line of the left knee and the left shoulder and the included angle between the connecting line of the right knee and the right shoulder and the Y.
The amount of muscle contraction includes: rectus femoris contraction, biceps femoris contraction, vastus lateralis contraction, tibialis anterior contraction, and gastrocnemius contraction.
The muscle tendon length comprises: rectus femoris tendon length, biceps femoris tendon length, lateral femoral musculature tendon length, anterior tibial musculature tendon length, and gastrocnemius tendon length.
The first step is as follows: in the standard action database establishment, the analysis and processing of the data flow information means that: the method comprises the steps of acquiring a human body whole body skeleton three-dimensional data coordinate by adopting a Kinect skeleton tracking technology, processing data flow information of a standard action template, calculating angles of all joint points of the lower limb in each frame of image, obtaining a sequence of all joint angles, further obtaining related angle characteristic quantities, and analyzing the relationship between the length of muscle tendons of the lower limb and the muscle contraction quantity of the lower limb.
The second step is as follows: in the establishment of a human-computer interaction interface, the function selection interface is used for selecting a training project and selecting the age group and gender information of a user;
the training action guide interface is used for demonstrating standard training actions, sending out voice prompts to guide the actions of a user and simultaneously displaying real-time training images of the user;
the data analysis and evaluation interface is used for displaying the training action completion condition of the user and the standard reaching percentage of the training action;
the action reproduction comparison interface is used for comparing and displaying the reproduction image of the standard action and the reproduction image of the user training action.
The action simulation software is OpenSim human lower limb muscle simulation software.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the design method of the human body lower limb muscle training system based on the Kinect, the skeleton tracking technology of Kinect software is utilized, the three-dimensional data coordinates and the related angle characteristic quantity of the skeleton of the whole body of the human body are obtained, the action data of the lower limb can be acquired without wearing sensing equipment, meanwhile, a human-computer interaction interface is designed to realize human-computer interaction during human body lower limb muscle training, so that the training is clear, and the training effect is enhanced. Therefore, the action data of the lower limbs can be collected without wearing sensing equipment, and meanwhile, the training effect is enhanced by utilizing a human-computer interaction interface.
2. In the design method of the human body lower limb muscle training system based on the Kinect, the standard actions of different ages and genders are classified, and meanwhile, the standard action judgment standard and the lower limb muscle training effect judgment standard are set according to the genders and the ages, and an objective evaluation system is made according to the difference of the ages and the genders. Therefore, this design sets an evaluation system according to age and sex, and the reliability of the determination result is high.
3. The invention relates to a design method of a human body lower limb muscle training system based on Kinect, which is characterized in that a human body lower limb bone muscle simulation model is obtained by analyzing and processing lower limb bone data stream information collected by a Kinect somatosensory camera, collected lower limb actions are reproduced, changes of bones and muscles of lower limbs are reflected visually, the length of muscle tendons and the muscle contraction quantity of the lower limbs are calculated to evaluate a training effect, the evaluation effect of each muscle is corresponding to the corresponding muscle of the lower limb action reproduction model to generate a reproduction image of a user training action, and the reproduction image of the user training action and the reproduction image of the corresponding standard action are output to an action reproduction contrast interface to be displayed, so that differences between detailed users and standard actions can be reflected visually, and the user can observe, compare and improve the training action conveniently. Therefore, the design can intuitively reproduce the training process, the images are reproduced by contrasting and displaying the actions, and the user can observe the contrast and improve the training actions conveniently.
Drawings
Fig. 1 is a control flow chart of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the following description and embodiments in conjunction with the accompanying drawings.
A design method of a human body lower limb muscle training system based on Kinect is based on the following equipment: the method comprises the following steps: the Kinect motion sensing camera is in signal connection with the computer, and a Kinect program, motion simulation software and a format conversion program matched with the Kinect and the motion simulation software are loaded in the computer;
the lower limb muscle exercising method comprises the following steps:
the first step is as follows: establishing a standard action database, selecting the crowd with no physical disabilities and normal lower limb functions, finishing the determination of the standard human lower limb training actions according to the gender and age grouping from the anatomical perspective, classifying the determined lower limb actions according to the lower limb expansion change conditions of different lower limb training actions, establishing the standard human lower limb action database, and storing the images of the standard human lower limb training actions into the standard human lower limb action database;
acquiring lower limb skeleton data flow information of each classified standard human lower limb training action by using a Kinect somatosensory camera, analyzing and processing the data flow information, and recording relevant angle characteristics corresponding to the standard human lower limb training action; extracting lower limb muscle joint coordinates and related angle characteristics, importing the lower limb muscle joint coordinates and the related angle characteristics into action simulation software, establishing a standard action human body lower limb skeleton muscle simulation model, extracting lower limb muscle joint coordinates and related angle characteristic data, converting the lower limb muscle joint coordinates and the related angle characteristic data into a data format which can be identified by the action simulation software, and then performing inverse dynamics simulation by utilizing the action simulation software to obtain standard muscle tendon length and standard muscle contraction quantity in an action process;
corresponding the obtained quasi muscle tendon length and the standard muscle contraction quantity of each muscle to corresponding muscles of a standard action human lower limb skeletal muscle simulation model to generate a reproduced image of a standard action;
storing the standard action related angle characteristics, the standard muscle tendon length, the standard muscle contraction quantity, the standard action human lower limb skeleton simulation model and the reproduced image of the standard action into a standard human lower limb action database as corresponding standard human lower limb training action database information;
the second step is that: establishing a human-computer interaction interface, establishing a visual human-computer interaction interface, and realizing human-computer interaction during human lower limb muscle training, wherein the human-computer interaction interface comprises: the system comprises a function selection interface, a training action guide interface, a data analysis and evaluation interface and an action reproduction comparison interface;
the third step: establishing an action training judgment system, establishing a human body lower limb muscle training judgment standard, and quantifying a training effect:
establishing standard action judgment standards, and setting standard range values of the relevant angle characteristic data of each standard action according to different age groups;
when the action training judging system carries out lower limb training, analyzing and processing lower limb skeleton data flow information acquired by the Kinect somatosensory camera, comparing the processed corresponding relevant angle characteristic data with a standard action standard database of a corresponding age by using an algorithm, judging that the action reaches the standard if a value obtained after comparison is within a standard range value, and otherwise, judging that the action does not reach the standard;
when training is finished, calculating the standard reaching percentage of the training actions, and outputting the standard reaching condition of each training action and the standard reaching percentage of the training actions to a data analysis and evaluation interface;
the fourth step: establishing an action reproduction system, establishing human body lower limb action reproduction simulation, and further judging an action training effect:
establishing an action reproduction system, reproducing each collected completed action process in action simulation software, and conveniently paying attention to the action details of a trainer;
when the action reproduction system receives a group of data stream information of training actions, analyzing and processing the lower limb skeleton data stream information acquired by the Kinect somatosensory camera, extracting the coordinates and angles of the muscle joints of the lower limbs, importing the coordinates and angles into action simulation software, modifying the parameters of a human lower limb skeleton muscle simulation model to obtain a lower limb action reproduction model, and reproducing the actions finished by a trainer;
the fifth step: establishing a training effect analysis system, analyzing the training effect of the human lower limb muscles, and quantitatively analyzing the effect:
establishing a lower limb muscle training effect evaluation standard, and setting different evaluation range values according to different age groups for standard muscle tendon length and standard muscle contraction quantity data of each standard action;
establishing a training effect analysis system, extracting lower limb muscle joint coordinates and related angle characteristic data, converting the lower limb muscle joint coordinates and the related angle characteristic data into data formats which can be identified by action simulation software, performing inverse dynamics simulation by utilizing the action simulation software, calculating the force and moment required by each muscle in the lower limb movement, further calculating the length of muscle tendons of the lower limb and muscle contraction quantity data, comparing the obtained length of the muscle tendons of the lower limb and the muscle contraction quantity data with the evaluation range value of the training effect evaluation standard of the corresponding age, obtaining a training effect evaluation result, and further quantifying the evaluation of the training effect;
and corresponding the obtained evaluation effect of each muscle to the corresponding muscle of the lower limb action reproduction model obtained in the fourth step to generate a reproduction image of the training action of the user, and outputting the reproduction image of the training action of the user and the reproduction image of the corresponding standard action to an action reproduction comparison interface for displaying.
The relevant angular features include: the left knee joint angle, the right knee joint angle, the hip opening angle, the included angle between the connecting line of the left hip joint and the right hip joint and the X axis, the included angle between the left thigh and the X axis, the included angle between the right thigh and the X axis, the included angle between the connecting line of the left knee and the left shoulder and the included angle between the connecting line of the right knee and the right shoulder and the Y.
The amount of muscle contraction includes: rectus femoris contraction, biceps femoris contraction, vastus lateralis contraction, tibialis anterior contraction, and gastrocnemius contraction.
The muscle tendon length comprises: rectus femoris tendon length, biceps femoris tendon length, lateral femoral musculature tendon length, anterior tibial musculature tendon length, and gastrocnemius tendon length.
The first step is as follows: in the standard action database establishment, the analysis and processing of the data flow information means that: the method comprises the steps of acquiring a human body whole body skeleton three-dimensional data coordinate by adopting a Kinect skeleton tracking technology, processing data flow information of a standard action template, calculating angles of all joint points of the lower limb in each frame of image, obtaining a sequence of all joint angles, further obtaining related angle characteristic quantities, and analyzing the relationship between the length of muscle tendons of the lower limb and the muscle contraction quantity of the lower limb.
The second step is as follows: in the establishment of a human-computer interaction interface, the function selection interface is used for selecting a training project and selecting the age group and gender information of a user;
the training action guide interface is used for demonstrating standard training actions, sending out voice prompts to guide the actions of a user and simultaneously displaying real-time training images of the user;
the data analysis and evaluation interface is used for displaying the training action completion condition of the user and the standard reaching percentage of the training action;
the action reproduction comparison interface is used for comparing and displaying the reproduction image of the standard action and the reproduction image of the user training action.
The action simulation software is OpenSim human lower limb muscle simulation software.
The principle of the invention is illustrated as follows:
the method uses a skeleton tracking technology of a Kinect somatosensory camera to acquire non-contact human body actions, and combines a human-computer interactive interface to acquire standard action sample data and train the targeted human body lower limb actions; the Kinect equipment comprises an RGB color camera, and the function of the RGB color camera is similar to that of a common camera; an infrared emitter for emitting infrared rays; and the infrared depth camera is used for acquiring a depth image of an object for subsequent bone tracking.
The Kinect software performs characterization processing on data, taking a left leg as an example, and comprises the following specific steps: the direction perpendicular to the horizontal plane is defined as the Y axis, the direction facing the human body is defined as the Z axis, and the direction horizontal to Kinect is defined as the X axis; because the coordinate data of each joint point acquired by the Kinect in real time is greatly influenced by individual difference, the coordinate change of the joint in the motion process is converted into angle change for reducing errors of the type. First, the lower limbs are calculatedEuclidean distance from hip joint to knee joint
The distance D from the knee joint to the ankle joint and the distance D from the ankle joint to the hip joint are calculated in the same wayBC(x, y, z) and DCA(x, y, z), and calculating the included angle of the joint according to the trigonometric cosine law
Wherein A, B and C represent the coordinates of a left hip joint point, a left knee joint point and a left ankle joint point; the system mainly adopts a bone data stream acquisition mode, and defines a transformemoothparameters structure body in the Kinect SDK for smoothing in order to prevent the jumping of bone points; taking the motion angles of the hip joint, the knee joint and the ankle joint as characteristic values, establishing a template R according to an angle sequence, and averaging 60 frames, namely 2 seconds as the total length of the motion template, because the Kinect can obtain about 30 frames of data per second; assuming the action to be measured T, calculating the cumulative distance between T and each frame sequence of the action template R: d [ T, R ]]＝∑d[T，R]If D [ T, R]If the matching is less than the set threshold value, the matching is considered to be successful, otherwise, the matching is not successful.
Example lower limb training action:
1. bending knees at standing positions: the left foot and the right foot are alternately lifted to the same height of the knee and the hip, and the whole process is upright, belonging to low-intensity body-building sports. 2. Stretching the ankle at the standing position: the action lifts the left foot to the front of the body from the upright state until the ankle joint expands, belonging to the low-intensity body-building exercise. 3. Running in situ: the action requires that the upper body is upright and the feet alternately step in place, belongs to middle-strength fitness exercise, exercises all muscle groups of the lower limbs and trains the heart-lung function and the endurance of the human body. 4. Weight bearing squat method: the two feet are closed, the gravity center is placed in the palm of the front foot, the chest is contained, the abdomen is drawn, the whole body is relaxed, the high-strength body-building head can not be tilted backward or inclined, the two legs are closed all the time, the high-strength body-building head slowly ascends after being squatted thoroughly, the operation is repeated for a plurality of times, the high-strength body-building head belongs to the high-strength body-building motion, all muscle groups of the lower limbs are exercised, and the training for the explosive force of the muscle of the human body is performed.
Action training evaluation example:
the action training judging system is used for preliminarily judging the finishing condition of the action of a trainer by utilizing a program, namely judging whether the action is correct or incorrect. For a specific action, establishing a template R ═ { R (1), R (2), R (3), R (4) }, R (n) as a static frame sequence, and assuming an action T to be measured, calculating a cumulative distance between T and each frame sequence of the action template R: d [ T, R ] ═ Σ D [ T, R ], assign the template label of the minimum cumulative distance value to the movement T to be measured, if the label is the same as the measured movement name, consider matching successful, otherwise unsuccessful; if D [ T, R ] is less than the set action threshold value corresponding to the corresponding age of the action, the matching is considered to be successful, otherwise, the matching is unsuccessful;
the motion reappearing system and the training effect analysis system are used for specifically researching the motion change of the lower limbs and the muscle force applying condition of the lower limbs, a motion simulation software OpenSim is introduced to establish a human body three-dimensional skeleton muscle model, and the muscle strength of the human body lower limb exercise is obtained by utilizing the inverse dynamics and residual error reduction of the OpenSim and a muscle calculation control tool;
the OpenSim software is an open source software platform developed by Stanford University, the theory of the OpenSim software is derived from a Hill equation and a Hill muscle tripartite model, the motion data of a tester is introduced, and then reverse dynamics simulation and forward dynamics simulation are carried out to analyze the characteristics of lower limb muscles, wherein an xia value muscle model, Gait2392_ Simbody.osim, is adopted;
according to the method, the Kinect and a human-computer interaction interface are used for performing correct and wrong matching of actions according to the action condition of a trainer, and the actions and muscle contraction conditions of the trainer are reproduced through OpenSim simulation, so that the training is more targeted, and a more ideal training effect is achieved; the training method comprises the following steps: 1. the patient stands on the front side and is within the recognizable range of the Kinect whole body, and an example action video is played; 2. the subject can start to simulate the example video to make a set of complete actions, and the actions are stored in the computer through the USB data line for data storage, and the actions are judged correctly and incorrectly by adopting a DTW algorithm; 3. in order to objectively evaluate the action of the lower limbs of the human body, OpenSim software is introduced, and muscle stretching simulation is performed on the motion data acquired by Kinect. Simulation experiment results show that the human motion model established in the method can perform real-time muscle simulation reappearance on the actions of the lower limbs of the human body and evaluate training results.
When the body finishes actions similar to lower limbs such as squatting and rising, the main strength source of the body provides effective support for the moment of the hip joint, and a series of powerful muscles such as iliocortical muscles, gluteus maximus, rectus femoris, biceps femoris, semitendinosus and semimembranosus around the hip joint.
In addition, the hip joint contribution degree tends to decrease with increasing age, the knee joint contribution degree gradually increases with increasing age, and the hip joint and knee joint contribution degrees in the centrifugal stage between the age group of 20 to 29 years, the age group of 50 to 59 years and the age group of 60 to 69 years are significantly different. According to research, when a human body is trained on lower limbs, the contribution degrees of lower limb joints of all age groups (29-69) are hip joints > knee joints > ankle joints in the order, particularly in the centrifugal stage in the squatting process, the contribution degree of the hip joints is reduced along with the increase of the age, the contribution degree of the knee joints is increased along with the increase of the age, the contribution degree of the ankle joints is increased along with the increase of the age in all stages of the age groups of 20-29 and 30-39, and the contribution degree of the ankle joints is reduced along with the increase of the age in all age groups of 40-69. Investigation studies simultaneously find that the discharge time sequence of the muscles of the lower limbs of the age groups of 20-39 years and 30-39 years is tibialis anterior → rectus femoris → biceps femoris → gastrocnemius lateral head. The discharge timing of each muscle of the lower limbs of the age group of 40-59 years is tibialis anterior → biceps femoris → rectus femoris → gastrocnemius lateral head, which proves that the stimulation of the muscles is brought by the lower limb training. Meanwhile, statistical results show that the percentage of exercise time of the left rectus femoris and the right biceps femoris of the two age groups of 20-29 years old and 70-79 years old is significantly different, and further prove that the muscle strength of the old is lost and the necessity of setting standards in different age groups is met.
In the fourth step: and an application programming interface of OpenSim is connected with Matlab, the coordinates and angles of the muscle joints of the lower limbs are imported into OpenSim by utilizing the Matlab, and parameters of a human lower limb skeletal muscle simulation model are modified by the OpenSim to obtain a lower limb action reproduction model.
The human body lower limb skeletal muscle simulation model adopts motion simulation software OpenSim to establish a human body three-dimensional skeletal muscle model, and utilizes inverse dynamics and residual error reduction of OpenSim and a muscle calculation control tool to obtain muscle strength for human body lower limb exercise. Firstly, three-dimensional coordinate data of human body joint points acquired by Kinect are reproduced in Matlab, then an application programming interface of OpenSim is connected with Matlab, so that bone data in the Matlab can be transmitted to OpenSim, and finally, personalized muscle and bone model scaling is carried out on a lower limb training object through OpenSim software according to the bone point data, so that a human body lower limb bone and muscle simulation model is established.
In the fifth step: and (3) converting kinematic and kinetic data into a data format identified by OpenSim through Visual 3D software by utilizing the lower limb joint coordinates and angles extracted by Kinect to perform inverse dynamic simulation. The sto file is generated under a working directory after the simulation is finished, the file respectively contains dynamic (joint torque) data of three joints of the hip, the knee and the ankle of the lower limb of the human body, and the joint torque is calculated by a Residual Reduction Algorithm (RRA);
and (4) carrying out muscle force calculation through an OpenSim model so as to analyze the change condition of muscle force around the joint of the lower limb in the motion. Modifying internal parameters of the original model by using Notepad + + text editor software to enable the new model, namely the object 01.osim, to have the mass, the height size and the strength of a human body representing an experimental participant; the degree of freedom and the muscle geometry are suitable for the muscle groups of the study, namely rectus femoris, biceps femoris, vastus lateralis, tibialis anterior and gastrocnemius; the motion curve is smoothed using Inverse Kinematics (IK) and using a Residual Reduction Algorithm (RRA).
And the human motion model established by the simulation experiment result performs real-time muscle simulation reappearance on the action of the lower limbs of the human body, compares the muscle contraction with the muscle contraction amount in a standard action library, and performs objective evaluation on the subsequent training result. And (3) creating a three-dimensional lower limb skeletal muscle model by utilizing OpenSim software through real-time transmitted position and angle information of joint points of a training action human body for calculating muscle contraction change data under action. And comparing the data with muscle contraction change data in a standard human body lower limb action database by using a variance method, wherein the variance values in different ranges are divided into three classes, namely excellent action effect and the like, so as to further evaluate the training effect.
Example 1:
a design method of a human body lower limb muscle training system based on Kinect is based on the following equipment: the method comprises the following steps: the Kinect motion sensing camera is in signal connection with the computer, and a Kinect program, motion simulation software and a format conversion program matched with the Kinect and the motion simulation software are loaded in the computer;
the lower limb muscle exercising method comprises the following steps:
the first step is as follows: establishing a standard action database, selecting the crowd with no physical disabilities and normal lower limb functions, finishing the determination of the standard human lower limb training actions according to the gender and age grouping from the anatomical perspective, classifying the determined lower limb actions according to the lower limb expansion change conditions of different lower limb training actions, establishing the standard human lower limb action database, and storing the images of the standard human lower limb training actions into the standard human lower limb action database;
acquiring lower limb skeleton data flow information of each classified standard human lower limb training action by using a Kinect somatosensory camera, analyzing and processing the data flow information, and recording relevant angle characteristics corresponding to the standard human lower limb training action; extracting lower limb muscle joint coordinates and related angle characteristics, importing the lower limb muscle joint coordinates and the related angle characteristics into action simulation software, establishing a standard action human body lower limb skeleton muscle simulation model, extracting lower limb muscle joint coordinates and related angle characteristic data, converting the lower limb muscle joint coordinates and the related angle characteristic data into a data format which can be identified by the action simulation software, and then performing inverse dynamics simulation by utilizing the action simulation software to obtain standard muscle tendon length and standard muscle contraction quantity in an action process;
corresponding the obtained quasi muscle tendon length and the standard muscle contraction quantity of each muscle to corresponding muscles of a standard action human lower limb skeletal muscle simulation model to generate a reproduced image of a standard action;
storing the standard action related angle characteristics, the standard muscle tendon length, the standard muscle contraction quantity, the standard action human lower limb skeleton simulation model and the reproduced image of the standard action into a standard human lower limb action database as corresponding standard human lower limb training action database information;
the second step is that: establishing a human-computer interaction interface, establishing a visual human-computer interaction interface, and realizing human-computer interaction during human lower limb muscle training, wherein the human-computer interaction interface comprises: the system comprises a function selection interface, a training action guide interface, a data analysis and evaluation interface and an action reproduction comparison interface;
the third step: establishing an action training judgment system, establishing a human body lower limb muscle training judgment standard, and quantifying a training effect:
establishing standard action judgment standards, and setting standard range values of the relevant angle characteristic data of each standard action according to different age groups;
when the action training judging system carries out lower limb training, analyzing and processing lower limb skeleton data flow information acquired by the Kinect somatosensory camera, comparing the processed corresponding relevant angle characteristic data with a standard action standard database of a corresponding age by using an algorithm, judging that the action reaches the standard if a value obtained after comparison is within a standard range value, and otherwise, judging that the action does not reach the standard;
when training is finished, calculating the standard reaching percentage of the training actions, and outputting the standard reaching condition of each training action and the standard reaching percentage of the training actions to a data analysis and evaluation interface;
the fourth step: establishing an action reproduction system, establishing human body lower limb action reproduction simulation, and further judging an action training effect:
establishing an action reproduction system, reproducing each collected completed action process in action simulation software, and conveniently paying attention to the action details of a trainer;
when the action reproduction system receives a group of data stream information of training actions, analyzing and processing the lower limb skeleton data stream information acquired by the Kinect somatosensory camera, extracting the coordinates and angles of the muscle joints of the lower limbs, importing the coordinates and angles into action simulation software, modifying the parameters of a human lower limb skeleton muscle simulation model to obtain a lower limb action reproduction model, and reproducing the actions finished by a trainer;
the fifth step: establishing a training effect analysis system, analyzing the training effect of the human lower limb muscles, and quantitatively analyzing the effect:
establishing a lower limb muscle training effect evaluation standard, and setting different evaluation range values according to different age groups for standard muscle tendon length and standard muscle contraction quantity data of each standard action;
establishing a training effect analysis system, extracting lower limb muscle joint coordinates and related angle characteristic data, converting the lower limb muscle joint coordinates and the related angle characteristic data into data formats which can be identified by action simulation software, performing inverse dynamics simulation by utilizing the action simulation software, calculating the force and moment required by each muscle in the lower limb movement, further calculating the length of muscle tendons of the lower limb and muscle contraction quantity data, comparing the obtained length of the muscle tendons of the lower limb and the muscle contraction quantity data with the evaluation range value of the training effect evaluation standard of the corresponding age, obtaining a training effect evaluation result, and further quantifying the evaluation of the training effect;
and corresponding the obtained evaluation effect of each muscle to the corresponding muscle of the lower limb action reproduction model obtained in the fourth step to generate a reproduction image of the training action of the user, and outputting the reproduction image of the training action of the user and the reproduction image of the corresponding standard action to an action reproduction comparison interface for displaying.
The relevant angular features include: the left knee joint angle, the right knee joint angle, the hip opening angle, the included angle between the connecting line of the left hip joint and the right hip joint and the X axis, the included angle between the left thigh and the X axis, the included angle between the right thigh and the X axis, the included angle between the connecting line of the left knee and the left shoulder and the included angle between the connecting line of the right knee and the right shoulder and the Y.
The amount of muscle contraction includes: rectus femoris contraction, biceps femoris contraction, vastus lateralis contraction, tibialis anterior contraction, and gastrocnemius contraction.
The muscle tendon length comprises: rectus femoris tendon length, biceps femoris tendon length, lateral femoral musculature tendon length, anterior tibial musculature tendon length, and gastrocnemius tendon length.
Example 2:
example 2 is substantially the same as example 1 except that:
the first step is as follows: in the standard action database establishment, the analysis and processing of the data flow information means that: the method comprises the steps of acquiring a human body whole body skeleton three-dimensional data coordinate by adopting a Kinect skeleton tracking technology, processing data flow information of a standard action template, calculating angles of all joint points of the lower limb in each frame of image, obtaining a sequence of all joint angles, further obtaining related angle characteristic quantities, and analyzing the relationship between the length of muscle tendons of the lower limb and the muscle contraction quantity of the lower limb.
The action simulation software is OpenSim human lower limb muscle simulation software.
Example 3:
example 3 is substantially the same as example 2 except that:
the second step is as follows: in the establishment of a human-computer interaction interface, the function selection interface is used for selecting a training project and selecting the age group and gender information of a user;
the training action guide interface is used for demonstrating standard training actions, sending out voice prompts to guide the actions of a user and simultaneously displaying real-time training images of the user;
the data analysis and evaluation interface is used for displaying the training action completion condition of the user and the standard reaching percentage of the training action;
the action reproduction comparison interface is used for comparing and displaying the reproduction image of the standard action and the reproduction image of the user training action.
Claims (5)
1. A design method of a human body lower limb muscle training system based on Kinect is characterized by comprising the following steps:
the method for designing the lower limb muscle training system and judging the effect is based on the following equipment: the method comprises the following steps: the Kinect motion sensing camera is in signal connection with the computer, and a Kinect program, motion simulation software and a format conversion program matched with the Kinect and the motion simulation software are loaded in the computer;
the lower limb muscle exercising method comprises the following steps:
the first step is as follows: establishing a standard action database, selecting the crowd with no physical disabilities and normal lower limb functions, finishing the determination of the standard human lower limb training actions according to the gender and age grouping from the anatomical perspective, classifying the determined lower limb actions according to the lower limb expansion change conditions of different lower limb training actions, establishing the standard human lower limb action database, and storing the images of the standard human lower limb training actions into the standard human lower limb action database;
acquiring lower limb skeleton data flow information of each classified standard human lower limb training action by using a Kinect somatosensory camera, analyzing and processing the data flow information, and recording relevant angle characteristics corresponding to the standard human lower limb training action; extracting lower limb muscle joint coordinates and related angle characteristics, importing the lower limb muscle joint coordinates and the related angle characteristics into action simulation software, establishing a standard action human body lower limb skeleton muscle simulation model, extracting lower limb muscle joint coordinates and related angle characteristic data, converting the lower limb muscle joint coordinates and the related angle characteristic data into a data format which can be identified by the action simulation software, and then performing inverse dynamics simulation by utilizing the action simulation software to obtain standard muscle tendon length and standard muscle contraction quantity in an action process;
corresponding the obtained quasi muscle tendon length and the standard muscle contraction quantity of each muscle to corresponding muscles of a standard action human lower limb skeletal muscle simulation model to generate a reproduced image of a standard action;
storing the standard action related angle characteristics, the standard muscle tendon length, the standard muscle contraction quantity, the standard action human lower limb skeleton simulation model and the reproduced image of the standard action into a standard human lower limb action database as corresponding standard human lower limb training action database information;
the second step is that: establishing a human-computer interaction interface, establishing a visual human-computer interaction interface, and realizing human-computer interaction during human lower limb muscle training, wherein the human-computer interaction interface comprises: the system comprises a function selection interface, a training action guide interface, a data analysis and evaluation interface and an action reproduction comparison interface;
the third step: establishing an action training judgment system, establishing a human body lower limb muscle training judgment standard, and quantifying a training effect:
establishing standard action judgment standards, and setting standard range values of the relevant angle characteristic data of each standard action according to different age groups;
when the action training judging system carries out lower limb training, analyzing and processing lower limb skeleton data flow information acquired by the Kinect somatosensory camera, comparing the processed corresponding relevant angle characteristic data with a standard action standard database of a corresponding age by using an algorithm, judging that the action reaches the standard if a value obtained after comparison is within a standard range value, and otherwise, judging that the action does not reach the standard;
when training is finished, calculating the standard reaching percentage of the training actions, and outputting the standard reaching condition of each training action and the standard reaching percentage of the training actions to a data analysis and evaluation interface;
the fourth step: establishing an action reproduction system, establishing human body lower limb action reproduction simulation, and further judging an action training effect:
establishing an action reproduction system, reproducing each collected completed action process in action simulation software, and conveniently paying attention to the action details of a trainer;
when the action reproduction system receives a group of data stream information of training actions, analyzing and processing the lower limb skeleton data stream information acquired by the Kinect somatosensory camera, extracting the coordinates and angles of the muscle joints of the lower limbs, importing the coordinates and angles into action simulation software, modifying the parameters of a human lower limb skeleton muscle simulation model to obtain a lower limb action reproduction model, and reproducing the actions finished by a trainer;
the fifth step: establishing a training effect analysis system, analyzing the training effect of the human lower limb muscles, and quantitatively analyzing the effect:
establishing a lower limb muscle training effect evaluation standard, and setting different evaluation range values according to different age groups for standard muscle tendon length and standard muscle contraction quantity data of each standard action;
establishing a training effect analysis system, extracting lower limb muscle joint coordinates and related angle characteristic data, converting the lower limb muscle joint coordinates and the related angle characteristic data into data formats which can be identified by action simulation software, performing inverse dynamics simulation by utilizing the action simulation software, calculating the force and moment required by each muscle in the lower limb movement, further calculating the length of muscle tendons of the lower limb and muscle contraction quantity data, comparing the obtained length of the muscle tendons of the lower limb and the muscle contraction quantity data with the evaluation range value of the training effect evaluation standard of the corresponding age, obtaining a training effect evaluation result, and further quantifying the evaluation of the training effect;
and corresponding the obtained evaluation effect of each muscle to the corresponding muscle of the lower limb action reproduction model obtained in the fourth step to generate a reproduction image of the training action of the user, and outputting the reproduction image of the training action of the user and the reproduction image of the corresponding standard action to an action reproduction comparison interface for displaying.
2. The design method of human body lower limb muscle training system based on Kinect as claimed in claim 1, wherein:
the relevant angular features include: the left knee joint angle, the right knee joint angle, the hip opening angle, the included angle between the connecting line of the left hip joint and the right hip joint and the X axis, the included angle between the left thigh and the X axis, the included angle between the right thigh and the X axis, the included angle between the connecting line of the left knee and the left shoulder and the included angle between the connecting line of the right knee and the right shoulder and the Y.
The amount of muscle contraction includes: rectus femoris contraction, biceps femoris contraction, vastus lateralis contraction, tibialis anterior contraction, and gastrocnemius contraction.
The muscle tendon length comprises: rectus femoris tendon length, biceps femoris tendon length, lateral femoral musculature tendon length, anterior tibial musculature tendon length, and gastrocnemius tendon length.
3. The design method of human body lower limb muscle training system based on Kinect as claimed in claim 2, wherein:
the first step is as follows: in the standard action database establishment, the analysis and processing of the data flow information means that: the method comprises the steps of acquiring a human body whole body skeleton three-dimensional data coordinate by adopting a Kinect skeleton tracking technology, processing data flow information of a standard action template, calculating angles of all joint points of the lower limb in each frame of image, obtaining a sequence of all joint angles, further obtaining related angle characteristic quantities, and analyzing the relationship between the length of muscle tendons of the lower limb and the muscle contraction quantity of the lower limb.
4. The design method of human lower limb muscle training system based on Kinect as claimed in claim 1, 2 or 3, wherein:
the second step is as follows: in the establishment of a human-computer interaction interface, the function selection interface is used for selecting a training project and selecting the age group and gender information of a user;
the training action guide interface is used for demonstrating standard training actions, sending out voice prompts to guide the actions of a user and simultaneously displaying real-time training images of the user;
the data analysis and evaluation interface is used for displaying the training action completion condition of the user and the standard reaching percentage of the training action;
the action reproduction comparison interface is used for comparing and displaying the reproduction image of the standard action and the reproduction image of the user training action.
5. The design method of human body lower limb muscle training system based on Kinect as claimed in claim 4, wherein:
the action simulation software is OpenSim human lower limb muscle simulation software.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110177730.9A CN112966370B (en) | 2021-02-09 | 2021-02-09 | Design method of human body lower limb muscle training system based on Kinect |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110177730.9A CN112966370B (en) | 2021-02-09 | 2021-02-09 | Design method of human body lower limb muscle training system based on Kinect |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112966370A true CN112966370A (en) | 2021-06-15 |
| CN112966370B CN112966370B (en) | 2022-04-19 |
Family
ID=76284499
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110177730.9A Active CN112966370B (en) | 2021-02-09 | 2021-02-09 | Design method of human body lower limb muscle training system based on Kinect |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112966370B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113506145A (en) * | 2021-09-08 | 2021-10-15 | 南京麦豆健康管理有限公司 | Postpartum exercise course evaluation system and method based on data analysis |
| CN114724681A (en) * | 2022-04-24 | 2022-07-08 | 福州大学 | Cervical vertebra rehabilitation training biomechanical simulation analysis method based on Opensim |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104524742A (en) * | 2015-01-05 | 2015-04-22 | 河海大学常州校区 | Cerebral palsy child rehabilitation training method based on Kinect sensor |
| WO2018120964A1 (en) * | 2016-12-30 | 2018-07-05 | 山东大学 | Posture correction method based on depth information and skeleton information |
| CN108785997A (en) * | 2018-05-30 | 2018-11-13 | 燕山大学 | A kind of lower limb rehabilitation robot Shared control method based on change admittance |
| CN109223441A (en) * | 2018-09-13 | 2019-01-18 | 华南理工大学 | A kind of human upper limb rehabilitation training and movement auxiliary system based on Kinect sensor |
| CN109331453A (en) * | 2018-08-07 | 2019-02-15 | 燕山大学 | The virtual rehabilitation system and training method interacted based on EMG feedback with Kinect |
| CN111437583A (en) * | 2020-04-10 | 2020-07-24 | 哈尔滨工业大学 | Badminton basic action auxiliary training system based on Kinect |
-
2021
- 2021-02-09 CN CN202110177730.9A patent/CN112966370B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104524742A (en) * | 2015-01-05 | 2015-04-22 | 河海大学常州校区 | Cerebral palsy child rehabilitation training method based on Kinect sensor |
| WO2018120964A1 (en) * | 2016-12-30 | 2018-07-05 | 山东大学 | Posture correction method based on depth information and skeleton information |
| CN108785997A (en) * | 2018-05-30 | 2018-11-13 | 燕山大学 | A kind of lower limb rehabilitation robot Shared control method based on change admittance |
| CN109331453A (en) * | 2018-08-07 | 2019-02-15 | 燕山大学 | The virtual rehabilitation system and training method interacted based on EMG feedback with Kinect |
| CN109223441A (en) * | 2018-09-13 | 2019-01-18 | 华南理工大学 | A kind of human upper limb rehabilitation training and movement auxiliary system based on Kinect sensor |
| CN111437583A (en) * | 2020-04-10 | 2020-07-24 | 哈尔滨工业大学 | Badminton basic action auxiliary training system based on Kinect |
Non-Patent Citations (3)
| Title |
|---|
| FREIVOGEL S 等: "Improved Walking Ability and Reduced Therapeutic Stress with An Electromechanical Gait Device", 《JOURNAL OF REHABILITATION》 * |
| 朱丽等: "老年人动作识别系统研究", 《计算机工程与应用》 * |
| 李德春: "基于Kinect下肢康复训练机器人人体步态分析", 《中国优秀硕士学位论文全文数据库（信息科技辑）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113506145A (en) * | 2021-09-08 | 2021-10-15 | 南京麦豆健康管理有限公司 | Postpartum exercise course evaluation system and method based on data analysis |
| CN114724681A (en) * | 2022-04-24 | 2022-07-08 | 福州大学 | Cervical vertebra rehabilitation training biomechanical simulation analysis method based on Opensim |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112966370B (en) | 2022-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6703199B2 (en) | Operating state evaluation system, operating state evaluation device, operating state evaluation server, operating state evaluation method, and operating state evaluation program | |
| CN112966370B (en) | Design method of human body lower limb muscle training system based on Kinect | |
| CN109919034A (en) | A kind of identification of limb action with correct auxiliary training system and method | |
| Zago et al. | Multi-segmental movements as a function of experience in karate | |
| CN109621331A (en) | Fitness-assisting method, apparatus and storage medium, server | |
| Willmann et al. | Home stroke rehabilitation for the upper limbs | |
| KR102013705B1 (en) | Apparatus and method for recognizing user's posture in horse-riding simulator | |
| KR102259011B1 (en) | user interface service system and method for user motion correction direction | |
| CN114093032A (en) | Human body action evaluation method based on action state information | |
| Yang et al. | Hand rehabilitation using virtual reality electromyography signals | |
| CN116271757A (en) | Auxiliary system and method for basketball practice based on AI technology | |
| Hirasawa et al. | Promotion system for home-based squat training using openpose | |
| CN115985462A (en) | Rehabilitation and intelligence-developing training system for children cerebral palsy | |
| Cynarski et al. | Improving the movements of basic karate techniques with the use of motion capture and mathematical modeling. The concept of a research project | |
| CN108256403B (en) | Device and method for digitizing limb movement characteristics | |
| Lianzhen et al. | Athlete Rehabilitation Evaluation System Based on Internet of Health Things and Human Gait Analysis Algorithm | |
| TWI693925B (en) | Device and method of quantifying characteristics of body and limb motions | |
| Lisboa et al. | A study for postural evaluation and movement analysis of individuals | |
| CN114973066A (en) | Online and offline fitness interaction method and system | |
| JP6978811B1 (en) | Training management system, training management method and training management program | |
| CN117672456A (en) | Lightweight visual posture assessment and non-contact emotion guiding rehabilitation system | |
| Kanungo et al. | Recognizing Physiotherapy Exercises using Machine Learning | |
| CN114360060B (en) | Human body action recognition and counting method | |
| CN109621335B (en) | Sports suggestion system based on big data | |
| TWI784243B (en) | Method of taekwondo poomsae movement detection and comparison |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |