CN107392939A - Indoor sport observation device, method and storage medium based on body-sensing technology - Google Patents
Indoor sport observation device, method and storage medium based on body-sensing technology Download PDFInfo
- Publication number
- CN107392939A CN107392939A CN201710649478.0A CN201710649478A CN107392939A CN 107392939 A CN107392939 A CN 107392939A CN 201710649478 A CN201710649478 A CN 201710649478A CN 107392939 A CN107392939 A CN 107392939A
- Authority
- CN
- China
- Prior art keywords
- user
- motion
- sports
- type
- action
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
- G06T7/251—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments involving models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
- G06T2207/10021—Stereoscopic video; Stereoscopic image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30008—Bone
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
Indoor sport observation device, method and storage medium based on body-sensing technology, described device include:Somatosensory device, for obtaining and exporting RGB image and depth information;Identification module is moved, for according to depth information, extracting the characteristic vector of human motion, matching and identify the type of sports of user, inquire about exercise intensity and the movable information of user is calculated;Library module is moved, including preserves the action database of various motion models;Exercise intensity module, preserve the exercise intensity parameter of various different motion types.The present invention can calculate the motion state and type of user, show the movable informations such as the calorie of consumption, oxygen demand, run duration, facilitate user to check, it is succinct convenient to have, compatibility by force and it is more recreational the characteristics of.
Description
Technical field
The application is related to image recognition and compares field, and indoor sport is entered based on body-sensing technology specifically, being related to one kind
The device and method of row observation.
Background technology
In recent years, as work and the increase of life stress, people are increasingly keen to move, it is desirable in this way
To change the physical and mental statuse of oneself.Motion is divided into indoor and outdoors, but due to traffic, weather and air quality reason, it is indoor
Motion is more welcomed by the general public.Indoor sport time span is controllable, and will not be limited by weather, either hot
Summer or the winter of bitter cold, air quality is bad in the winter time in addition is, interior can equally reach preferable air quality.
Such as a kind of intelligent Yoga mat of Chinese patent 201520733691, a kind of dress for doing Yoga indoors this method provide
Put.This motion consumption that can only monitor Yoga, function are single, it is impossible to evoke the enthusiasm of user's persistent movement.Moreover, once only
The motion state of a people can be observed, it is recreational not strong, it can not meet the diversified demand of indoor sport.
And some other motion monitoring equipment generally require to wear some accessories with user, or special
Sports equipment on carry out, on the one hand the use feeling degree to user is bad, still further aspect, and above-mentioned special sports equipment is not
Only cost is larger, and more space-consuming indoors.
Therefore, when moving indoors how untrammeled motion, do not wear any auxiliary equipment, and can be with
Any different forms of motion is applicable, reduces occupation area of equipment, movement entertainment is improved, turns into prior art urgent need to resolve
Technical problem.
The content of the invention
It is an object of the invention to propose a kind of indoor sport observation device and method based on body-sensing technology, can calculate
Go out the motion state and type of user, show the movable informations such as the calorie of consumption, oxygen demand, run duration, facilitate user to look into
See.
To use following technical scheme up to this purpose, the present invention:
A kind of indoor sport observation device based on body-sensing technology, including:
Somatosensory device, for shooting user, measurement space 3 d measurement data, and scheme to motion identification module output RGB
The depth information of picture and image;
Identification module is moved, for according to the depth information per two field picture, extraction to form the n dimensions of the human motion of present frame
Characteristic vector, and with the motion model comparison match in action database, identify the type of sports of user, and inquire about exercise intensity
Module, the exercise intensity of the type of sports is obtained, and the fortune of user is calculated according to the body weight of exercise intensity and user
Dynamic information:
Library module, including action database are moved, the motion mould of various type of sports is preserved in the action database
Type;
Exercise intensity module, the exercise intensity parameter of various different motion types is preserved, so that motion identification module is looked into
Ask and the exercise intensity of the type of sports of active user is calculated.
Optionally, it is described to be used to form people according to the depth information per two field picture, extraction in the motion identification module
Body motion n dimensional feature vectors, and with the motion model comparison match in action database, identify that the type of sports of user is specific
For:According to the depth information of every two field picture, human body 3D skeletons are extracted, the joint point data of key is filtered out from human body 3D skeletons,
And from key artis extracting data motion characteristic value, construction feature sequence vector, by the characteristic vector sequence normalizing
Change, form the n dimension normalization characteristic vectors of the human action of present frame, carry out action recognition, it is special to calculate current n dimension normalization
Sign vector and the distance of the characteristic vector of motion model, if distance is less than corresponding threshold value, then it is assumed that action matching, continuous
With multiple actions, the type of sports of user is judged.
Optionally, in the motion identification module,
Also need to judge whether be children in extraction human body 3D skeletons, if children, then to the joint of human body 3D skeletons
Coordinate is normalized, and the normalization includes limb size normalizing, reference zero normalizing and direction normalizing;
The crucial joint point data, include the joint point data of pin, knee and stern;
The characteristic vector includes body height, movement velocity, movement angle and movement locus;
The action recognition uses AP clustering algorithms, the characteristic vector of the n dimensions normalization characteristic vector and motion model
Distance be Euclidean distance.
Optionally, the action database is by the way of training, collection input in advance;
The body weight of the user is calculated by height, the measurements of the chest, waist and hips of somatosensory device acquisition user or voluntarily inputted.
Optionally, the somatosensory device includes infrared transmitter, infrared remote receiver, RGB cameras and 3D body-sensing chips.
The invention also discloses a kind of indoor sport observation procedure based on body-sensing technology, comprise the following steps:
Step S110:User, measurement space 3 d measurement data are shot, and exports the depth letter of RGB image and image
Breath;
Step S120:According to the depth information of every two field picture, human body 3D skeletons are extracted, key is filtered out from human body 3D skeletons
Joint point data;
Step S130:From the artis extracting data motion characteristic value of key, construction feature sequence vector, by the spy
Sequence vector normalization is levied, forms the n dimension normalization characteristic vectors of the human action of present frame;
Step S140:Using n dimension normalization characteristic vectors, action recognition is carried out, it is special to calculate current n dimension normalization
Sign vector and the distance of the characteristic vector of motion model in action database, if distance is less than corresponding threshold value, then it is assumed that dynamic
Match, the multiple actions of continuous coupling, judge the type of sports of user;
Step S150:Exercise intensity data are inquired about according to the type of sports of the user, obtain the fortune of the type of sports
Fatigue resistance, and the movable information of user is calculated according to the body weight of exercise intensity and user.
Optionally, in the step s 120, also need to judge whether be children in extraction human body 3D skeletons, if children,
Then the joint coordinates of human body 3D skeletons are normalized, normalization include limb size normalizing, reference zero normalizing and
Direction normalizing;
The crucial joint point data, mainly joint data including pin, knee and stern;
In step s 130, the characteristic vector includes body height, movement velocity, movement angle and movement locus.
Optionally, in step S140, AP clustering algorithms are used when carrying out action recognition, the n ties up normalization characteristic
Vector and the distance of the characteristic vector of motion model are Euclidean distance;
The action database is by the way of training, collection input in advance.
Optionally, in step S150, the body weight of the user is calculated by the height of user, measurements of the chest, waist and hips in RGB image,
Or voluntarily set.
The present invention further discloses a kind of storage medium, and for storing computer executable instructions, the computer can
Execute instruction performs above-mentioned method when being executed by processor.
The present invention has the advantage that:
First, more succinctly facilitating, realized by existing body-sensing camera and corresponding application software, cost-effective, side
Portable belt, plug and play.
Second, compatible more type of sports, multi-motion type can be observed, such as Yoga, dancing, chin-up, bow
Sleeping support, remain where one is etc., more type of sports can be inputted in action database by action training.
Third, it is more recreational, using body-sensing camera, normalized n dimensional feature vectors are calculated, therefore can be same
When observe more people, and can be combined together with game, user can not only see the movable information of individual in motion,
It can also be played, both relax body and mind, also reached the purpose of body-building, killed two birds with one stone.
Brief description of the drawings
Fig. 1 is the use state figure according to the indoor sport observation device of the specific embodiment of the invention;
Fig. 2 is the module map according to the indoor sport observation device of the specific embodiment of the invention;
Fig. 3 is the module map according to the somatosensory device of the specific embodiment of the invention;
Fig. 4 is the flow chart according to the indoor sport observation procedure of the specific embodiment of the invention.
The technical characteristic that reference in figure refers to respectively is:
10th, somatosensory device;20th, identification module is moved;30th, library module is moved;40th, exercise intensity module;11st, infrared emission
Device;12nd, infrared remote receiver;13rd, RGB cameras;14th, 3D body-sensings chip.
Embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention, rather than limitation of the invention.It also should be noted that in order to just
Part related to the present invention rather than entire infrastructure are illustrate only in description, accompanying drawing.
Generally require to wear some accessories with user relative to existing motion monitoring equipment, or special
Sports equipment on carry out.Body-sensing technology can realize the accurate detect and track to personage by body-sensing technology, and to dynamic
Work, personage's movement locus, motion state are detected and analyzed, and and then obtain the movable information of human body.
The present invention uses somatosensory device, the depth information and RGB image information of user is gathered in real time, in body-sensing camera face
Before, user can arbitrarily do the motion for wanting to do, such as Yoga, chin-up, push-up, remain where one is, dance, to personage
Accurate detection, tracking, realize the detection and analysis to movement posture and personage's movement locus, and the action number with storing before
Compared according to storehouse, identify the type of sports and state of user, and the type of sports pair is inquired about by the body weight of user accordingly
The exercise intensity answered, the movable informations such as the calorie of the consumption of user, oxygen demand, run duration are calculated, and carried out accordingly
It has been shown that, to facilitate user to check.
Embodiment 1:
Shown referring to Fig. 1 makes according to the indoor sport observation device based on body-sensing technology of the specific embodiment of the invention
With state diagram, user moves in face of somatosensory device 10 as can be seen from the figure, and observation device calculates and on the display apparatus
Output campaign information, the movable information include but is not limited to the calorie of the consumption of user, oxygen demand, run duration.
Fig. 2 shows the module map of indoor sport observation device, and the observation device includes:
Somatosensory device 10, for shooting user, measurement space 3 d measurement data, and exported to motion identification module 20
The depth information of RGB image and image;
Identification module 20 is moved, for forming the n of the human motion of present frame according to the depth information per two field picture, extraction
Dimensional feature vector, and with the motion model comparison match in action database, identify the type of sports of user, and it is strong to inquire about motion
Module is spent, obtains the exercise intensity of the type of sports, and be calculated user's according to the body weight of exercise intensity and user
Movable information, exemplary, the exercise intensity includes the calorie of user's consumption, oxygen demand;
Library module 30, including action database are moved, the motion of various type of sports is preserved in the action database
Model;
Exercise intensity module 40, the exercise intensity parameter of various different motion types is preserved, for moving identification module
20 inquire about and the exercise intensity of the type of sports of active user are calculated.
Specifically, in the motion identification module, for according to the depth information per two field picture, extraction to form human body fortune
Dynamic n dimensional feature vectors, and with the motion model comparison match in action database, the type of sports for identifying user is specially:
According to the depth information of every two field picture, human body 3D skeletons are extracted, skeleton is the data of the important artis of human body, from human body 3D bones
Frame filter out key joint point data, and from key artis extracting data motion characteristic value, construction feature vector sequence
Row, the characteristic vector sequence is normalized, and is formed the n dimension normalization characteristic vectors of the human action of present frame, is acted
Identification, current n dimension normalization characteristic vectors and the distance of the characteristic vector of motion model are calculated, if corresponding apart from being less than
Threshold value, then it is assumed that action matching, the multiple actions of continuous coupling, judge the type of sports of user.
Specifically, also need to judge whether be children in extraction human body 3D skeletons, for example, by the size of skeleton whether be
Children, if children, build is smaller, then the joint coordinates of human body 3D skeletons is normalized, and is carried by normalization
High subsequent action accuracy of identification;Normalization includes limb size normalizing, reference zero normalizing and direction normalizing.
The crucial joint point data, for running, the action such as jump/Yoga, mainly including the joint such as pin, knee and stern number
According to the joint point data by extracting key can optimize motion characteristic, improve efficiency of algorithm.
The characteristic vector includes body height, movement velocity, movement angle and movement locus, and characteristic vector is returned
One change being capable of the Recognition Different that brings of less human body otherness and stability.
AP clustering algorithms, the feature of the n dimensions normalization characteristic vector and motion model are used when carrying out action recognition
The distance of vector is Euclidean distance, that is, uses Euclidean distance as measuring similarity.
The action database can be by the way of training, collection input in advance.Specific mode can be with obtaining
It is identical to take the mode of family motion characteristics sequence vector, the above method is repeated several times, so as to train to obtain a variety of fortune
Movable model.Exemplary, motion library module can include action training submodule, and the spy of user movement is obtained by being repeated several times
The identical step of sequence vector is levied, Category learning is carried out to action.
Further, the body weight of the user can be calculated by height, the measurements of the chest, waist and hips of somatosensory device acquisition user,
Can voluntarily input, with prevent environment it is poor when, body-sensing camera calculates the larger situation of body weight error.
In the present invention, somatosensory device can use the somatosensory device of prior art, referring to Fig. 3, the somatosensory device 10
Including infrared transmitter 21, infrared remote receiver 22, RGB cameras 23 and 3D body-sensings chip 24, the three of space can be exported in real time
Dimension data measures and shooting RGB image, and the depth information of every two field picture is calculated.
Therefore, the present invention passes through the somatosensory device with 3D functions, the human body 3D skeleton images of real-time capture user, extraction
Crucial artis characteristic vector, characteristic vector is input in action database, finds out the action of matching degree highest, continuous identification
Multiple actions, current type of sports is judged, so as to calculate the movable information of user.
It should be noted that the type of action that user needs to carry out can be set in advance in specifically used, sort out in advance,
To reduce the amount of calculation of action recognition, it can also automatically be compared by observation device, compared with a variety of type of sports
Compared with so as to finding matching degree highest type of sports.
In the present invention, it can be a central processing unit to move identification module 20, entered by code in run memory
The various calculating of row, it can be different types of database stores information to move library module 30 and exercise intensity module 40, be stored in
It is called in the calculation for motion identification module 20 so that corresponding result is calculated in memory.
Therefore, the indoor sport observation device of the invention based on body-sensing technology has as follows excellent compared with prior art
Point:
First, more succinctly facilitate, traditional motion consumption observation procedure is and expensive, it is necessary to special equipment, than
It is heavier, it is not portable.The present invention can be realized by existing body-sensing camera and corresponding application software, cost-effective,
It is convenient for carrying, plug and play.
Second, compatible more type of sports, traditional motion consumption observation procedure, equipment is all special a, equipment
A kind of type of sports can only be observed, motion is single, easily uninteresting.The present invention can observe multi-motion type, such as Yoga, jump
Dance, chin-up, push-up, remain where one is etc., more motions can be inputted in action database by action training
Type.
Third, more recreational, traditional motion consumption observation procedure, all when one uses, and trip can not be combined
Play, time have been grown just dull.The present invention is due to using body-sensing camera, being calculated normalized n dimensional feature vectors, therefore
More people can be observed simultaneously, and can be combined together with game, and user can not only see the motion of individual in motion
Information, additionally it is possible to played, both relax body and mind, also reached the purpose of body-building, killed two birds with one stone.
Embodiment 2:
Referring to Fig. 4, a kind of flow chart of the indoor sport observation procedure based on body-sensing technology is shown, this method can fit
For the indoor sport observation device in embodiment 1, this method comprises the following steps:
Step S110:User, measurement space 3 d measurement data are shot, and exports the depth letter of RGB image and image
Breath;
Step S120:According to the depth information of every two field picture, human body 3D skeletons are extracted, skeleton is the important artis of human body
Data, from human body 3D skeletons filter out key joint point data;
Step S130:From the artis extracting data motion characteristic value of key, construction feature sequence vector, by the spy
Sequence vector normalization is levied, forms the n dimension normalization characteristic vectors of the human action of present frame;
Step S140:Using n dimension normalization characteristic vectors, action recognition is carried out, it is special to calculate current n dimension normalization
Sign vector and the distance of the characteristic vector of motion model in action database, if distance is less than corresponding threshold value, then it is assumed that dynamic
Match, the multiple actions of continuous coupling, judge the type of sports of user;
Step S150:Exercise intensity data are inquired about according to the type of sports of the user, obtain the fortune of the type of sports
Fatigue resistance, and the movable information of user is calculated according to the body weight of exercise intensity and user.
Further, in the step s 120, also need to judge whether be children in extraction human body 3D skeletons, such as pass through bone
Whether the size of frame is children, if children, build is smaller, then place is normalized to the joint coordinates of human body 3D skeletons
Reason, subsequent action accuracy of identification is improved by normalizing;Normalization includes limb size normalizing, reference zero normalizing and direction and returned
One;The crucial joint point data, for running, the action such as jump/Yoga, mainly including the joint such as pin, knee and stern data.
In step s 130, the characteristic vector includes body height, movement velocity, movement angle and movement locus.
In step S140, AP clustering algorithms, the n dimensions normalization characteristic vector and fortune are used when carrying out action recognition
The distance of the characteristic vector of movable model is Euclidean distance, that is, uses Euclidean distance as measuring similarity.The action database can
In a manner of using training, collection input in advance.Specific mode can be with the characteristic vector sequence of acquisition user movement
Mode it is identical, i.e. step S110-S130, above steps may be repeated multiple times, so as to train to obtain a variety of motion models.
In step S150, the body weight of the user can be calculated by the height of user, measurements of the chest, waist and hips in RGB image,
Can voluntarily set, with prevent environment it is poor when, body-sensing camera calculates the larger situation of body weight error.
Exemplary, the movable information of user can be supplied to user to be observed by showing on the display apparatus.
Embodiment 3:
The present invention further discloses a kind of storage medium, and for storing computer executable instructions, the computer can
Execute instruction performs the method described in embodiment 2 when being executed by processor.
As skilled in the art will be aware of, various aspects of the invention may be implemented as system, method or meter
Calculation machine program product.Therefore, various aspects of the invention can take following form:Complete hardware embodiment, complete software
Embodiment (including firmware, resident software, microcode etc.) or herein generally can referred to as " circuit ", " module " or
The embodiment that software aspects are combined with hardware aspect of " system ".In addition, the aspect of the present invention can take following shape
Formula:The computer program product realized in one or more computer-readable mediums, computer-readable medium have thereon
The computer readable program code of realization.
Any combination of one or more computer-readable mediums can be utilized.Computer-readable medium can be computer
Readable signal medium or computer-readable recording medium.Computer-readable recording medium can be such as (but not limited to) electronics,
Magnetic, optical, electromagnetism, infrared or semiconductor system, device, or foregoing any appropriate combination.Meter
The more specifically example (exhaustive to enumerate) of calculation machine readable storage medium storing program for executing will include the following:With one or more electric wire
Electrical connection, portable computer diskette, hard disk, random access memory (RAM), read-only storage (ROM), erasable compile
Journey read-only storage (EPROM or flash memory), optical fiber, portable optic disk read-only storage (CD-ROM), light storage device,
Magnetic memory apparatus or foregoing any appropriate combination.In the context of this document, computer-readable recording medium can be
It can include or store the program used by instruction execution system, device or combined command execution system, equipment or dress
Put any tangible medium of the program used.
Computer-readable signal media can include the data-signal propagated, and the data-signal of the propagation has wherein
Such as the computer readable program code of the part realization in a base band or as carrier wave.The signal of such propagation can use
Any form in diversified forms, include but is not limited to:Electromagnetism, optical or its any appropriate combination.It is computer-readable
Signal media can be following any computer-readable medium:It is not computer-readable recording medium, and can be to by instructing
The program that execution system, device use or combined command execution system, device use is communicated, propagated
Or transmission.
Including but not limited to wireless, wired, fiber optic cables, RF etc. or foregoing can be used any appropriately combined any
Suitable medium transmits the program code realized on a computer-readable medium.
Computer program code for performing for the operation of each side of the present invention can be with one or more programming languages
Any combination of speech is write, and the programming language includes:The programming language of object-oriented such as Java, Smalltalk, C++ etc.;
And conventional process programming language such as " C " programming language or similar programming language.Program code can be used as independent software package
Fully on the user computer, partly perform on the user computer;Partly exist on the user computer and partly
Performed on remote computer;Or fully perform on remote computer or server.In the latter case, can be by far
Journey computer by any type of network connection including LAN (LAN) or wide area network (WAN) to subscriber computer, or
It can be attached with outer computer (such as internet by using ISP).
The flow chart legend and/or frame of the methods of embodiments of the present invention, equipment (system) and computer program product
Figure describes various aspects of the invention.It will be appreciated that each block and flow chart figure of flow chart legend and/or block diagram
The combination of example and/or the block in block diagram can be realized by computer program instructions.These computer program instructions can be carried
The processor of all-purpose computer, special-purpose computer or other programmable data processing devices is supplied to, to produce machine so that (warp
By the computing device of computer or other programmable data processing devices) instruction created for implementation process figure and/or frame
The device for the function/action specified in segment or block.
These computer program instructions can also be stored in can instruct computer, other programmable data processing devices
Or in the computer-readable medium that runs in a specific way of other devices so that the instruction production stored in computer-readable medium
It is raw to include realizing the product of the instruction for the function/action specified in flow chart and/or block diagram or block.
Computer program instructions can also be loaded on computer, other programmable data processing devices or other devices
On, so as to perform a series of operable steps on computer, other programmable devices or other devices to produce computer reality
Existing process so that the instruction performed on computer or other programmable devices is provided for realizing in flow chart and/or frame
The process for the function/action specified in segment or block.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for those skilled in the art
For, the present invention can have various changes and change.All any modifications made within spirit and principles of the present invention, it is equal
Replace, improve etc., it should be included in the scope of the protection.
Claims (10)
1. a kind of indoor sport observation device based on body-sensing technology, including:
Somatosensory device, for shooting user, measurement space 3 d measurement data, and to motion identification module output RGB image with
And the depth information of image;
Identification module is moved, for forming the n dimensional features of the human motion of present frame according to the depth information per two field picture, extraction
Vector, and with the motion model comparison match in action database, identify the type of sports of user, and inquire about exercise intensity mould
Block, the exercise intensity of the type of sports is obtained, and the motion of user is calculated according to the body weight of exercise intensity and user
Information;
Library module, including action database are moved, the motion model of various type of sports is preserved in the action database;
Exercise intensity module, the exercise intensity parameter of various different motion types is preserved, for the inquiry of motion identification module simultaneously
The exercise intensity of the type of sports of active user is calculated.
2. indoor sport observation device according to claim 1, it is characterised in that:
Described to be used for according to the depth information per two field picture in the motion identification module, extraction forms the n dimensions of human motion
Characteristic vector, and with the motion model comparison match in action database, the type of sports for identifying user is specially:According to every frame
The depth information of image, extract human body 3D skeletons, from human body 3D skeletons filter out key joint point data, and from key pass
Motion characteristic value is extracted in node data, construction feature sequence vector, the characteristic vector sequence is normalized, forms present frame
Human action n dimension normalization characteristic vectors, carry out action recognition, calculate current n dimensions normalization characteristic vector and motion
The distance of the characteristic vector of model, if distance is less than corresponding threshold value, then it is assumed that action matching, the multiple actions of continuous coupling,
Judge the type of sports of user.
3. indoor sport observation device according to claim 2, it is characterised in that:
In the motion identification module,
Also need to judge whether be children in extraction human body 3D skeletons, if children, then to the joint coordinates of human body 3D skeletons
It is normalized, the normalization includes limb size normalizing, reference zero normalizing and direction normalizing;
The crucial joint point data, include the joint point data of pin, knee and stern;
The characteristic vector includes body height, movement velocity, movement angle and movement locus;
The action recognition uses AP clustering algorithms, the n dimensions normalization characteristic vector and the characteristic vector of motion model away from
From for Euclidean distance.
4. the indoor sport observation device according to any one in claim 1-3, it is characterised in that:
The action database is by the way of training, collection input in advance;
The body weight of the user is calculated by height, the measurements of the chest, waist and hips of somatosensory device acquisition user or voluntarily inputted.
5. indoor sport observation device according to claim 4, it is characterised in that:
The somatosensory device includes infrared transmitter, infrared remote receiver, RGB cameras and 3D body-sensing chips.
6. a kind of indoor sport observation procedure based on body-sensing technology, comprises the following steps:
Step S110:User, measurement space 3 d measurement data are shot, and exports the depth information of RGB image and image;
Step S120:According to the depth information of every two field picture, human body 3D skeletons are extracted, the pass of key is filtered out from human body 3D skeletons
Node data;
Step S130:From key artis extracting data motion characteristic value, construction feature sequence vector, by the feature to
Sequence normalization is measured, forms the n dimension normalization characteristic vectors of the human action of present frame;
Step S140:Using n dimension normalization characteristic vectors, carry out action recognition, calculate current n tie up normalization characteristic to
The distance of amount and the characteristic vector of motion model in action database, if distance is less than corresponding threshold value, then it is assumed that action
Match somebody with somebody, the multiple actions of continuous coupling, judge the type of sports of user;
Step S150:Exercise intensity data are inquired about according to the type of sports of the user, the motion for obtaining the type of sports is strong
Spend, and the movable information of user is calculated according to the body weight of exercise intensity and user.
7. indoor sport observation procedure according to claim 6, it is characterised in that:
In the step s 120, also need to judge whether be children in extraction human body 3D skeletons, if children, then to human body 3D bones
The joint coordinates of frame are normalized, and normalization includes limb size normalizing, reference zero normalizing and direction normalizing;
The crucial joint point data, mainly joint data including pin, knee and stern;
In step s 130, the characteristic vector includes body height, movement velocity, movement angle and movement locus.
8. indoor sport observation procedure according to claim 6, it is characterised in that:
In step S140, AP clustering algorithms, the n dimensions normalization characteristic vector and motion mould are used when carrying out action recognition
The distance of the characteristic vector of type is Euclidean distance;
The action database is by the way of training, collection input in advance.
9. indoor sport observation procedure according to claim 6, it is characterised in that:
In step S150, the body weight of the user is calculated by the height of user, measurements of the chest, waist and hips in RGB image, or is voluntarily set
Put.
10. a kind of storage medium, for storing computer executable instructions,
Computer executable instructions perform claim when being executed by processor requires the method described in any one in 6-9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710649478.0A CN107392939A (en) | 2017-08-01 | 2017-08-01 | Indoor sport observation device, method and storage medium based on body-sensing technology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710649478.0A CN107392939A (en) | 2017-08-01 | 2017-08-01 | Indoor sport observation device, method and storage medium based on body-sensing technology |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107392939A true CN107392939A (en) | 2017-11-24 |
Family
ID=60344535
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710649478.0A Pending CN107392939A (en) | 2017-08-01 | 2017-08-01 | Indoor sport observation device, method and storage medium based on body-sensing technology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107392939A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108970084A (en) * | 2018-06-29 | 2018-12-11 | 西安深睐信息科技有限公司 | A kind of moving scene analogy method of Behavior-based control identification |
CN109086659A (en) * | 2018-06-13 | 2018-12-25 | 深圳市感动智能科技有限公司 | A kind of Human bodys' response method and apparatus based on multimode road Fusion Features |
CN109815907A (en) * | 2019-01-25 | 2019-05-28 | 深圳市象形字科技股份有限公司 | A kind of sit-ups attitude detection and guidance method based on computer vision technique |
CN110377824A (en) * | 2019-07-15 | 2019-10-25 | 贝壳技术有限公司 | Information-pushing method, device, computer readable storage medium and electronic equipment |
CN110719455A (en) * | 2019-09-29 | 2020-01-21 | 深圳市火乐科技发展有限公司 | Video projection method and related device |
CN110826422A (en) * | 2019-10-18 | 2020-02-21 | 北京量健智能科技有限公司 | System and method for obtaining motion parameter information |
CN111240486A (en) * | 2020-02-17 | 2020-06-05 | 河北冀联人力资源服务集团有限公司 | Data processing method and system based on edge calculation |
CN111260678A (en) * | 2018-11-30 | 2020-06-09 | 百度在线网络技术(北京)有限公司 | Gymnastics assistant learning method and device, storage medium and terminal equipment |
CN112245892A (en) * | 2020-09-25 | 2021-01-22 | 北京汇洋时代科技有限公司 | System for testing movement by using 3D camera |
TWI783679B (en) * | 2020-09-14 | 2022-11-11 | 日商亞西斯特有限公司 | Telescopic motion measurement device and telescopic motion measurement program |
CN117653995A (en) * | 2022-08-18 | 2024-03-08 | 山东海天智能工程有限公司 | Brain-computer interaction technology rehabilitation training device adopting neuromorphic perception movement system |
WO2024131796A1 (en) * | 2022-12-19 | 2024-06-27 | FindSatoshi Lab Limited | Method for detecting validity of human body movement |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101519775B1 (en) * | 2014-01-13 | 2015-05-12 | 인천대학교 산학협력단 | Method and apparatus for generating animation based on object motion |
CN105748039A (en) * | 2016-02-03 | 2016-07-13 | 天脉聚源(北京)传媒科技有限公司 | Method and device for calculating exercise energy consumption |
CN105930767A (en) * | 2016-04-06 | 2016-09-07 | 南京华捷艾米软件科技有限公司 | Human body skeleton-based action recognition method |
-
2017
- 2017-08-01 CN CN201710649478.0A patent/CN107392939A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101519775B1 (en) * | 2014-01-13 | 2015-05-12 | 인천대학교 산학협력단 | Method and apparatus for generating animation based on object motion |
CN105748039A (en) * | 2016-02-03 | 2016-07-13 | 天脉聚源(北京)传媒科技有限公司 | Method and device for calculating exercise energy consumption |
CN105930767A (en) * | 2016-04-06 | 2016-09-07 | 南京华捷艾米软件科技有限公司 | Human body skeleton-based action recognition method |
Non-Patent Citations (1)
Title |
---|
范凯熹,等: "《信息交互设计》", 28 February 2015 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109086659A (en) * | 2018-06-13 | 2018-12-25 | 深圳市感动智能科技有限公司 | A kind of Human bodys' response method and apparatus based on multimode road Fusion Features |
CN109086659B (en) * | 2018-06-13 | 2023-01-31 | 深圳市感动智能科技有限公司 | Human behavior recognition method and device based on multi-channel feature fusion |
CN108970084A (en) * | 2018-06-29 | 2018-12-11 | 西安深睐信息科技有限公司 | A kind of moving scene analogy method of Behavior-based control identification |
CN111260678A (en) * | 2018-11-30 | 2020-06-09 | 百度在线网络技术(北京)有限公司 | Gymnastics assistant learning method and device, storage medium and terminal equipment |
CN109815907A (en) * | 2019-01-25 | 2019-05-28 | 深圳市象形字科技股份有限公司 | A kind of sit-ups attitude detection and guidance method based on computer vision technique |
CN109815907B (en) * | 2019-01-25 | 2023-04-07 | 深圳市象形字科技股份有限公司 | Sit-up posture detection and guidance method based on computer vision technology |
CN110377824B (en) * | 2019-07-15 | 2020-06-05 | 贝壳找房(北京)科技有限公司 | Information pushing method and device, computer readable storage medium and electronic equipment |
CN110377824A (en) * | 2019-07-15 | 2019-10-25 | 贝壳技术有限公司 | Information-pushing method, device, computer readable storage medium and electronic equipment |
CN110719455A (en) * | 2019-09-29 | 2020-01-21 | 深圳市火乐科技发展有限公司 | Video projection method and related device |
CN110826422A (en) * | 2019-10-18 | 2020-02-21 | 北京量健智能科技有限公司 | System and method for obtaining motion parameter information |
CN111240486A (en) * | 2020-02-17 | 2020-06-05 | 河北冀联人力资源服务集团有限公司 | Data processing method and system based on edge calculation |
CN111240486B (en) * | 2020-02-17 | 2021-07-02 | 河北冀联人力资源服务集团有限公司 | Data processing method and system based on edge calculation |
TWI783679B (en) * | 2020-09-14 | 2022-11-11 | 日商亞西斯特有限公司 | Telescopic motion measurement device and telescopic motion measurement program |
CN112245892A (en) * | 2020-09-25 | 2021-01-22 | 北京汇洋时代科技有限公司 | System for testing movement by using 3D camera |
CN117653995A (en) * | 2022-08-18 | 2024-03-08 | 山东海天智能工程有限公司 | Brain-computer interaction technology rehabilitation training device adopting neuromorphic perception movement system |
WO2024131796A1 (en) * | 2022-12-19 | 2024-06-27 | FindSatoshi Lab Limited | Method for detecting validity of human body movement |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107392939A (en) | Indoor sport observation device, method and storage medium based on body-sensing technology | |
CN111144217B (en) | Motion evaluation method based on human body three-dimensional joint point detection | |
CN109863535B (en) | Motion recognition device, storage medium, and motion recognition method | |
CN110448870B (en) | Human body posture training method | |
WO2021051579A1 (en) | Body pose recognition method, system, and apparatus, and storage medium | |
US10803762B2 (en) | Body-motion assessment device, dance assessment device, karaoke device, and game device | |
KR101227569B1 (en) | Body Segments Localization Device and Method for Analyzing Motion of Golf Swing | |
CN106203503B (en) | A kind of action identification method based on bone sequence | |
US10186041B2 (en) | Apparatus and method for analyzing golf motion | |
CN110334573B (en) | Human motion state discrimination method based on dense connection convolutional neural network | |
CN105229666A (en) | Motion analysis in 3D rendering | |
WO2017161734A1 (en) | Correction of human body movements via television and motion-sensing accessory and system | |
JP6943294B2 (en) | Technique recognition program, technique recognition method and technique recognition system | |
CN105868779B (en) | A kind of Activity recognition method based on feature enhancing and Decision fusion | |
US20220207921A1 (en) | Motion recognition method, storage medium, and information processing device | |
JP7014304B2 (en) | Recognition method, recognition program, recognition device and learning method | |
KR100907704B1 (en) | Golfer's posture correction system using artificial caddy and golfer's posture correction method using it | |
KR102369945B1 (en) | Device and method to discriminate excersice stance using pressure | |
CN113409651B (en) | Live broadcast body building method, system, electronic equipment and storage medium | |
US20150051512A1 (en) | Apparatus and method for recognizing user's posture in horse-riding simulator | |
Ting et al. | Kinect-based badminton movement recognition and analysis system | |
CN111353345B (en) | Method, apparatus, system, electronic device, and storage medium for providing training feedback | |
CN111353347B (en) | Action recognition error correction method, electronic device, and storage medium | |
CN113673327B (en) | Penalty hit prediction method based on human body posture estimation | |
CN109663325B (en) | Scoring system and scoring method for batting sports |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171124 |