CN116680963A - Method, system and equipment for constructing human plantar force based on computer vision - Google Patents
Method, system and equipment for constructing human plantar force based on computer vision Download PDFInfo
- Publication number
- CN116680963A CN116680963A CN202310959679.6A CN202310959679A CN116680963A CN 116680963 A CN116680963 A CN 116680963A CN 202310959679 A CN202310959679 A CN 202310959679A CN 116680963 A CN116680963 A CN 116680963A
- Authority
- CN
- China
- Prior art keywords
- human body
- segment
- human
- plantar force
- computer vision
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 75
- 230000033001 locomotion Effects 0.000 claims abstract description 92
- 230000001133 acceleration Effects 0.000 claims abstract description 48
- 238000006073 displacement reaction Methods 0.000 claims abstract description 47
- 230000008569 process Effects 0.000 claims abstract description 24
- 238000009432 framing Methods 0.000 claims abstract description 16
- 238000010276 construction Methods 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 13
- 239000003550 marker Substances 0.000 claims description 10
- 230000003068 static effect Effects 0.000 claims description 10
- 230000001020 rhythmical effect Effects 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 4
- 230000001413 cellular effect Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 description 8
- 238000005259 measurement Methods 0.000 description 7
- 230000009471 action Effects 0.000 description 3
- 230000037396 body weight Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000003595 spectral effect Effects 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 3
- 230000033764 rhythmic process Effects 0.000 description 2
- 210000000689 upper leg Anatomy 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 239000002390 adhesive tape Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 244000309466 calf Species 0.000 description 1
- 238000013016 damping Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
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
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Abstract
The application relates to the biomechanics field, and provides a method, a system and a device for constructing artificial plantar force based on computer vision, which are based on a human body multi-rigid body model, divide the human body multi-rigid body model into a plurality of segments and acquire the mass ratio of each segment in the human body multi-rigid body model; establishing a quantitative relation between plantar force and human body movement; recording the motion process of the human body by using the camera equipment to obtain a motion video; carrying out framing treatment on the motion video, and acquiring displacement data of each segment of the human body, which corresponds to each segment in the human body multi-rigid body model one by one, based on a computer vision method; based on displacement data, acquiring the acceleration of each segment of the human body by using a finite difference method; based on the acceleration and the established quantitative relation, the artificial plantar force is calculated. The application can record the motion process of the human body by utilizing universal camera equipment such as a mobile phone or a camera, and further obtain accurate human foot sole force, has low cost and high precision, and can realize large-scale acquisition.
Description
Technical Field
The application belongs to the technical field of biomechanics, and particularly relates to a method, a system and equipment for constructing artificial plantar force based on computer vision.
Background
More and more public buildings such as stadiums, music stands and the like adopt large-span structures, and the structure is characterized by light weight, low frequency and small damping, and is sensitive to artificial load. The artificial load refers to the dynamic action of the plantar force applied to the supporting structure when a user of the building performs actions such as walking, jumping, running, dancing, bouncing, suddenly standing/sitting, going up and down stairs and the like, and the dynamic action is easy to cause the vibration of a building floor, a pedestrian bridge, a long cantilever structure, a stadium stand, a flexible stair and other large-span soft engineering structures, and can cause the problem of comfort and even safety of the structure when serious. In the civil construction field, anthropogenic plantar force is a basic input parameter for engineering structure anthropogenic vibration analysis. In the field of rehabilitation medicine, human health conditions can be diagnosed by comparing the morphological differences of the actually measured plantar force and the standard plantar force (healthy human body). Therefore, it is important to obtain accurate plantar force. At present, the traditional method mainly obtains the plantar force by directly measuring special equipment such as a force measuring table (plate) or a force measuring insole and the like. But such direct measurement methods are expensive in equipment, costly to test, and disadvantageous for large sample collection.
Disclosure of Invention
The application aims to overcome the defects of the prior art, and provides a human plantar force construction method, a system and equipment based on computer vision, which are used for acquiring a human motion track through a computer vision method, acquiring human acceleration based on a finite difference method, constructing plantar force based on a quantitative relation between plantar force and human motion, and realizing large-scale acquisition, wherein the test cost is low, the precision is high.
In a first aspect, the present application provides a computer vision-based artificial plantar force construction method, comprising the steps of:
dividing the human body multi-rigid body model into a plurality of segments based on the human body multi-rigid body model, and acquiring the mass ratio of each segment in the human body multi-rigid body model; establishing a quantitative relation between plantar force and human body movement:
wherein:is the human plantar force; />The quality corresponds to each segment of the human body; />The mass center acceleration of each segment of the human body; />Is the static weight of the human body; n is the total number of segments in the human body multi-rigid body model; i is the segment number, t is the time; the mass of each segment of the human body is +.>The total mass of the human body is combined with the mass ratio of each segment in the human body multi-rigid body model to determine;
recording the motion process of the human body by using the camera equipment to obtain a motion video;
carrying out framing treatment on the motion video, and acquiring displacement data of each segment of the human body, which corresponds to each segment in the human body multi-rigid body model one by one, based on a computer vision method;
based on the displacement data, acquiring the acceleration of each segment of the human body by using a finite difference method;
based on the acceleration and the established quantitative relation, a human plantar force is calculated.
Further, the step of recording the motion process of the human body by using the image pickup device to obtain the motion video includes: the method comprises the steps of setting mark points at positions of corresponding centroids of all segments of a human body, which correspond to all segments in a human body multi-rigid-body model one by one; the human body performs rhythmic flexion and extension movements, the image pickup device is used for shooting the movement process of the human body, and in the shooting process, the shooting picture is ensured to cover all the mark points so as to obtain a movement video for recording the displacement of all the mark points.
Further, the duration of the motion video is 20s-30s.
Further, the step of framing the motion video and acquiring displacement data of each segment of the human body corresponding to each segment in the human body multi-rigid body model one by one based on a computer vision method comprises the following steps:
carrying out framing treatment on the motion video; extracting the characteristic of the mark point of each frame of photo after framing to obtain the position coordinate of the pixel space of the mark point of each frame of photo, and converting the position coordinate into the position coordinate of the real space based on coordinate transformation; and extracting the characteristic of the mark points and converting coordinates of all the photos to obtain displacement data of the mark points of each segment of the human body in a real space.
Further, the step of acquiring the acceleration of each segment of the human body by using the finite difference method based on the displacement data comprises the following steps:
obtaining a displacement time course curve of the mark points of each segment of the human body based on the displacement data of the mark points of each segment of the human body in the real space; based on the displacement time course curve of the mark points of each segment of the human body, the speed time course curve and the acceleration time course curve of the mark points of each segment of the human body are obtained by adopting two finite differences.
Further, the image pickup apparatus includes a cellular phone or a camera.
In a second aspect, the present application also proposes a computer vision-based artificial plantar force construction system, comprising:
the model dividing module is used for dividing the human body multi-rigid body model into a plurality of segments and acquiring the mass ratio of each segment in the human body multi-rigid body model;
the image pickup device is used for recording the motion process of a human body so as to obtain a motion video;
the first processing module is used for carrying out framing processing on the motion video and acquiring displacement data of each segment of the human body based on computer vision;
the second processing module is used for acquiring the acceleration of each segment of the human body by utilizing a finite difference method based on the displacement data; and
the third processing module is used for calculating the artificial plantar force based on the acceleration and the established quantitative relation between plantar force and human body movement;
wherein the quantitative relation is:
wherein:is the human plantar force; />The quality corresponds to each segment of the human body; />The mass center acceleration of each segment of the human body; />Is the static weight of the human body; n is the total number of segments in the human body multi-rigid body model; i is the segment number, t is the time; the mass of each segment of the human body is +.>The total mass of the human body is combined with the mass ratio of each segment in the human body multi-rigid body model.
Further, the human body multi-rigid body model also comprises a plurality of mark points which are arranged at the positions of the corresponding centroids of the human body segments corresponding to the segments in the human body multi-rigid body model one by one.
Further, the image pickup apparatus includes a cellular phone or a camera.
In a third aspect, the present application also proposes a computer vision-based artificial plantar force construction apparatus, comprising: one or more processors; and a storage device for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the computer vision-based artificial plantar force construction method according to any one of the above-described aspects.
The beneficial effects of the application include: the human body movement process is recorded through the camera equipment, the displacement data of the human body is obtained based on a computer vision method, the acceleration of the human body is obtained based on a finite difference method, and the artificial plantar force is calculated based on the established quantitative relation between plantar force and human body movement. The motion process of the human body can be recorded by using universal camera equipment such as a mobile phone or a camera, so that accurate human foot sole force is obtained, the cost is low, the precision is high, and large-scale collection can be realized.
Drawings
FIG. 1a shows the time course and spectral curves of the measured normalized plantar force at different motor frequencies.
Fig. 1b shows the time course and spectrum of the measured acceleration of the human body at different movement frequencies.
Fig. 2 is a graph showing the correlation coefficient between plantar force and human acceleration, for example, the flexion and extension rhythm.
FIG. 3 is a flow chart of a method for constructing artificial plantar force based on computer vision according to the present application.
Fig. 4 is a schematic diagram of a human body multi-rigid body model and centroid distribution of each segment thereof according to the present application.
Figure 5a shows the measured plantar force at a frequency of 1.5Hz and the displacement time course of the landmark points of each segment of the human body.
Fig. 5b shows the measured plantar force at a frequency of 2.5Hz and the displacement time course of the landmark points of each segment of the human body.
Fig. 5c shows the measured plantar force at a frequency of 3.5Hz and the displacement time course of the landmark points of each segment of the human body.
Fig. 6a is a graph of the displacement time course of the marker point numbered 1 when the human body performs bouncing motion at the frequency of 2.5 Hz.
Fig. 6b is a velocity profile for a marker point numbered 1 obtained using the finite difference method.
Fig. 6c is an acceleration time course curve of marker point number 1 obtained by finite difference method.
FIG. 7a shows a graph of plantar force constructed by the method of the present application versus measured plantar force at a frequency of 1.5 Hz.
FIG. 7b shows a graph of plantar force constructed by the method of the present application versus measured plantar force at a frequency of 2.5 Hz.
FIG. 7c shows a graph of plantar force constructed by the method of the present application versus measured plantar force at a frequency of 3.5 Hz.
Detailed Description
The application is described in further detail below with reference to the drawings and specific examples.
First, the theoretical basis and technical support for the feasibility of the application are described:
taking the rhythmic flexion-extension movements (bounces) of the human body as an example, fig. 1a shows the time course and spectral curves of the measured normalized plantar forces (plantar forces/static body weights) at different movement frequencies (2 Hz, 2.4Hz and 3 Hz). Fig. 1b shows the time course and spectral curves of the measured acceleration of the human body (back waist) at different movement frequencies (2 Hz, 2.4Hz and 3 Hz). Analysis of plantar force and human acceleration under flexion-extension rhythms shows that both of the plantar force and the human acceleration show typical periodic-like characteristics in time course, and the frequency spectrum distribution is very similar to that of the plantar force and the human acceleration, as shown in fig. 1a and 1 b.
The correlation coefficient between plantar force and human acceleration is calculated, and the calculation result is shown in fig. 2. The definition of the correlation coefficient is shown in formula (1).
Formula (1)
Wherein:is the covariance between variables X and Y, < ->The standard deviation of these two variables is respectively.
Fig. 2 shows that the plantar force and the acceleration of the human body all show good correlation (the correlation coefficient is close to 1 and is shown in fig. 2) at a plurality of movement frequencies. The analysis results of the plantar force and the human body movement data under other movement forms also show similar rules, namely, obvious correlation between the plantar force of the human body and the movement of the mass center of the human body is shown.
Based on the analysis results, it can be seen that: the plantar force and the human body movement acceleration show good correlation, and not only are time-course curve shapes similar, but also have similar frequency spectrum characteristics. The high correlation between plantar force and acceleration of the body demonstrates that plantar force changes are caused by oscillations of the body's centroid. We will further demonstrate in the subsequent implementation steps: the inertial effect of the moving mass of the human body is a major source of plantar force. Therefore, once the quantitative relationship between plantar force and human motion is determined, plantar force measurements can be converted into human motion measurements.
Based on the theoretical basis and technical support, the embodiment of the application provides a human plantar force construction method based on computer vision. The core idea is as follows: based on the intrinsic law that the human plantar force is derived from the inertial effect of a moving human body, the direct force measurement is converted into the measurement of human body movement (acceleration). The method comprises the steps of using a human body multi-rigid body model provided by ISO to lay mark points on main sections (rigid body mass centers) of a human body, recording displacement in the motion process of the human body through universal camera equipment such as a mobile phone or a common camera, and obtaining acceleration of each section in the motion state based on a finite difference method. Based on Newton's second law, a quantitative relation between plantar force and human body movement is established, and then plantar force reconstruction is achieved.
Specifically as shown in fig. 3, the method for constructing the artificial plantar force based on computer vision provided by the application comprises the following steps:
s1, dividing the human body multi-rigid body model into a plurality of segments based on the human body multi-rigid body model, and acquiring the mass ratio of each segment in the human body multi-rigid body model; establishing a quantitative relation between plantar force and human body movement:
formula (2)
Wherein:is the human plantar force; />The quality corresponds to each segment of the human body; />The mass center acceleration of each segment of the human body; />Is the static weight of the human body; n is the total number of segments in the human body multi-rigid body model; i is the segment number, t is the time; the mass of each segment of human body>The total mass of the human body is combined with the mass ratio of each segment in the human body multi-rigid body model.
The method specifically comprises the following steps:
s1.1, dividing the human body multi-rigid body model into a plurality of segments based on the human body multi-rigid body model, and obtaining the mass ratio of each segment in the human body multi-rigid body model.
The embodiment of the application adopts a human body multi-rigid body model recommended by international ISO standard (shown in figure 4). The ISO standard divides the human body into 8 main segments of head, torso, upper arm, lower arm, hand, thigh, shank, and foot, etc., according to the human body skeletal structure. The ratio of the mass of each segment to the weight is shown in Table 1.
TABLE 1 mass ratio of each segment of human body multiple rigid body model
Wherein the value in parentheses indicates the number of corresponding segments, for example "hand (2)" indicates that there are 2 segments corresponding to "hand". Obviously, the sum of the mass fractions of all segments is equal to the static body weight of the human body, where the mass fraction of the torso exceeds 50% of the total weight. It should be noted that the mass ratio of each segment in the body weight of the human body shown in table 1 is a result of statistical averaging of large samples, and covers representative factors of age, sex, height, weight and the like of the samples, that is, fig. 4 can be regarded as a standard multi-rigid body model of the human body.
S1.2, establishing a quantitative relation between plantar force and human body movement.
The main starting points or principles of this embodiment are: plantar forces are derived from inertial effects of the moving mass of the human body. Based on the thought, the embodiment starts from a plantar force source and establishes a quantitative relation between plantar force and human body mass center motion based on strict mathematical derivation. The specific deduction process is as follows:
taking rhythmic flexion and extension movement (bouncing) of a human body as an example, based on Newton's second law, the human body is vertically stressed and balanced in the movement process, and a quantitative relation between plantar force and human body movement is established:
formula (2)
Wherein:is human plantar force, unit N; />The weight of each segment of the human body is kg; />The mass center acceleration of each segment of the human body is in units of +.>；/>Is the static weight of human body, unit N; n is the total number of segments in the human body multi-rigid body model, and as can be seen from fig. 4 and table 1, the total number of segments in the human body multi-rigid body model is 14, so that the value of n is 14; i is the segment number, t is the time; quality corresponding to each segment of human body/>The total mass of the human body is combined with the mass ratio of each section in the human body multi-rigid body model, the section is taken as a trunk as an example, the mass ratio of the trunk is 52.2 percent, when the total mass of the human body is 50kg, the section number corresponding to the trunk is 2, and the mass m of the trunk is 2 =50/>52.2%。
As is known from the formula (2), the human plantar force fluctuates in a static weight, in which the dynamic load is partially caused by the inertial force in the motion state of each segment of the human body. Because the weight proportion of each segment of the human body is known, the construction of plantar force can be realized based on the formula (2) by measuring the movement acceleration of each segment of the human body.
S2, recording the motion process of the human body by using the image pickup equipment so as to obtain a motion video.
The method specifically comprises the following steps: the marking points are arranged at the positions of the corresponding centroids of the human body sections corresponding to the sections in the human body multi-rigid body model one by one. The human body performs rhythmic flexion and extension movements, the image pickup device is used for shooting the movement process of the human body, and in the shooting process, the shooting picture is ensured to cover all the mark points so as to obtain a movement video for recording the displacement of all the mark points. It should be noted that each segment of the human body is different from each segment in the human body multi-rigid body model.
The marking point can be a round hard card or a round sticker, the diameter is about 2cm, and the marking is stuck on the corresponding centroid position of each segment of the human body by double faced adhesive tape, such as the black dot position shown in fig. 4. The numbers 1-8 on the black dot side in fig. 4 correspond to the number of the marker point and can also be regarded as the segment number i. In connection with the classification of the segments in table 1, the segment corresponding to number 1 in fig. 4 is the landmark of the head, the segment corresponding to number 2 is the landmark of the torso, the segment corresponding to number 3 is the landmark of the upper arm, the segment corresponding to number 4 is the landmark of the lower arm, the segment corresponding to number 5 is the landmark of the hand, the segment corresponding to number 6 is the landmark of the thigh, the segment corresponding to number 7 is the landmark of the calf, and the segment corresponding to number 8 is the landmark of the foot.
The color of the mark point generally ensures that the shot picture has stronger contrast in RGB color space, and black can be selected. And carrying out a human body movement test, and shooting by adopting a mobile phone or a common camera in the movement process, wherein the picture can cover all mark points of a human body in the shooting process. The duration of the motion video is preferably 20s-30s, and the motion period of more than 10 motion periods can be ensured during the human body motion test.
S3, framing the motion video, and acquiring displacement data of each segment of the human body, which corresponds to each segment in the human body multi-rigid body model one by one, based on a computer vision method.
And (3) carrying out framing processing on the motion video based on the motion video obtained in the step (S2).
And extracting the characteristic of the mark point of each frame of photo after framing to obtain the position coordinate of the pixel space of the mark point of each frame of photo, and converting the position coordinate into the position coordinate of the real space based on coordinate transformation.
Extracting the characteristic of the mark points and converting the coordinates of all the photos to obtain displacement data u of the mark points of each segment of the human body in the real space i (t)。
S4, based on displacement data u i (t) obtaining the acceleration a of each segment of the human body by using a finite difference method i (t)。
And (3) obtaining a displacement time course curve of the mark points of each segment of the human body based on the displacement data of the mark points of each segment of the human body in the real space.
Based on the displacement time course curve of the mark points of each segment of the human body, the speed time course curve and the acceleration time course curve of the mark points of each segment of the human body are obtained by adopting two finite differences. According to the acceleration time curve, the acceleration a can be obtained i (t)。
S5, determining the corresponding mass of each segment of the human body by combining the total mass of the human body with the mass ratio of each segment in the human body multi-rigid body model in table 1 based on the acceleration ai (t)Andand (5) calculating the human plantar force according to the established quantitative relation.
In this example, a verification test was performed using rhythmic flexion and extension exercise (bouncing). Subjects (males, 25 years old, weighing 68.5 kg) performed bouncing movements under metronome guidance at 1.5Hz, 2.5Hz and 3.5Hz, respectively. As shown in fig. 4, black mark points (diameter 2 cm) are stuck on the mass center positions of all the segments of the human body, so that the reliable connection between the mark points and the human body is ensured. The mobile phone is used for carrying out video acquisition on the motion process of the human body, and the sampling frame rate is 30 frames/s.
And extracting the characteristic of the mark point of each frame of photo after framing, firstly obtaining the position coordinates of the pixel space, and then converting the position coordinates into the position coordinates of the real space based on coordinate transformation.
Figure 5a shows the measured plantar force at a frequency of 1.5Hz and the displacement time course of the landmark points of each segment of the human body.
Fig. 5b shows the measured plantar force at a frequency of 2.5Hz and the displacement time course of the landmark points of each segment of the human body.
Fig. 5c shows the measured plantar force at a frequency of 3.5Hz and the displacement time course of the landmark points of each segment of the human body.
It is worth noting that since the marker point numbered 8 is located on the foot, the displacement during bouncing is close to 0, and is not shown in the displacement curve.
After the displacement time-course curve of each mark point is obtained, the speed and acceleration time-course curve of the mark point can be obtained by adopting a finite difference method. Taking the displacement of the marker point numbered 1 during 2.5Hz motion as an example, the velocity and acceleration time course curves obtained using the finite difference method are given.
Fig. 6a is a graph of the displacement time course of the marker point numbered 1 when the human body performs bouncing motion at the frequency of 2.5 Hz.
Fig. 6b is a velocity profile for a marker point numbered 1 obtained using the finite difference method.
Fig. 6c is an acceleration time course curve of marker point number 1 obtained by finite difference method.
Based on step S5, the artificial plantar force construction is completed. And comparing the constructed plantar force with the actually measured plantar force of the traditional method.
FIG. 7a shows a graph of plantar force constructed by the method of the present application versus measured plantar force at a frequency of 1.5 Hz.
FIG. 7b shows a graph of plantar force constructed by the method of the present application versus measured plantar force at a frequency of 2.5 Hz.
FIG. 7c shows a graph of plantar force constructed by the method of the present application versus measured plantar force at a frequency of 3.5 Hz.
Obviously, the reconstructed plantar force under different movement frequencies is very consistent with the actually measured plantar force, and the method provided by the application has better precision, and is convenient and feasible.
The application converts the traditional plantar force measurement into the human body movement (acceleration) measurement based on the quantitative relation between the human plantar force and the human body movement.
The traditional method needs to rely on special force measuring equipment, has high technical threshold and high test cost, is limited by the number of test instruments, and is difficult to carry out large-scale acquisition. The method can complete human motion acquisition by adopting general equipment such as a mobile phone or a common camera, and further complete the construction of plantar force, thereby greatly reducing the test threshold, being convenient and fast, ensuring good measurement precision, being convenient for large-scale acquisition of data and having wide popularization and application prospects.
Based on the same inventive concept, the application also provides a human plantar force construction system based on computer vision, which comprises the following steps:
the model dividing module is used for dividing the human body multi-rigid body model into a plurality of segments and acquiring the mass ratio of each segment in the human body multi-rigid body model;
the plurality of mark points are arranged at the positions of the corresponding centroids of the human body sections corresponding to the sections in the human body multi-rigid body model one by one;
the image pickup device is used for recording the motion process of a human body so as to obtain a motion video; the camera equipment comprises a mobile phone or a camera;
the first processing module is used for carrying out framing processing on the motion video and acquiring displacement data of each segment of the human body based on computer vision;
the second processing module is used for acquiring the acceleration of each segment of the human body by utilizing a finite difference method based on the displacement data; and
the third processing module is used for calculating the artificial plantar force based on the acceleration and the established quantitative relation between plantar force and human body movement;
wherein, the quantitative relation formula is:
wherein:is the human plantar force; />The quality corresponds to each segment of the human body; />The mass center acceleration of each segment of the human body; />Is the static weight of the human body; n is the total number of segments in the human body multi-rigid body model; i is the segment number, t is the time; the mass of each segment of human body>The total mass of the human body is combined with the mass ratio of each segment in the human body multi-rigid body model.
Based on the same inventive concept, the application also provides a computer vision-based artificial plantar force construction device, which comprises:
one or more processors; and
and a storage means for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement a computer vision-based artificial plantar force construction method.
The above description is only a preferred embodiment of the present application, and the protection scope of the present application is not limited to the above examples, and all technical solutions belonging to the concept of the present application belong to the protection scope of the present application. It should be noted that modifications and adaptations to the application without departing from the principles thereof are intended to be comprehended by those skilled in the art and are intended to be within the scope of the application.
Claims (10)
1. The human plantar force construction method based on computer vision is characterized by comprising the following steps of:
dividing the human body multi-rigid body model into a plurality of segments based on the human body multi-rigid body model, and acquiring the mass ratio of each segment in the human body multi-rigid body model; establishing a quantitative relation between plantar force and human body movement:
;
wherein:is the human plantar force; />The quality corresponds to each segment of the human body; />The mass center acceleration of each segment of the human body; />Is the static weight of the human body; n is the total number of segments in the human body multi-rigid body model; i is the segment number, t is the time; the mass of each segment of the human body is +.>The total mass of the human body is combined with the mass ratio of each segment in the human body multi-rigid body model to determine;
recording the motion process of the human body by using the camera equipment to obtain a motion video;
carrying out framing treatment on the motion video, and acquiring displacement data of each segment of the human body, which corresponds to each segment in the human body multi-rigid body model one by one, based on a computer vision method;
based on the displacement data, acquiring the acceleration of each segment of the human body by using a finite difference method;
based on the acceleration and the established quantitative relation, a human plantar force is calculated.
2. The method for constructing human plantar force based on computer vision according to claim 1, wherein the step of recording a human body movement process using an image capturing apparatus to obtain a movement video comprises:
the method comprises the steps of setting mark points at positions of corresponding centroids of all segments of a human body, which correspond to all segments in a human body multi-rigid-body model one by one;
the human body performs rhythmic flexion and extension movements, the image pickup device is used for shooting the movement process of the human body, and in the shooting process, the shooting picture is ensured to cover all the mark points so as to obtain a movement video for recording the displacement of all the mark points.
3. The computer vision-based artificial plantar force construction method according to claim 2, wherein the duration of the motion video is 20s-30s.
4. The method for constructing human plantar force based on computer vision according to claim 2, wherein the step of framing the motion video and acquiring displacement data of each segment of the human body corresponding to each segment in the human body multi-rigid body model one by one based on the computer vision method comprises:
carrying out framing treatment on the motion video;
extracting the characteristic of the mark point of each frame of photo after framing to obtain the position coordinate of the pixel space of the mark point of each frame of photo, and converting the position coordinate into the position coordinate of the real space based on coordinate transformation;
and extracting the characteristic of the mark points and converting coordinates of all the photos to obtain displacement data of the mark points of each segment of the human body in a real space.
5. The method for constructing human plantar force based on computer vision according to claim 4, wherein the step of obtaining the acceleration of each segment of the human body using the finite difference method based on the displacement data comprises:
obtaining a displacement time course curve of the mark points of each segment of the human body based on the displacement data of the mark points of each segment of the human body in the real space;
based on the displacement time course curve of the mark points of each segment of the human body, the speed time course curve and the acceleration time course curve of the mark points of each segment of the human body are obtained by adopting two finite differences.
6. The computer vision-based artificial plantar force construction method according to claim 1, wherein the image pickup apparatus includes a cellular phone or a camera.
7. A computer vision-based human plantar force construction system, comprising:
the model dividing module is used for dividing the human body multi-rigid body model into a plurality of segments and acquiring the mass ratio of each segment in the human body multi-rigid body model;
the image pickup device is used for recording the motion process of a human body so as to obtain a motion video;
the first processing module is used for carrying out framing processing on the motion video and acquiring displacement data of each segment of the human body based on computer vision;
the second processing module is used for acquiring the acceleration of each segment of the human body by utilizing a finite difference method based on the displacement data; and
the third processing module is used for calculating the artificial plantar force based on the acceleration and the established quantitative relation between plantar force and human body movement;
wherein the quantitative relation is:
;
wherein:is the human plantar force; />The quality corresponds to each segment of the human body; />The mass center acceleration of each segment of the human body; />Is the static weight of the human body; n is the total number of segments in the human body multi-rigid body model; i is the segment number, t is the time; the mass of each segment of the human body is +.>The total mass of the human body is combined with the mass ratio of each segment in the human body multi-rigid body model.
8. The computer vision based artificial plantar force construction system of claim 7, further comprising a plurality of marker points disposed at positions of corresponding centroids of segments of the human body in one-to-one correspondence with the segments in the human body multi-rigid body model.
9. The computer vision based artificial plantar force constructing system of claim 7, wherein the image capturing apparatus includes a cell phone or a camera.
10. A computer vision-based human plantar force construction apparatus, comprising:
one or more processors; and
storage means for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the computer vision-based artificial plantar force construction method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310959679.6A CN116680963A (en) | 2023-08-01 | 2023-08-01 | Method, system and equipment for constructing human plantar force based on computer vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310959679.6A CN116680963A (en) | 2023-08-01 | 2023-08-01 | Method, system and equipment for constructing human plantar force based on computer vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116680963A true CN116680963A (en) | 2023-09-01 |
Family
ID=87791346
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310959679.6A Pending CN116680963A (en) | 2023-08-01 | 2023-08-01 | Method, system and equipment for constructing human plantar force based on computer vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116680963A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104808510A (en) * | 2015-02-16 | 2015-07-29 | 中国航天员科研训练中心 | Virtual astronaut multilayer movement control simulation method |
| CN109394231A (en) * | 2018-12-10 | 2019-03-01 | 刘坤 | One kind stand moving equilibrium monitoring with dynamic analysis system |
| CN113837150A (en) * | 2021-11-25 | 2021-12-24 | 湖南大学 | Non-contact tire pressure obtaining method based on computer vision and related device |
| CN116358622A (en) * | 2023-01-10 | 2023-06-30 | 中国地质大学（武汉） | Human-bridge mutual feedback displacement monitoring and early warning system and method based on vision technology |
-
2023
- 2023-08-01 CN CN202310959679.6A patent/CN116680963A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104808510A (en) * | 2015-02-16 | 2015-07-29 | 中国航天员科研训练中心 | Virtual astronaut multilayer movement control simulation method |
| CN109394231A (en) * | 2018-12-10 | 2019-03-01 | 刘坤 | One kind stand moving equilibrium monitoring with dynamic analysis system |
| CN113837150A (en) * | 2021-11-25 | 2021-12-24 | 湖南大学 | Non-contact tire pressure obtaining method based on computer vision and related device |
| CN116358622A (en) * | 2023-01-10 | 2023-06-30 | 中国地质大学（武汉） | Human-bridge mutual feedback displacement monitoring and early warning system and method based on vision technology |
Non-Patent Citations (1)
| Title |
|---|
| 张梦诗等: "基于动作捕捉与刚体动力学的步行荷载重构与仿真", 应用基础与工程科学学报, vol. 21, no. 5, pages 961 - 972 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108095725B (en) | Human body motion capability integrated test device and use method thereof | |
| US9524424B2 (en) | Calculation of minimum ground clearance using body worn sensors | |
| Gindre et al. | Reliability and validity of the Myotest® for measuring running stride kinematics | |
| US20140330171A1 (en) | Method and device for monitoring postural and movement balance for fall prevention | |
| CN108968965A (en) | Portable body balance evaluation and test training system | |
| CN107871116A (en) | For the method and system for the postural balance for determining people | |
| Lord et al. | Concurrent validity of accelerometry to measure gait in Parkinsons Disease | |
| Silsupadol et al. | Smartphone-based assessment of gait during straight walking, turning, and walking speed modulation in laboratory and free-living environments | |
| CN109528201A (en) | A kind of balanced capacity detection method | |
| CN108338790B (en) | Gait analysis and fall assessment system | |
| Yang et al. | A smart environment-adapting timed-up-and-go system powered by sensor-embedded insoles | |
| JP2006101973A (en) | Apparatus for health management | |
| US20170296113A1 (en) | Combined device that measures the body weight and balance index | |
| KR101446653B1 (en) | Health and rehabilitation game apparatus, system and game method | |
| Anwary et al. | Validity and consistency of concurrent extraction of gait features using inertial measurement units and motion capture system | |
| WO2017075953A1 (en) | Accelerometer-based method and device for predicting heart rate during exercise | |
| CN108542398A (en) | A kind of earth's surface gait and balance detecting device | |
| JP2020192307A (en) | Lower limb muscle strength evaluation method, lower limb muscle strength evaluation program, lower limb muscle strength evaluation device, and lower limb muscle strength evaluation system | |
| CN116680963A (en) | Method, system and equipment for constructing human plantar force based on computer vision | |
| Caramia et al. | Spatio-temporal gait parameters as estimated from wearable sensors placed at different waist levels | |
| Qi | Volleyball action extraction and dynamic recognition based on gait tactile sensor | |
| Zhang et al. | A Pilot Study of Dynamic Stability Indices for Potential Application of Identifying Older Adult Fallers | |
| US10548510B2 (en) | Objective balance error scoring system | |
| JP2020151470A (en) | Walking evaluation device, walking evaluation method, and program | |
| CN110801230A (en) | Evaluation method for lower limb fracture rehabilitation effect |
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 |